Frequently Asked Questions

CityData.AI is a big data and geospatial artificial intelligence company that creates and maintains dynamic urban digital twins for smarter, sustainable, and resilient cities. The company specializes in mobility intelligence in the form of people counting data, density patterns, and movement trends for cities, business districts, parks, and public spaces. We help local governments and civic organizations understand how people move, visit, and interact with different urban areas so they can make informed, data-driven decisions regarding parks and recreation, economic development, community engagement, mobility infrastructure, transit and transportation, public works operations, and resource allocation. Founded in October 2020, CityData.AI is a profitable company and a self-sustaining enterprise.

CityData.AI is headquartered in the technology hub of the San Francisco Bay Area. Our physical office is located at 548 Market Street, San Francisco, California 94104, USA. The company has registered offices in Canada, Mexico, India, and Singapore as well.

CityData, Inc. is a fully registered United States C-Corporation, operating as "CityData.AI" The company is incorporated in the State of Delaware. We are headquartered in Silicon Valley. California. Our physical office is located at 548 Market Street, San Francisco, California 94104, USA. While we maintain our primary business operations within the U.S, our company has registered offices in Canada, Mexico, India, and Singapore as well.

CityData.AI bootstrapped its way to success. Our relentless focus on our civic customers, our endless iterations to achieve product-market fit, and our fiscal discipline to live within our means has ensured profitability for the past many years. While we did raise a modest seed round from institutional investors like True, Orbit, Saama Capital, and angels, CEOs and industry stalwarts in the govtech vertical, we have not had the need for additional funding and we do not anticipate that to change.

CityData.AI is a global entity that serves civic customers internationally. We have active deployments and provide mobility insights in the United States, Canada, Mexico, and major regions in South America, Europe, and Asia, by adapting our data models to local geography and behavior. We have comprehensive mobility data for publishing urban digital twins in every country around the world, with a few exceptions. Over the past 4 years, our customer portfolio has grown to include cities, academic institutions and research organizations in the US, Mexico, Canada, Brazil, Colombia, Argentina, Chile, Peru, UK, Spain, Italy, France, the Netherlands, Denmark, the UAE, Saudi Arabia, Oman, India, Thailand, Malaysia, Taiwan, Singapore, Indonesia, Vietnam, the Philippines, Japan, Australia, and New Zealand.

CityData was founded by Apurva Kumar in October 2020. Apurva is the CEO, CPO, and acting CTO at CityData.

CityData is built upon a bedrock of deep academic and technical expertise, embodied by our Founder and CEO, Apurva Kumar. His background provides the scientific DNA that informs every aspect of our company's methodology and technology. He holds a Master of Science in Engineering and Earth Sciences from Stanford University, with first-principles understanding of spatial data, geography, and the statistical modeling of real-world systems. This academic foundation is complemented by decades of experience in geostatistics, data science, mobile analytics, and machine learning. Apura’s career includes foundational roles at pioneering technology firms and leadership positions at Fortune 100 companies in Silicon Valley, where he architected and launched massive-scale geospatial data platforms and AI-powered geospatial inference engines. His work over the years has empowered urban planners, transportation engineers, and economic developers to make smarter data-driven decisions for local communities. This expertise is coupled with a profound personal commitment to the civic mission. Apurva describes himself as a "geospatial big data and AI nerd who likes to simulate the activity and movement of populations to build mobility twins or digital replicas for smarter, sustainable, resilient cities".

Apurva’s areas of expertise include: Geospatial AI, Geospatial data standards,, Cloud data warehouses, Cloud functions, Transportation big data, Agent-based modeling, Urban air mobility standards, Large-scale multimodal transport simulations, Place-based visitation patterns for economic development, planning, and operations, Computer-vision algorithms, Real-time crowd-counting algorithms, Massive-scale global mobility datasets, Climate impact and disaster resiliency.

You can find Apurva’s Linkedin profile here: https://www.linkedin.com/in/apukumar/

CityData has a stellar team of professionals who are passionate about civic innovation. Our team members are alums of top-tier academic institutions like Stanford University, the University of Southern California, the University of California Irvine, the Indian Institute of Technology, and ETH Zurich. The team has deep expertise in mobility data intelligence, big data + AI / ML, geospatial analysis for location data, large-scale data compute, cloud infrastructure, devops, AWS and GCP. You can read more about the team members on the company website at this link: https://www.citydata.ai/company

CityData’s products, solutions and big data coverage is extensive, spanning the United States, Canada, Mexico, and major regions in South America, Europe, and Asia. CityData.AI can activate urban digital twins for virtually any municipality or region in the world. Since October 2020, CityData’s customer portfolio has included cities, academic institutions and research organizations in the US, Mexico, Canada, Brazil, Colombia, Argentina, Chile, Peru, UK, Spain, Italy, France, the Netherlands, Denmark, the UAE, Saudi Arabia, Oman, India, Thailand, Malaysia, Taiwan, Singapore, Indonesia, Vietnam, the Philippines, Japan, Australia, and New Zealand.

CityData.AI offers a comprehensive suite of products designed for specific civic challenges and targeted use cases:

• CityFlow.AI
  • CityFlow.AI combines massive mobility datasets with agent-based simulation. We give transportation agencies the tools to plan, budget, and operate with certainty. CityFlow empowers DoTs to master the complexity of modern mobility and decode movement patterns through granular origin-destination matrices and multimodal split analysis for walking, driving, and public transit. Our advanced agent-based modeling and simulation engine enables you to test infrastructure scenarios, validate budgets and predict outcomes with precision before you build. From managing crowd surges at major festivals, concerts, and sporting events, to planning evacuation routes for natural disasters, CityFlow provides the foresight you need to build resilient cities. Revolutionize transportation planning and operations with AI. Don't just track traffic. Predict it.
• CitySim.AI
  • CitySim.AI empowers transportation engineers and urban planners with next-generation agent-based modeling (ABM). We generate high-fidelity synthetic populations by fusing real-world mobility data with census demographics and travel surveys to create a realistic digital twin of urban movement. Our high-resolution engine simulates complex "what-if" scenarios, allowing you to predict traffic and congestion outcomes with precision. Evaluate the impact of infrastructure changes—from road expansions to new transit lines—and stress-test your network against major events or natural disasters. Validate intervention strategies in a virtual world before committing real-world resources.
• CityEconomy.AI
  • CityEconomy.AI powers data-driven decisions for Economic Development Organizations. We transform how cities understand their commercial vitality. Move beyond intuition and measure the true heartbeat of your business districts by providing granular metrics on footfall, visitation trends, and customer loyalty in your commercial zones and downtown districts. Our platform reveals detailed trade areas and historical spending so you can get the full picture of your local economy. We analyze cross-visitation patterns to understand where your visitors go before and after they enter your district. You can use our platform to validate marketing strategies, drive smarter policy, arm your team with the evidence needed to recruit new retail, attract businesses, and secure development. Don't guess what drives your economy. Prove it with data. Attract the right investment and build thriving communities with CityEconomy.
• CityEvents.AI
  • CityEvents.AI estimates the attendance, movement trends, demographics, before-and-after behaviors, and spending patterns for events, festivals, recreational programs, sporting, and entertainment activities, at major and minor stadiums, arenas, convention centers, fairgrounds, parks, libraries, community centers, and business districts. The CityEvents product is similar to the CityEconomy product with the main difference being the specific focus on visitors to events and their origin points across the country and around the world.
• CityParks.AI
  • CityParks.AI replaces outdated manual counting with precision analytics for parks and recreation departments. We combine privacy-compliant mobility data with ground-truth sensor calibration to measure what matters. Instantly access accurate metrics on visitation counts, movement trends, dwell times, hourly vs daily insights, weekday vs weekend patterns, weather correlations, origin blocks and demographics of visitors. Our big data and AI platform provides the continuous digital measurement you need to optimize your parks, trails, green spaces, water bodies, natural resources, amenities, facilities, community centers, and event venues. Stop guessing. Start planning with data. Start operating with confidence.
• CitySensor.AI
  • CitySensor.AI provides the ultimate ground truth for urban analytics. Municipal agencies can deploy our privacy-compliant, ruggedized IoT sensors to measure reality with precision. Whether mounted on streetlight poles, parking lots, parks, trailheads, or community centers, CitySensor accurately counts people, vehicles, and animals. The sensors also provide real-time visibility by capturing attendance at festivals, concerts, and public spaces instantly. Our hardware utilizes reliable 4G, 5G, or WiFi connectivity to transmit data and verification snapshots directly to your cloud dashboard. CitySensor is the essential companion to CityParks, CityEconomy, CityEvents, and CityFlow because it validates and re-calibrates the AI models powering all our products. Combine the scale of big data with the precision of on-site measurement.
• CitySurvey.AI
  • CitySurvey.AI is an automated solution in the form of an iOS and Android mobile app that uses computer vision AI for counting and surveying persons, bicycles, animals, and vehicles such as cars, vans, trucks, buses, trains, scooters, and motorcycles. The app is designed to run continuous autonomous surveys without intervention and relay the survey results to the cloud, where metrics are computed and displayed in an online dashboard. The app can be set up on a phone with a tripod for quick surveys and instant counts at parks, trailheads, arenas, stadiums, fairgrounds, events and festivals. The resulting ground truth can be correlated with the crowdsourced mobility data to refine the accuracy and expansion factors of the visitaiton and movement AI models utilized by CityParks, CityEvents, CityFlow, and CityEconomy.
• CityOps.AI
  • CityOps.AI is the agentic AI command center for your mobile workforce. We connect managers to field crews, sales teams, and fleet drivers through a privacy-first, opt-in network. CityOps uses location-aware intelligence to trigger contextual notifications and predefined workflows the moment a worker enters a job site or trade area. You can track routes, measure time-on-site, monitor progress in real-time, and seamlessly export data to your existing task management and payroll systems. Automate the physical world with our geospatial workflow cloud. Optimize field operations, slash costs, and bridge the gap between headquarters and the field. Master your operations with CityOps.
• CityChat.AI
  • CityChat.AI revolutionizes civic engagement through location-aware Agentic AI. We connect residents to their local government with "Cici," a multilingual, 24/7 digital concierge that delivers hyper-local intelligence while collecting real-time constituent feedback. Cici instantly surfaces critical updates including 311 reports, construction permits, crime incidents, community events, and park details. Beyond simple notifications, our platform drives active participation through localized surveys, polls, and neighborhood alerts. Available on iOS, Android, and as a seamless web integration, CityChat builds the transparent, responsive communication channel essential for engaged citizens and thriving cities.
• CityMedia.AI
  • CityMedia empowers city agencies to convert and activate sponsors for local events based on footfall data and visitor attribution insights gathered from previous events. Agencies already have a wealth of knowledge sourced from the visitation insights available through CityParks, CityEconomy, and CityEvents. The insights are translated into actionable sponsorship requests by identifying the businesses and entities most likely to benefit from a surge in footfall visits from city-organized events and festivals. This in turn drives the sponsorship strategy for funding such city-organized events.

CityData.AI caters to many different government agencies. Therefore CityData offers fine-tuned products that cater to the specific needs of each agency. For example, CityParks is finetuned for the requirements and expectations of Parks and Recreation agencies. CityEconomy is designed to meet the needs of Economic Development and Urban Planning agencies. CityFlow caters to the data needs of Mobility Operators and Transportation Agencies. CityOps is meant for Public Works and Field Operations while CityChat is primarily focused on Community Engagement.

CityData.AI derives its insights from a massive pool of anonymized, privacy-compliant mobile location signals. These signals are sourced from mobile apps and SDKs that collect geospatial data with explicit user consent. This raw data is then cleaned, aggregated, and processed through our proprietary machine learning algorithms to represent population-wide movement trends.

Our primary data sources include anonymized privacy-compliant Location-Based Services (LBS) data from mobile applications, connected vehicles, and IoT sensor feeds. CityData.AI collect first party data. We also partner with certified third-party data aggregators who adhere to strict privacy frameworks, ensuring we receive only de-identified, consent-based signals. We do not collect personal attributes like names, emails, phone numbers, home addresses, dates of birth, local or national identifiers. We only collect the hashed mobile identifier also know as the hashed MAID from the apps where end users have consented to sharing their anonymized location data points.

CityData.AI cannot identify specific apps or developers for various reasons. For first-party data sources, we are contractually obligated to withhold the identities of the participating app data provides. For third party data sources like intermediaries and aggregators, we may not be aware of the exact identities of the participating app data providers because the intermediaries typically strip away app-specific identifiers before the data reaches us. This ensures that our analysis remains neutral and focused on aggregate movement based on boarder app categories, rather than user behavior within a specific application.

CityData.AI does not extract demographic data from individuals directly. Instead, CityData.AI uses a privacy-safe inference model. We identify the "common evening location" (presumed home area) of an anonymized device at the Census Block Group level. We then apply the aggregate census data (such as median income, age distribution, and ethnicity) of that Block Group to the visitor profile, providing statistical demographic insights without ever accessing personal user profiles.

Privacy is central to the CityData.AI methodology. We ensure compliance by strictly adhering to "Privacy by Design" principles. We never process Personally Identifiable Information (PII). All data is aggregated into large groups to hide individuals, and we apply techniques like differential privacy and Laplacian noise to geospatial coordinates to prevent re-identification of any single user.

CityData.AI does not track individual users or build profiles on specific people. Our system focuses exclusively on aggregate patterns—processing billions of data points to understand how groups of people move through a city or park, rather than monitoring the specific movements of any single individual.

CityData.AI is fully compliant with the California Consumer Privacy Act (CCPA). We respect the data rights of California residents and have implemented all necessary protocols to ensure our data handling practices meet the stringent requirements of this legislation. CityData.AI is a registered data broker in the State of California and is listed on the Attorney General’s website. You can look us up here: https://cppa.ca.gov/data_broker_registry/

CityData.AI monitors and complies with state-specific privacy legislation, including Oregon Senate Bill 2008B. We continuously update our compliance frameworks to align with the evolving privacy laws across all jurisdictions in which we operate.

To protect individual privacy, CityData.AI employs multiple layers of anonymization. We hash mobile device identifiers (MAIDs) so they cannot be traced back to a person. We also "perturb" or blur GPS coordinates slightly and aggregate data into grid cells or time buckets. Finally, we filter out data samples that fall below a certain threshold (k-anonymity) to ensure no single individual can be isolated in the dataset.

The data provided by CityData.AI is completely anonymous. It is stripped of all personal identifiers such as names, phone numbers, and email addresses before it is ever processed by our algorithms. Additionally, the data is aggregated to only reveals visitation counts and movement patters for a cohort, but never for an individual.

CityData.AI maintains a strict policy against collecting data from children. We do not knowingly collect, process, or store data from devices associated with individuals under the age of 18.

Minors are not directly tracked via their own devices. However, their presence in parks or public spaces is statistically inferred based on household composition data from the census (e.g., "Households with children under 18") associated with the adult devices that visit the location. This allows us to estimate family visitation without digital surveillance of minors.

CityData.AI is aware of two limitations or blind spots. As a policy, we do not collect data from children. Minors are not directly tracked via their own devices. Similarly, we have seen a reduction in data collected from senior citizens due to a drop in usage of mobile apps and digital services for the +65 years cohort. However, their presence in parks or commercial areas or events or public spaces is statistically inferred based on household composition data from the census (e.g., "Households with children under 18", “Households with residents of +65 age”) associated with the devices that visit each location. This allows us to estimate family visitation and senior visitation without digital surveillance of minors or seniors.

Yes. Our datasets and insights are strictly siloed. Using Looker Studio access configurations and BigQuery row/column-level security, data is only accessible by authorized City staff members and essential CityData support employees.

No data products or reports created by City staff are shared with other organizations unless explicitly authorized by the City. While CityData utilizes advanced tools like Vertex AI and Gemini Pro APIs to help staff extract analytical insights, customer data is never used to train Google's foundation models.

Additionally, CityData reserves the right to publish a blog post, case study, and press wire related to this agreement, without revealing any confidential information. The Customer’s identity will never be disclosed without explicit consent.

Yes, CityData is fully compliant with all applicable privacy regulations and the aforementioned security controls. Furthermore, our frontend applications are ADA compliant, adhering to WCAG 2.1 AA accessibility standards.

CityData's Privacy Policy Summary:

CityData does not collect any personally identifiable information (“PII”) such as full names, email addresses, exact home addresses, telephone numbers, identification cards, credit cards, vehicle registration, driver's licenses, or dates of birth. We collect and analyze relevant non-PII, anonymized, and aggregated mobility data across services.

Mobile publishers who partner with CityData are required to implement strict opt-ins and opt-outs to obtain affirmative consent from end users before data collection. CityData warrants that data activities comply fully with applicable laws, including all U.S. state consumer privacy laws (e.g., CCPA, CalOPPA), GDPR (EU), PDPA (Singapore), and LGPD (Brazil).

Unlike marketing companies that monetize citizen data unethically, CITYDATA strictly does not collect, provide, or sell citizen data for advertising purposes to brands, fast-food chains, shopping malls, or commercial real estate owners. CityData is built exclusively for Government use cases. You can read more here: https://blog.citydata.ai/making-mobility-data-privacy-compliant-for-civic-use-case/

Maryland’s data privacy act (MODPA) bans the sale of sensitive data including geolocation data and stipulates that companies may only collect such data if strictly necessary to provide a product or a service. For companies like CityData that curate anonymized granular geolocation data with consent and opt-in, this law could impede the ability to make detailed inferences about visits to subzones within parks, like the pickleball courts or the picnic benches. However, CityData is still able to infer visits to larger spaces like parks and downtowns and event venues. CityData also benefits from 6 years of granular historical data. We have done extensive work for the Maryland-National Capital Park and Planning Commission (MNCPPC) to analyze the parks systems in Prince George’s County and Montgomery County. We have also partnered with KABOOM!, a non-profit focused on parks and playgrounds in underserved communities. All of our historical data has been used to train our AI visitation model which enables us to continue making accurate visitation and movement insights for parks, trails, event venues, and commercial spaces in Maryland. You can read more about about AI model on our blog website here: https://blog.citydata.ai/gravity-model-for-place-visitation-inferences/

The biggest misconception about big data companies is that the results are always accurate and represent 100% of what happened in the park or on the trail. You must understand that big data companies like CityData can only collect data for 70% to 80% of the population. Even Google and Apple cannot collect 100% of the data in any city or county or state. So there is always a gap that needs to be filled. That gap is filled using expansion or scale-up factors in an AI model. When CityData says that we inferred 100 visits to the park, it means we most likely have actual data for around 70 visitors and we applied an expansion model to estimate 100 visitors. That best way to verify if the visitor counts are indeed correct is to validate it against ground truth. You may have registration data for a few events or sensor data for a few weeks, such data is sufficient for CityData to verify and validate if the model results align with the ground truth. Note that models can be fine tuned to improve accuracy and confidence.

States like Maryland have passed privacy law restricting businesses from selling precise geolocation data or processing it without strict consumer consent. Maryland defines "geolocation data" as information identifying a consumer, mobile device, or vehicle within a 1,750-foot radius. However many apps continue to use our technology to legally collect precise geolocation data with consumer consent and opt-in because the Maryland regulation allows apps to collect such data if it is a "necessity" for the functioning of the app. For example, navigation apps, delivery apps, continue to collect such data with consent. Why is CityData in a different category from what these regulations were designed for? Because (1) we do not sell geolocation data about consumers (2) do not identify individual consumers, we aggregate data across parks and open spaces (3) there are no personal attributes in our dataset, we do not collect names, emails, phone numbers, home addresses, or anything that would be considered personal, any data we do hold is completely hashed and secure (4) we intentionally do not work with marketing agencies or advertising networks who may "target" consumers within geofences with ads and offers and (5) we strictly stay clear of wall street firms, investment banks, hedge funds, we only work with local and regional government agencies for public sector use cases based on anonymized and aggregated geolocation data.

CityData.AI provides unique dashboards for each product. The dashboards provide a rich set of mobility insights for business places, commercial areas, downtowns, event venues, facilities, parks, trails, and natural resources. The insights include total visitor counts, timeseries charts of visitation, repeat visitor estimates, average dwell time, peak visitation hours, hourly trends, daily trends, weekday vs weekend trends, and seasonal or annual trends. CityData also provides detailed land visitation heatmaps for a deeper understanding of the usage patterns for micro-areas and sub-zones within each place of interest. Additionally, you can view movement patterns, such as where visitors travel from (Origin-Destination) at the census block level, county, state, and nationwide levels. You can also study the routes that visitors take to reach each place of interest along with the mode of transport, trip speeds, distances traveled, journey times, as well as the origin-destination patterns. And last but not the least, CityData correlates footfalls with weather and provides insights into the variations in visitation patterns with temperature, pressure, rainfall, snowfall, humidity, and visibility levels.

The CityParks dashboard allows for robust comparative analysis. You can select multiple parks, trails, water bodies, community centers, facilities, event venues, or natural resources to view their metrics side-by-side. This feature is particularly useful for understanding how different assets in your system perform relative to one another or how they share visitors.

The CityEconomy dashboard allows for robust comparative analysis. You can select multiple businesses, trade zones, commercial areas, business improvement districts, and event venues, to view their metrics side-by-side. This feature is particularly useful for understanding how different assets in your system perform relative to one another or how they share visitors.

The CityFlow dashboard allows for robust comparative analysis. You can select multiple road segments, or intersections to view their metrics side-by-side. This feature is particularly useful for understanding how different parts of the transportation and transit infrastructure perform relative to one another and how each asset reacts to specific intervention measures.

The dashboard is highly customizable to fit your workflow. You can choose specific date ranges, toggle between different metric views (e.g., switching from daily to monthly granularity), filter by visitor type (e.g., residents vs. tourists), and configure the layout of scorecards to prioritize the data most relevant to your department.

CityData.AI facilitates easy reporting by allowing users to export charts and data directly. You can download visualization images for presentations or export the underlying data in CSV or Excel formats for further analysis or inclusion in external reports.

RESPONSE: Yes, CityData accommodates a wide variety of data export and integration options to ensure our cloud-native datasets work seamlessly with your preferred local or third-party workflows. Our capabilities include:

• Unlimited on-demand, user-defined PDF reports.
• Unlimited downloadable CSV data files.
• Export and import of spatial data via shapefiles in KML/KMZ, .shp, or GeoJSON formats.
• Direct access to our Google Cloud datastore (via BigQuery) with API access for direct SQL queries.
• Seamless API integrations to push data into third-party environments, such as Azure Cloud.
• Embedded analytics, allowing Looker Studio dashboards and charts to be embedded directly into the Customer’s website upon request.
• Export pipelines are readily compatible with third-party analytics and visualization tools like Excel, Tableau, PowerBI, ESRI ArcGIS, QGIS, and SAS.

Users who prefer to work with raw data or third-party tools can access CityData.AI data without relying on the dashboard interface. We provide data exports in standard formats (CSV, JSON, Excel) and offer API integration options for automated ingestion into your internal business intelligence systems. We can also provide access to the BigQuery tables in Google Cloud to enable our customers to execute direct SQL queries against our massive datasets and derive answers to very specific questions.

There is no need to download or install any software. CityData.AI is a fully cloud-native, web-based platform that is accessible from any modern web browser (Chrome, Edge, Firefox, Safari) on any device with an internet connection.

CityData.AI is committed to accessibility and strives to ensure our web platforms are ADA compliant. We follow WCAG (Web Content Accessibility Guidelines) to make our data accessible to all users. Please contact our support team if you require a specific Accessibility Conformance Report.

CityData.AI values customer feedback and actively collaborates with clients to improve the platform. If you have a specific need for a new chart type or data listing, you can propose it to our customer success team, and we frequently incorporate such requests into our product roadmap. We are also willing to create custom dashboards through professional services engagements for customers who prefer different layouts, UX branding, additional data tables, or charts.

CityData.AI uses a default minimum dwell time threshold of 4 minutes to register a visit within a geofence. This means a person (represented by a mobile device) must have stayed within the geofence for at least 4 minutes to count as a valid visit to that place. However, the minimum threshold of 4 minutes is entirely configurable. Certain categories of places like local or neighborhood parks are well suited for the 4 minute threshold. However, we would recommend a higher threshold of 10 minutes for regional or state parks. For economic development use cases, the dwell time threshold can vary by category. Cafes and restaurants would have a lower dwell time threshold than grocery stores and home improvement stores, which in turn might have a lower dwell time threshold than large shopping malls.

CityData.AI measures the distance traveled by visitors from their origin point to the geofence. For local visitors, the distance is measured from the origin point within the county and is labeled as “Local Distance”. The distance traveled by out-of-county visitors who live within the State is labeled as “Regional Distance”. And the distance traveled by visitors from other states within the country is labeled as “National Distance”.

CityData.AI computes the number of raw visits to parks and places using crowdsourced mobility data and applies an expansion ratio based on our gravity AI model. The expansion ratio provides a realistic view of the actual number of people who may have visited the park or place. CityData then computes the trip routes for visitors and their modes of transport: driving, walking, cycling, and public transit. The trips are based on the interpolated raw data to infer the trajectory paths of visitors along the transport network. Given the nature of this process, we typically do not apply an expansion factor or a scale-up to the trips shown within your CityParks or CityEconomy dashboards. However, trips are indeed scaled-up in your CityFlow and CitySim dashboards.

Spending patterns data is based on open datasets including the US Census, ACS, Bureau of Labor Statistics, and the Consumer Expenditure Survey. The open datasets are refreshed once a year or once every two years. These open datasets are combined with specific proprietary datasets that represent anonymized spending patterns from credit card transactions to create a unified spending patterns view for every county across the United States. The proprietary datasets are refreshed once a quarter or once every 6 months. The spend categories reflect the taxonomy derived from the public and private datasets. You can leverage our spending patterns data to understand the consumer spending in absolute dollar value for each census block group or neighborhoods. You can also utilize this dataset to understand relative spending by comparing specific categories (like “food at home”, “entertainment”, “alcohol”, “travel”) across census block groups. Finally you can compute correlations between categories to see if people who spend more on category A also spend more on categories B and C.

The heatmaps use pre-aggregated anonymized GPS density data and is not dynamically recalculated for short-term date filters. Daily or weekly selections may update visitation charts, but the heatmaps are designed to show overall spatial patterns rather than short-term fluctuations or single-event movement.

The quarterly filter aggregates movement data into three-month periods (Q1–Q4). This provides a more stable and representative spatial pattern by reducing short-term noise. Heatmaps are meant to show overall concentration areas across a season, not changes from specific days, weeks, or events.

In Looker Studio, the heatmap does not dynamically recalculate density at every zoom level. When zoomed out, it appears as a smooth gradient. When zoomed in closely, the visualization may show clustered or grid-like patterns. This is the expected behavior of the platform’s rendering logic.

The mobility data is normalized on a scale of 0 to 1 and then converted into a heat map. The heat map visualization dynamically splits the data based on the absolute minimum and maximum values from the normalized set:

Min (0%): Tied to the absolute lowest value in the datasets. It renders blue/green

Mid (50%): Calculated as the exact midpoint between the minimum and maximum values in the dataset. It renders yellow/orange.

Max (100%): Tied to the absolute highest value in the dataset. It renders red.

Any value falling between these percentages is assigned a color based on linear interpolation. Heat maps are representative visualization. They do not capture all of the data for a place or a polygon. The AI model may show higher counts by virtue of the scale-up process than what the actual heat map might represent

To get started with CityData.AI, simply fill our the online form for a free consultation. We will assign you a dedicated customer success manager and business account manager. They will request a call with you to define the scope of your project. We will work with you to identify the specific parks, trails, business districts, commercial zones, event venues, road networks, public transit networks, tollways, bridges, or other areas that you wish to analyze and then activate your account, upload the appropriate data insights for your use cases, and configure your dedicated dashboard instance.

You can define custom geofences by providing us with standard GIS data files (such as Shapefiles, KML, or GeoJSON) that outline the boundaries of your areas. Your inhouse GIS team might be able to readily provide you with these GIS data files from software platforms like ESRI ArcGIS. If such GIS data files are not available, you can draw the shapes on the map using services like Google My Maps (mymaps.google.com). Our team can set you up with a Google account for drawing shapes. Alternatively, you can provide a list of park names or addresses, and CityData’s geospatial team can map them for you as a free courtesy service.

You are welcome to draw your own geofences using tools like Google Earth, Google My Maps, or ArcGIS. Once you have drawn the shapes, you can export them as KML or GeoJSON files and send them to CityData.AI for ingestion into our cloud data platform.

Geofences are not static and can be modified if park boundaries or business district boundaries change or if you wish to refine your analysis. However, because data is pre-processed against these shapes, modifying a shape often requires re-running the historical data analysis to ensure accuracy, and might incur additional effort.

Minor modifications to shapes are typically included as part of your active subscription and support. However, extensive re-mapping or frequent large-scale changes that require significant data reprocessing may incur additional service fees depending on the terms of your contract.

You can swap out existing shapes for new ones to keep your analysis relevant. Similar to modifications, swapping a few shapes is generally covered, while a complete overhaul of your monitored locations might be treated as a new setup or incur a configuration and data reprocessing fee.

Submitting your own official park shapes is actually the preferred method. This ensures that the analysis matches your department's exact definitions of park boundaries, including or excluding specific amenities like parking lots or maintenance yards as desired.

CityData.AI is not limited to parks, trails, green spaces, recreational facilities, campgrounds, water bodies, and natural resources that are typically included in our CityParks product. Our technology platform is location-agnostic and can analyze any geofenced area, including downtown zones, business improvement districts (BIDs), retail corridors, industrial zones, trade areas, event venues, and mixed-use developments, using our CityEconomy and CityEvents products.

Currently, shape ingestion involves a data processing pipeline that includes cleaning, calibration, and historical backfilling, and ground truthing to produce the most accurate results using our big data and AI platform. Therefore, you cannot get immediate results instantly after uploading a shape; the process typically takes a 5 to 15 business days to generate the full historical dataset. That said, you can ingest a shape and view a modeled estimate immediately based on historical data, such estimates are revised later when the real-world data is processed with higher confidence scores.

CityData.AI is built on a scalable cloud infrastructure capable of processing unlimited places and unlimited geofences. Whether you need to analyze five parks or five thousand city blocks or five hundred thousand road segments or way IDs, our platform can handle the volume without performance degradation.

You can request updates to your custom shapes as often as necessary to reflect real-world changes. We recommend reviewing your geofences annually or semi-annually to ensure they remain aligned with your current city planning zones.

CityData.AI provides an "Unlimited users" licensing model for your organization. You can grant access to as many team members, stakeholders, or inter-departmental colleagues as needed without paying for extra seats. In fact, we encourate our municipal customers to share the data dashboards will all stakehoders, colleagues, and internal users. And we welcome feedback and diverse perspectives that can help us improve our big data platform and our AI inference engine.

While the standard dashboard is secured for internal staff use, CityData.AI can configure public-facing views or "transparency portals." These are simplified, read-only versions of the dashboard designed to share key metrics with the community while protecting sensitive internal data.

Yes. While most of CityData's products are turnkey, we typically propose a compact and structured 5 to 6-week implementation plan to launch your data dashboards and insights from the time of contract signing:

Weeks 1 - 2: Setup, Ingestion & Geofencing

• Conduct project kickoff meeting(s) with Customer staff, stakeholders, and partners.
• Define geofences and custom areas for all Places of Interest (POIs).
• Ingest existing Customer GIS shapefile definitions securely into our BigQuery cloud environment.
• Ingest any Customer-owned “Ground Truth” datasets (e.g., traffic, sensor, or parking counts; surveys; sales tax data) that the Customer is voluntarily willing to share.
• Draw and ingest custom-defined shapes if pre-existing GIS shapefiles are not available.
• Set up the Google Cloud data pipelines for the Customer using CityData’s anonymized crowdsourced mobility, visitation, and demographics data.
• Finalize desired metrics to include in the dashboard.
• Deliver 2x optional CitySensor hardware devices to the Customer and collaborate to identify deployment locations (parks, trails, open spaces, etc.).

Weeks 3 - 4: Computations, Hardware Testing & Initial Review

• Compute visitation counts and movement trends using the anonymized crowdsourced data pipeline.
• Customer’s team, Public Works, or a preferred local vendor visits the sites to install and operationalize the sensors.
• Test the sensors to ensure cellular data connectivity and real-time data transmission for capturing pedestrian, bicyclist, or vehicle counts.
• Calibrate and fine-tune the crowdsourced visitation data models to ensure any deviations are within an acceptable range compared to the ground truth.
• Provide initial data insights and an early draft of the Looker Studio dashboard for Customer review and feedback.
• Publish the initial CityParks dashboard.

Weeks 5 - 6: Refinement, Delivery & Training

• Provide a second version of the dashboard for Customer review.
• Meet to discuss and gather additional feedback/feature requests.
• Finalize calibration and fine-tune the dashboard based on feedback.
• Review, publish, and deliver the final, production-ready dashboard.
• Provide comprehensive training for all Customer staff members and stakeholders.
• Publish the online URL/link for the dashboard, updated with the Customer’s branding/logo.
• Publish an online external and public access dashboard with Customer branding (if requested).
• Integrate and embed the dashboard within the Customer’s website if desired.

CityData infers the visits to a park, trail, or downtown zone. CityData then proceeds to infer the origin or home census block of visitors based on their overnight stay patterns. If a visitor to a park lives within the same county, they are labeled as “local visitor”. If a visitor lives within the same state (but not within the county), they get labeled as “regional visitor”. If a visitor lives outside the state, then they are a "national visitor”. Note that CityData can upon request also infer international visitors from other countries..

CityData.AI primarily provides high-fidelity historical data. Our core datasets have a latency of 2-3 days. However, in all our years of providing big data services to municipal agencies, we have only received requests for refreshing data insights on a monthly frequency because our customers acknowledge their inability to react faster to data insights even if we were to refresh the dashboards more frequently. The one exception is events. Some of our customers might expect results for events and festivals within 7-to-14 days. We can meet this timeline for our events-focused customers. To summarize, our standard engagement model focuses on analyzing confirmed movement patterns with a monthly cadence, which allows for greater accuracy and the elimination of noise. However, we do offer near-real-time solutions via our hardware sensors.

While a weekly delivery cadence is technically feasible, CityData.AI recommends a monthly cadence for mobility data. Weekly splits can be volatile due to weather or short-term anomalies, whereas monthly aggregation provides a statistically robust baseline for trend analysis. CityData.AI primarily provides high-fidelity historical data. Our core datasets have a latency of 2-3 days. That said, in all our years of providing big data services to municipal agencies, we have only received requests for refreshing data insights on a monthly frequency because our customers acknowledge their inability to react faster to data insights even if we were to refresh the dashboards more frequently. The one exception is events. Some of our customers might expect results for events and festivals within 7-to-14 days. We can meet this timeline for our events-focused customers. To summarize, our standard engagement model focuses on analyzing confirmed movement patterns with a monthly cadence, which allows for greater accuracy and the elimination of noise. However, we do offer near-real-time solutions via our hardware sensors.

A monthly cadence is suggested because it aligns with standard urban planning and reporting cycles. It also ensures that the data has a sufficient sample size to be statistically significant and allows for the necessary time to clean, calibrate, and verify the data against ground truth sources.

The CityData.AI dashboard is refreshed on a monthly basis. Typically, data for the previous month is processed, quality-checked, and published to your dashboard by the 10th day of the current month.

For purely mobile app data, there is always a latency in aggregation. However, CityData.AI can provide real-time visitation insights by deploying CitySensor hardware. These physical sensors process counts on the edge and upload data instantly, providing live crowd intelligence.

Our historical data archives generally extend back to January 2019 or 2020, depending on the region. This deep history allows you to perform crucial longitudinal studies, such as comparing current visitation trends against pre-pandemic levels.

Once a new month ends, it takes approximately two to three weeks to ingest the raw data, apply privacy filters, run calibration models, and publish the final insights. For new customer setups, processing historical backfills for new geofences typically takes one to two weeks.

CityData.AI obtains demographic insights by correlating the visitor's inferred home location (Census Block Group) with open census data. We do not obtain demographics from the user's phone or apps directly; rather, we apply the statistical characteristics of their neighborhood to the aggregate visitor profile.

Upon subscribing, you will receive access to the dashboard and historical data usually within one week of providing your geofences. On an ongoing basis, new monthly data is delivered automatically by the middle of the following month.

Access to historical data is a core component of our offering. We believe that understanding the past is essential for planning the future, so we provide extensive historical baselines as part of our standard analytics packages.

CityData.AI allows you to access the processed data behind the charts. While we do not provide raw, device-level traces due to privacy restrictions, you can self-serve by exporting the granular, aggregated datasets (e.g., daily counts, hourly distributions) directly from the dashboard in CSV or Excel formats.

Events are an integral part of our analysis. CityData.AI captures the visitation spikes associated with festivals, concerts, and public gatherings within your geofences, allowing you to measure the impact and reach of specific community events.

For specific event reports, CityData.AI can often prioritized processing to provide insights within 5 to 15 business days after the event concludes, allowing for timely post-event reporting and impact assessment.

Providing event dates is highly recommended. When you share a calendar of events (past or upcoming), it allows our data scientists to contextualize anomalies in the data, validate peak counts, and generate specific "Event Impact Reports" that isolate event traffic from normal baseline park usage.

To ensure the highest accuracy for events, you can provide CityData.AI with any available ground truth data, such as ticket sales, gate counts, or manual clicker tallies. We use this data to calibrate our models specifically for the high-density crowds typical of large festivals.

Both the CitySurvey app and the CitySensor are excellent tools for event counting. The app allows staff to perform roving counts or surveys using any mobile phone while the sensors can be temporarily or permanently deployed at entry points to provide a continuous, automated count of attendees.

Accuracy of geolocation data is measured in three distinct ways:

1. Horizontal accuracy of GPS data: Each GPS geolocation data point sourced from mobile apps or connected vehicles has a “horizontal accuracy” metric attached to it. The horizontal accuracy typically varies from 1 meter to 100 meters and represents the accuracy of the GPS data point on the map. A (latitude, longitude) data point with 10 meter accuracy indicates that the device producing that data point is within a circle of 10 meter radius from the (latitude, longitude) pair.

2. Population coverage of the mobility dataset: The mobility datasets for a city, county or state must be compared with the resident population based on the latest census data to understand the population coverage and therefore the accuracy of the inferences made using such mobility data.

3. Hyper-local ground truth correlations: The visit inference for a park or trail or downtown zone must be compared with confirmed counts measured using manual surveys, event registration data, traffic sensor data, or videos from camera feeds, to infer accuracy relative to the ground truth.

Our model measures accuracy along all three vectors to estimate an overall accuracy for the visitation results. Our goal is to achieve an accuracy of 80% or higher, however that may not always be achievable based on the characteristics of the specific place and the built environment.

Yes. Leveraging BigQuery and Vertex AI, CITYDATA has strict quality control procedures in place for all raw mobility datasets. The data is continuously scanned via our algorithms to identify anomalies, hardware issues, unrealistic movement patterns, or fraudulent activities.

To ensure forensic-level data hygiene, our algorithms categorize observations using several highly specific quality flags:

Too Many Devices at Location: Identifies an improbable number of devices with the exact same latitude/longitude on a given day, flagging it for further analysis.

Radio Derived Data: Indicates location was likely derived from cell tower triangulation rather than GPS or WiFi, which generally provides moderate to low accuracy depending on tower availability.

Mobile Network: Indicates the signal was gathered on a mobile network rather than a non-mobile IP address.

Accuracy Tiers (High / Moderate / Low): Automatically categorizes horizontal accuracy ranges based on the signal's metadata.

US Flag: Uses approximations over time to flag if a device spends the majority of its time in the US, adjusting dynamically if travel patterns change.

Lat-Lon on Grid Location: Flags signals falling exactly on grid coordinates, indicating the location is likely estimated, imprecise, or spoofed.

Spoof Location: Specifically flags locations that appear mathematically or behaviorally spoofed.

Over Capacity Device: Truncates and flags signals from suspicious devices exhibiting abnormal density (e.g., exceeding 2,500 signals/day).

Hypervelocity (HyperV) Indicators: Flags impossible speed vectors, including moving too fast within a spatial-temporal cluster, jumping outside a cluster, or overlapping/interleaving clusters (often due to shifted timestamps or spoofing).

Timeshift: Flags observations that appear erroneously adjusted to local time rather than Coordinated Universal Time (UTC).

Implausible Movement: Identifies physically impossible travel distances between two timestamps (e.g., 5,000 meters traveled in 0 seconds).

Approximated Signal: Indicates the location is derived from the centerpoint of a Geohash, with approximated timeframes (morning, afternoon, etc.).

Replayed Data: Flags historical data patterns replaying on future dates with manipulated timestamps.

"Ground truth" refers to real-world verification data collected on-site, such as manual counts, traditional surveys, click-counting, sensor readings, or ticket sales. CityData.AI uses this tangible data to benchmark our digital mobility signals, ensuring that our AI models accurately reflect reality.

Providing ground truth data is necessary for calibration. Mobile data penetration rates can vary by region and demographic. By comparing our digital signal counts against your physical ground truth counts, we can calculate a precise "expansion factor" to scale our data up to the total population with high confidence.

We recommend providing at least one week of ground truth data per month, including both weekdays and weekends. Ideally, this data should cover different times of the day (morning, afternoon, evening) to capture the variability in visitor flow.

The most effective format for ground truth data is a simple CSV or Excel file containing three columns: Date, Time (in hourly or 15-minute intervals), and Count. This standardized format allows our system to ingest and correlate the data quickly.

CityData.AI employs a "ratio-based" calibration method. We calculate the correlation between the number of mobile devices observed in a block or a zone and the actual people counted on the ground. This ratio (the expansion factor) is then applied dynamically to our historical and future mobile data to derive total visitation estimates.

Yes. CITYDATA compares and correlates the people counts in our datasets with physical ground-truth counts to validate and calibrate our models. We actively collect and curate ground-truth data from open data sources wherever available, including:

• Transit station ridership counts
• Highway entry/exit counts
• Airports visitation counts
• Road survey counts
• Events venue attendance counts
• Parks & trails counts
• Tourism board and downtown association counts
• Business district counts
• Thermal satellite images (where available)
• Census population counts (US Census and ACS)

The CitySurvey app is available for iOS and Android. The app is indeed free to download and use for CityData.AI customers. It is designed as a companion tool to facilitate the easy collection of manual counts and qualitative survey responses from citizens.

The CitySurvey app is cross-platform and compatible with the vast majority of smartphones. It is available for download on both the Apple App Store (iOS) and the Google Play Store (Android).

CityData.AI offers CitySensor, a proprietary, ruggedized hardware device designed for permanent outdoor installation. It uses edge-computing computer vision AI to count pedestrians, cyclists, vehicles, and even animals continuously in real-time. The results are automatically uploaded to our cloud data platform and correlated with the crowdsourced mobility datasets.

The CitySensor is flexible regarding power. While it can be hardwired into mains power (e.g., on a light pole), we also offer low-power versions that can be deployed with solar panels and battery backups for remote trail locations.

The hardware sensor counts "visits" or passages rather than unique individuals. Because it does not store personal identities (like faces or phone IDs) for privacy reasons, it counts each time a person, vehicle, cyclist, or animal passes the sensor, which provides excellent data on flow and volume.

The CitySensor does not use facial recognition. It utilizes computer vision to detect the "shape" and movement of a human or bicycle but does not capture, store, or analyze facial features or biometric data.

The sensor is fully privacy-compliant. All image processing happens "on the edge" (on the device itself), and no video feeds are recorded or transmitted to the cloud. Only the numerical count data is transmitted, ensuring no personal data is ever at risk.

Yes, the CitySurvey app acts as a digital survey tool replacing the traditional clicker for manual surveys. The app is freely available to our clients to help them mobilize volunteers or staff for counting and ground truthing projects.

The app is built to be universally accessible and works seamlessly on both iOS and Android operating systems.

Data collected via the CitySurvey app or CitySensor hardware is considered the "Gold Standard" of ground truth. We prioritize this data in our calibration pipeline to refine the expansion ratios and fine-tune the accuracy of our broader, city-wide mobility models.

CityData.AI provides a comprehensive User Manual and Installation Guide for the CitySensor. This documentation covers everything from physical mounting instructions to configuring the software dashboard, ensuring your team can operate the hardware independently.

CityData.AI is designed to be interoperable. We can easily integrate with Esri ArcGIS and other standard GIS platforms. Our data exports (CSV, Excel, JSON, GeoJSON) are formatted to be directly ingestible as layers into your existing tools and geospatial environment.

We can deploy a customized, public-facing version of the dashboard. This "Community View" allows you to share selected success metrics and transparency data with the public while keeping sensitive operational data secure behind a login.

The platform includes a robust user management system. Administrators can create individual accounts for staff members, assign roles, and manage secure login credentials to ensure data security.

CityData.AI supports Enterprise Single Sign-On (SSO). We can integrate with your organization’s identity provider, including Microsoft Azure AD (Entra ID), Google Workspace, and LinkedIn, to streamline access management.

While the dashboard itself is a secure web app, we can provide you with static embed codes or public links for specific reports. Alternatively, many customers export the data to power their own website visualizations.

Embedding live, interactive maps usually requires a specific API setup or a public-facing dashboard module. Please consult with your Customer Success Manager to determine the best technical approach for embedding dynamic maps into your public portal.

Data portability is a key feature. You can download your historical and current data at any time for archival purposes or to perform advanced offline analysis in tools like Excel, SPSS, or R.

Importing data into Esri ArcGIS is seamless. Since our data is geospatial by nature and exported in standard formats like CSV (with lat/long) or GeoJSON, it can be dragged and dropped directly into ArcGIS Pro or ArcGIS Online.

CityData.AI exports are fully compatible with Microsoft PowerBI. You can connect PowerBI directly to our CSV exports or API to build custom, interactive business intelligence dashboards that combine our mobility data with your other internal datasets.

Our data is formatted to work perfectly with Tableau. You can import the datasets to visualize complex relationships, such as correlating park visitation with other city metrics managed in your Tableau environment.

Yes, CityData.AI supports the open-source community. Our standard data formats (CSV, GeoJSON, KML) are native to QGIS and other open-source geospatial tools, allowing for flexible and cost-effective analysis.

CityData.AI views every contract as a partnership. You will be assigned a dedicated Customer Success Manager (CSM) who will guide you through onboarding, assist with data interpretation, hold regular check-in meetings, and ensure you are getting maximum value from the platform.

Our core team is based in San Francisco (Pacific Time), but we have support staff and account managers distributed across multiple time zones to ensure timely responses for our global customer base in North America, Europe, and Asia.

We utilize a professional project management system to track all customer requests. Whether it's a new feature request, a shape modification, or a data inquiry, every task is logged, tracked, and visible to ensure accountability and timely resolution.

CityData.AI maintains a library of training resources, including video tutorials, step-by-step guides, and "playbooks" specifically designed for parks departments to help new users navigate the dashboard and understand the data.

We offer comprehensive training options. Standard onboarding includes live virtual training sessions for your team. For larger deployments, we can also arrange on-site workshops to train your analysts and leadership on leveraging mobility intelligence.

If you experience access issues, our Support Team is available via email (support@citydata.ai) and in-app chat. We can quickly assist with password resets, account unlocking, and troubleshooting browser compatibility issues.

CityData.AI offers a flexible pricing model designed to fit government budget cycles. We typically structure pricing based on the number of assets (parks/places) monitored and the size of the population, with options for annual subscriptions or project-based "Pay-as-you-go" fees.

To simplify procurement for civic clients, we highly recommend and support annual billing. This reduces administrative overhead for both parties, although monthly payment terms can be arranged if required by your agency.

CityData.AI supports "One-off" data purchases. If you do not require an ongoing subscription, you can purchase a comprehensive historical report for a specific time range to support a master plan or grant application.

Pricing is generally volume-based. While historical data is a premium asset due to its depth, bundling it with an ongoing subscription often results in the most cost-effective per-month rate. Please contact our sales team for a specific quote.

We accept a wide range of payment methods to accommodate municipal finance departments, including major credit cards, bank wire transfers, and ACH (Automated Clearing House) payments.

CityData.AI accepts traditional mailed checks. We understand that this is often the standard payment method for many government agencies and municipalities.

For most government and enterprise engagements, a Purchase Order (PO) is the standard procedure. We are experienced in working within the PO process and can provide all necessary vendor forms and W-9s to facilitate POs.

CityData.AI supports disaster resiliency by modeling population movement during crises. Our CITYOPS and mobility data can identify evacuation bottlenecks, measure the displacement of populations during floods or fires, and help emergency managers plan resilient infrastructure based on actual human behavior patterns during stress events.

CityData.AI is at the forefront of future mobility. Our CITYSIM product provides agent-based simulations that help cities plan for Urban Air Mobility (eVTOLs/drones). We analyze ground traffic and commuter demand to identify the optimal locations for vertiports and model the impact of air traffic on the urban canopy.

Citizen engagement is streamlined through CITYCHAT, our bilingual AI chatbot app. It allows cities to broadcast geofenced alerts, conduct "pulse" surveys, and enable citizens to report 311 issues (like potholes or broken lights) directly from their phones, creating a continuous feedback loop.

CityFlow** is our dedicated mobility product for transportation planners. It goes beyond simple counts to provide complex Origin-Destination (OD) matrices and trip trajectories. This helps Departments of Transportation (DOTs) understand where commuters live versus where they work, visualize traffic congestion, and plan more efficient bus and transit routes.

Economic development is a key use case for our CityEconomy product. By analyzing foot traffic in downtowns and commercial corridors, we help Economic Development Organizations (EDOs) prove the vitality of a district to prospective retailers, measure the "recovery" of downtowns post-pandemic, and quantify the economic impact of tourism.

CityData utilizes a fixed-fee subscription model for the CityFlow platform. This provides our customers with predictable budgeting for platform access, hosting, and data exports. Variables may include geographic extent (e.g., specific counties vs. statewide coverage) and optional professional service hours for custom data engineering billed at standard hourly rates.

Driving costs are primarily centered on the massive-scale data compute and cloud infrastructure required to ingest and archive +10 Terabytes of daily mobility data from over 10,000 global sources. Specialty items include the proprietary geospatial AI and machine-learning techniques used for spatiotemporal clustering. The data pipelines are fully automated for standard features.

CityData controls costs through a commercial-off-the-shelf (COTS) SaaS model, which removes the need for our customers to invest in internal hardware or data storage. Automated data pipelines and automated compute scale-up scale costs in a predictable manner while controlling overages..Algorithmic expansion ratios based on existing traffic sensors and existing cctv camera feeds allows for regional coverage without requiring the deployment of physical sensor deployment for every road segment.

CityData offers CityFlow, a commercial-off-the-shelf (COTS), cloud-based software-as-a-service (SaaS) solution designed to deliver high-fidelity "Mobility Intelligence" for smarter, more resilient urban environments. The platform serves as a standalone environment for multimodal movement analytics, regional planning, and transportation system analysis.

Multimodal Movement Intelligence: CityFlow transforms raw, crowdsourced mobility data from over 10,000 sources—including connected vehicles, mobile apps, and IoT sensors—into actionable analytics. It uses proprietary confidence score thresholds to accurately infer and distinguish between various transport modes, such as driving, walking, bicycling, and public transit (bus, metro, rail).

Origin-Destination (O-D) Matrices: The platform spatially and temporally clusters anonymized data points to infer O-D trip hops and matrices. These matrices are essential for understanding travel demand and identifying the top communities or neighborhoods from where people visit specific points of interest (POIs).

Route Trajectory Visualization: CityFlow generates exact geometries and routes along the regional and local roadway networks. This data includes detailed metrics for each individual road segment, such as speed, trip distance, and travel time for each trip, and aggregated across all trips..

Algorithmic Traffic Volume Scale-Up: CityData applies machine-learning-based scale-up techniques to align sample trip counts with expected regional population figures. This process uses census data, ground-truth measurements, traffic counters, and cctv feeds, to ensure the insights reflect the true pulse of the entire metropolitan area rather than just a panel sample.

Toll Network Correlations: Along the same lines, CityData applies machine-learning-based scale-up to align the trips and routes dataset with existing toll network data and hourly volume counts collected from toll stations, and entry/exit ramps on highways.

Demographic Attributes: The CityFlow trips and routes data is correlated to census demographics data at the census block group level across the entire region to produce a deep understanding of the demographic characteristics of the commuting population, include age distribution, income levels, household sizes, employment levels, occupation types, and spending patterns.

CityData ensures precision by utilizing location signals typically within 1 to 20 meters. CityData applies spatiotemporal clustering, requiring multiple data points from a device within a set time window to confirm a trip segment. As described above, CityData also utilizes census and acs population data, ground-truth measurements, traffic counters, cctv feeds, toll network data, and highway entry/exit ramp data to fine tune the accuracy of the CityFlow dataset.

CityData employs a streamlined 6-week go-live plan for the deployment of its Mobility Intelligence solution. Subject to the formal signature of the contract and the prompt availability of required data files, specifically GTFS and RT-GTFS feeds for public transit integration and multimodal trip interferences.

As a cloud-native platform, CityFlow is designed for seamless interoperability and does not require our customers to install or maintain any new proprietary software. The solution is "software agnostic," meaning it integrates directly into your existing ecosystem through the following categories:

Inbound Data Requirements: To provide transit-specific analytics, CityData requires access to our customers’s GTFS or RT-GTFS feeds and GIS shapefiles (e.g., road networks or custom zones).

Cloud & Warehouse Interoperability: CityFlow provides direct API integration with Google BigQuery, AWS S3, and Microsoft Azure, allowing our customers to ingest data tables directly into existing data lakes for automated processing.

Analytics & Visualization: For reporting and spatial analysis, the platform supports automated data exports to industry-standard tools including Google Looker Studio, ESRI ArcGIS, QGIS, Tableau, Power BI, and Excel.

Standard Browser Access: Authorized users can access the secure, interactive online dashboard using any standard web browser (e.g., Chrome, Edge, Safari, or Firefox) without additional plugins.

CityData.ai integrates a diverse array of traffic performance metrics and behavioral data into its solution, as illustrated in the comprehensive CityFlow dashboard. The platform provides high-fidelity insights across the following categories:

Multimodal Aggregated Performance Metrics

The dashboard displays real-time and historical analytics for Driving, Cycling, Walking, and Transit trips. For each mode, CityData captures:

• Total Trip Volume: Real-time and cumulative counts (e.g., 751.7M driving trips).

• Average Traffic Speed: Segment-specific and network-wide speed metrics (e.g., 51.2 km/h for driving).

• Average Travel Time/Duration: Detailed duration tracking (e.g., 22:10 minutes for driving).

• Average Trip Distance: Precision measurement of travel length (e.g., 21.4K meters for driving)

  Network-Level Aggregated Analytics

  CityData provides granular visibility into the road network's functional health:

• Route Aggregation & Volume: The "Route Aggregation" view identifies specific roadway segments (e.g., Texas Avenue) and tracks total trips and average speed at the link level.

• TMC and Segment Granularity: Performance data is provided for each individual road segment (using OSM Way IDs), enabling block-by-block bottleneck detection.

• Transit Specifics: For public transport, the platform monitors average speeds (32.1 km/h) and distances (13.8K), specifically calibrated against GTFS schedules.

  Origin-Destination (O-D) & Behavioral Insights

  The platform transforms raw data into high-level mobility intelligence:

• O-D Matrices: CityData clusters billions of data points to generate O-D matrices between specific Census Block Groups.

• Trip Distributions: Visualizations show temporal patterns of life via Average Trip Distribution by Hour of Day, Day of Week, and Day of Month.

• Micro-Geofencing: Analytics for specific points of interest (POIs) such as commercial zones, downtown districts, or transit hubs.

The CityFlow data is provided in the form of massive data files or through access to the data warehouse in Google Cloud BigQuery.

• Origin-Destination Movement
  • The OD Movement data provides the trip origins and destinations for movement across the administrative or geospatial grid.
• Trip Route Trajectories
  • The OD Movement data provides the trip origins and destinations for movement across the administrative or geospatial grid.
• Traffic Volumes
  • The Traffic Volumes data provides the aggregated traffic counts for each road segment with average speed and modality.
• Place Visits
  • The Place Visits data provides the anonymized and aggregated visits to places, business, downtowns, commercial zones, BIDs, and custom-defined geofences within the region of analysis

You can find the complete data dictionary with attribute definitions for each of our datasets at this link:

• CityData Dictionary
  • https://www.citydata.ai/dictionary

CityFlow data penetration tends to range from 35% to 65% for different road geometries like urban roads, rural roads, state routes, U.S. routes, freeway sections, arterials, and local roads. The penetration varies by county and by state. As described above, CityData applies algorithmic traffic volume scale-up, a machine-learning-based scale-up technique to align sample trip counts with expected regional population figures. This process uses census data, ground-truth measurements, traffic counters, cctv feeds, tollway data, and highway entry / exit counts to ensure the insights reflect the true pulse of the entire metropolitan area rather than just a panel sample.

CityData provides an intuitive, web-based CityFlow dashboard that empowers users to retrieve and analyze historical mobility data with ease. The platform is designed to transform massive datasets into interactive visualizations, allowing our customers staff to perform longitudinal studies and evaluate project success over time.

Deep Historical Insights

Extensive Data Archive: CityFlow offers an extensive historical baseline with data archives extending back to 2019, providing several years of consistent mobility patterns for comparison.

Long-Term Trend Analysis: Users can easily query and visualize multi-year trends to understand how infrastructure changes or policy decisions have impacted traffic flow and population movement.

Time-Series Visualization: The dashboard features dedicated time-series charts that allow users to toggle between hourly, daily, and monthly granularities to pinpoint specific historical anomalies or peak periods.

Direct SQL Access to Cloud Datawarehouse:

For deeper research, CityData provides direct API access to cloud datastores (e.g., Google BigQuery), allowing our customers to run custom SQL queries across the entire historical database without going through the UI.

CityData provides a secure, enterprise-grade user interface designed for seamless accessibility without the need for specialized local software or hardware. The CityFlow dashboard is a cloud-native platform accessible through any standard modern web browser—including Google Chrome, Microsoft Edge, Apple Safari, and Mozilla Firefox—ensuring that our customers personnel can retrieve critical mobility intelligence from any location with an internet connection.

Yes. CityData.ai provides a robust suite of analytical tools through the CityFlow dashboard, specifically engineered to measure and visualize the mobility performance of roadways and transform complex datasets into actionable visual insights:

Multimodal Performance Charts: The platform features interactive graphical summaries for Driving, Cycling, Walking, and Transit trips. For each mode, it tracks critical KPIs including Total Trip Counts, Average Speed (km/h), Average Distance (meters), and Average Duration (minutes).

Density and Heatmap Visualizations: High-resolution heatmaps and choropleth maps allow planners to identify "hot spots" of population presence and movement density across urban and rural corridors.

Granular Route Aggregation: Users can visualize Trip Route Trajectories directly on the road network. This tool provides segment-level detail (using OSM Way IDs), displaying total trips and average speeds for specific roadway links (e.g., Texas Avenue) to pinpoint localized bottlenecks.

Origin-Destination (O-D) Matrices: Interactive spatial maps and tables visualize the flow between specific census blocks or custom polygons, allowing for the analysis of regional travel demand and "patterns of life".

Temporal Trend Analysis: The interface includes time-series charts to evaluate performance by Hour of Day, Day of Week, and Day of Month, enabling "before-and-after" longitudinal studies of infrastructure projects.

Direct SQL Access to Cloud Datawarehouse: For deeper research, CityData provides direct API access to cloud datastores (e.g., Google BigQuery), allowing our customers to run custom SQL queries across the entire historical database without going through the UI.

Yes. CityData provides a highly flexible data architecture that enables the seamless export and download of mobility datasets. Authorized users have the ability to retrieve all elements of analyzed data in standard, non-proprietary formats to ensure compatibility with existing state systems and third-party tools.

Specific capabilities include:

Unlimited Exports: Users can generate unlimited downloadable CSV data files directly from the CityFlow dashboard.

Comprehensive Format Support: Beyond the online dashboard, CityData CityData provides direct API access to cloud datastores (e.g., Google BigQuery), allowing our customers to run custom SQL queries across the entire historical database without going through the UI. This allows our customers to download structured data CSV, XML, JSON, GeoJSON, and Parquet formats.

CityFlow platform provides a comprehensive suite of transportation performance indicators. Key metrics visually demonstrated in the dashboards include:

Trip Performance: Average speed (km/h), total trip count, average distance (meters/km), and average duration (minutes/hours).

System Reliability: Maximum distance and maximum trip duration filters to identify outliers and long-range travel patterns.

Traffic Operations: Total trips and average speeds for specific road segments to identify localized bottlenecks.

Multimodal Metrics: Modal split percentages (driving, walking, cycling, transit) and mode-specific speed/distribution profiles.

Origin-Destination (O-D) Indices: Top origin and destination blocks ranked by total trips, average distance, and average speed to support regional Planning Time Index analysis.

In addition to the CityFlow crowdsourced movement data, CityData also provides comprehensive simulation capabilities through the CitySim product. CitySim is an ABM simulation engine, inspired by the CitySim framework. CitySim is unique because it models individual agents rather than aggregate traffic flows. As a result, its key performance indicators (KPIs) focus heavily on the agent's experience (utility) and detailed temporal dynamics, in addition to standard network metrics. CitySim provides the below transportation performance indicators:

1. Agent Utility (Average Score)

This is the most fundamental CitySim-specific metric. The simulation runs iteratively to maximize this score.

What it measures: The overall "satisfaction" of agents with their daily plans, based on a scoring function that rewards performing activities (e.g., working, shopping) and penalizes travel time, cost, and waiting.

Why it matters: It indicates if the system has reached Nash Equilibrium (stability). If a policy change (e.g., a new train line) increases the average population score, it is generally considered a net benefit to society.

2. Modal Split (Mode Share)

What it measures: The percentage of trips or distance traveled by each mode of transport (car, public transit, walking, cycling, etc.).

Why it matters: It is often the primary target of transport policies (e.g., "Shift 10% of car trips to transit"). CitySim calculates this dynamically based on agent choices rather than static matrices.

3. Travel Time Distribution

What it measures: A histogram or cumulative curve showing how long trips take across the population, often segmented by mode or time of day.

Why it matters: Averages can be misleading. This metric reveals equity issues, such as whether a specific group of commuters suffers from extreme commute times (the "long tail" of the distribution).

4. Link Volumes (vs. Real-World Counts)

What it measures: The number of vehicles passing through specific road segments per hour.

Why it matters: This is the standard validation metric. Planners compare the "simulated volumes" against "real-world traffic counter data" (using the counts module) to prove the model is accurate before testing future scenarios.

5. Network Speed & Congestion Patterns

What it measures: Average speeds on network links at different times of day (e.g., 8:00 AM vs. 2:00 PM).

Why it matters: CitySim excels at capturing spillover effects (traffic jams growing backward). Visualizing this identifies bottlenecks where infrastructure upgrades are needed.

6. Vehicle Kilometers Traveled (VKT)

What it measures: The total distance driven by all vehicles in the system.

Why it matters: This is a direct proxy for infrastructure wear and tear, energy consumption, and overall traffic load. Reducing VKT while maintaining Agent Utility is a "Holy Grail" of urban planning.

7. Trip Distance Distribution

What it measures: How far agents travel for their activities.

Why it matters: It helps distinguish between short-distance trips (potentially walkable/cyclable) and long-distance commutes. It validates whether the "Destination Choice" module is working correctly.

8. Public Transit Ridership (Boardings & Occupancy)

What it measures: The number of agents boarding specific transit lines and the crowding levels inside vehicles.

Why it matters: Unlike static models, CitySim simulates overcrowding. If a bus is full, agents may be denied boarding (if configured), forcing them to wait for the next one or switch modes, providing realistic capacity analysis.

9. Wait Times (PT & On-Demand)

What it measures: The time an agent spends waiting for a bus, train, or on-demand vehicle (like Uber/Lyft).

Why it matters: Wait time is usually penalized more heavily in the scoring function than in-vehicle time. High wait times are a primary driver of mode shifts away from public transit.

10. Emissions (CO2, NOx, PM)

What it measures: The total pollutants generated based on vehicle trajectories, cold starts, and congestion levels.

Why it matters: Using the emissions contribution module, CitySim provides highly accurate environmental impact assessments because it accounts for "stop-and-go" waves rather than just average speeds.

CityData.ai provides highly granular temporal analysis, allowing our customers to examine mobility patterns across various time scales to identify peak periods, seasonal trends, and long-term changes. The platform offers the following time granularities from lowest to highest:

Hourly Granularity: CityFlow visualizes trip distributions across a 24-hour cycle, enabling the identification of morning and afternoon peak congestion periods.

Day-Part Granularity: Users can filter data by specific segments of the day, such as Morning, Afternoon, Evening, and Night, to analyze shift-based or event-driven movement.

Daily Granularity: The platform tracks visitation and movement trends on a 24-hour basis, facilitating direct comparisons between Weekdays and Weekends.

Weekly and Monthly Granularity: Data is aggregated into weekly and monthly summaries to provide a high-level view of regional mobility health.

Annual/Historical Granularity: Leveraging a model trained on 48 months of historical data dating back to 2019, CityData provides year-over-year longitudinal analysis to evaluate the long-term impact of infrastructure projects.

While the industry often utilizes TMC segmentation, CityData has optimized its platform to work natively and exclusively with OpenStreetMap (OSM) way IDs. This data-centric approach offers our customers several distinct advantages over traditional segmentation:

Comprehensive Functional Coverage: Unlike TMC codes, which are primarily focused on major arterials and freeways, the OSM network provides 100% coverage across the area of interest’s entire functional classification, including state routes, U.S. routes, and local rural roads often missing from traditional traffic datasets

Segment-Specific Mobility Intelligence: Every unique OSM way ID in the platform is associated with specific motorized and non-motorized metrics, such as counts, average speeds, and travel times, providing a detailed "Route Aggregation" view essential for corridor studies and rural road assessments.

Departmental Interoperability: While the internal architecture is based on OSM to ensure data-level depth, the output remains fully compatible with our customers’s GIS systems. OSM way IDs can be spatially mapped and conflated with any of the department’s internal roadway shapefiles or administrative layers.

Granular Precision: By utilizing OSM way IDs, the platform tracks performance metrics—including total trips and average speed—at a much higher resolution than standard TMC links, enabling the identification of localized bottlenecks at a block-by-block level.

CityData has optimized its platform to work natively and exclusively with OpenStreetMap (OSM) way IDs. This data-centric approach has no constraints. The platform is not limited by segment length. However, most experienced users and network cleaners recommend a minimum of 10 to 20 meters for stability for transportation simulations.

Gridlock Artifacts: Extremely short links (e.g., 8-10m) between intersections often create "gridlock cycles" where vehicles get stuck waiting for space on the next short link, which is blocked by a vehicle waiting for the first link.

Intersection simplification: In OpenStreetMap (OSM) data, complex intersections are often modeled with many tiny segments. It is standard practice to "simplify" the network by merging these clusters into a single node or a single longer link to prevent QSim errors.

Yes. CityData provides robust capabilities for high-volume data retrieval, ensuring that our customers is not limited by user interface constraints when conducting large-scale regional studies. Beyond standard dashboard exports, CityData offers the following methods for downloading large datasets. Users are provided with direct access to the back-end cloud data warehouse in Google Cloud using APIs or SQL Queries. Users can directly query and download the raw datasets from Google BigQuery. Additionally, CityData provides direct download access in AWS S3 buckets to pull massive historical movement datasets.

Mobility data within the CityFlow platform is typically refreshed on a monthly basis. This deliberate processing cycle is essential to ensure the highest fidelity in population calibration, allowing CityData to transform raw, crowdsourced data into a statistically accurate representation of the total population. By prioritizing this calibration phase, CityData provides our customers with a more reliable baseline for corridor studies, project evaluation, and long-term planning, rather than instantaneous operational alerts

CityData.ai provides 100% geographic and functional data coverage for both the the entire United States. By utilizing a data-centric approach integrated with the OpenStreetMap (OSM) network, the platform delivers comprehensive visibility across all FHWA Functional Classifications. The nationwide coverage ensures that most departments of transportation will have access to consistent mobility intelligence, supporting both local projects within a single state and multi-state corridor studies that require a unified national data baseline. The solution covers the following roadway hierarchies across the nation:

Interstate and Freeway Systems: High-fidelity analysis for the entire National Highway System (NHS), including rural and urban interstates and expressways.

U.S. and State Routes: Complete monitoring of major travel corridors and state-maintained highways, providing critical performance data for inter-regional passenger and freight movement.

Arterials and Collectors: Granular data for both principal and minor arterials, enabling precise bottleneck identification in both urbanized and rural areas.

Local Roads: Extensive coverage of the local road network, ensuring visibility into the "last mile" of travel which is often underserved by traditional hardware sensors.

Rural Connectivity: By utilizing algorithmic scale-up, the platform provides statistically significant movement profiles for low-volume rural routes in across the U.S. that typically lack physical loop detectors.

Yes. Because our movement data is a statistically significant sample—rather than 100% of actual movement—scale-up ratios and multipliers are calculated for each recorded trip. Using our predictive algorithms, we correlate our sample against ground-truth and census datasets. For example, if we record 50 trips per day between Neighborhood A and the downtown core, but our ground-truth analysis confirms the actual volume should be 100 trips, we assign a scale-up ratio of 2 for those trips. These dynamic multipliers are critical for downstream accuracy, especially if you intend to input our datasets into agent-based simulators, including CityData’s native agent-based simulator, CITYSIM.

Users can directly query and download the raw datasets from Google BigQuery using APIs or SQL queries. Additionally, CityData provides direct download access in AWS S3 buckets to pull massive historical movement datasets. Mobility data within the CityFlow platform is typically refreshed on a monthly basis. This deliberate processing cycle is essential to ensure the highest fidelity in population calibration, allowing CityData to transform raw, crowdsourced data into a statistically accurate representation of the total population. By prioritizing this calibration phase, CityData provides our customers with a more reliable baseline for corridor studies, project evaluation, and long-term planning, rather than instantaneous operational alerts

For a State Department of Transportation (DOT) RFP, our recommendation is that the evaluation of mobility data suppliers must go beyond simple "volume" metrics, and must also be scrutinized for equity (bias), usability (conflation), and privacy. Based on current Federal Highway Administration (FHWA) guidelines and industry standards for soliciting crowdsourced data (Apps, CV, IoT), here are the top 5 Key Performance Indicators (KPIs) that we recommend for our customers.

1. Demographic & Geographic Representativeness (Bias)

Why it matters: This is currently the most critical differentiator for state agencies. "Big Data" is notoriously biased (e.g., connected vehicles often skew newer/wealthier; app data often skews urban). If a DOT uses biased data to plan infrastructure, they risk violating Title VI or Justice40 equity requirements.

The KPI to ask for: Bias Correction Methodology & Correlation Coefficients.

How to evaluate:

Ask vendors to demonstrate how their data correlates with Census ACS (American Community Survey) data at the Block Group level.

Pass/Fail: Does the vendor have a documented normalization process to upweight under-represented groups (low-income, rural, elderly)?

2. Spatial Precision & Map Matching (LRS Conflation)

Why it matters: Raw GPS points are noisy. A "teleporting" car or a truck appearing to drive through a building is useless to a DOT. The data is only valuable if it snaps correctly to the Linear Referencing System (LRS) (the specific roadway network map the DOT uses).

The KPI to ask for: Conflation Rate & Positional Accuracy.

How to evaluate:

Conflation Rate: What percentage of raw data points are successfully snapped to a road segment? (Target >95%).

LRS Compatibility: Can the supplier deliver data linked to the state's specific ARNOLD (All Road Network of Linear Referenced Data) ID or TMC segment IDs, or do they only provide proprietary segment IDs?

3. Penetration Rate (Sample Size) vs. Observable Miles

Why it matters: "Billions of data points" is a vanity metric. A DOT needs to know if the data covers their specific roads, particularly rural arterials and lower-functional-class roads, not just interstates.

The KPI to ask for: Roadway Coverage Percent by Functional Class.

How to evaluate:

Do not ask for "Total Data Points." Ask for "Percent of State Roadway Miles with >5% daily sample rate."

Request separate coverage KPIs for Interstates, Arterials, and Collectors.

Metric: "Average daily vehicle observations per mile (Observations/Mile) on Functional Class 3 and 4 roads."

4. Trip Granularity & Mode Imputation

Why it matters: DOTs are moving from "Vehicle Throughput" to "Person Throughput." They need to distinguish between a vehicle, a bicycle, and a pedestrian. Crowdsourced data is often messy here.

5. Privacy Preservation & Re-identification Risk

Why it matters: State DOTs are subject to FOIA (Freedom of Information Act) requests. If a supplier provides data that allows a journalist to track a senator's car from home to the statehouse, the DOT faces a massive liability.

The KPI to ask for: K-Anonymity Threshold & Aggregation Methodology.

How to evaluate:

Privacy Floor: Does the vendor enforce a strict "minimum trip threshold" (e.g., they will not show data for a road segment unless at least x unique devices traversed it)?

Origin/Destination Fuzzing: Do they truncate trip start/end points to a census block or grid, rather than showing the exact driveway?

Metric: "Compliance with NIST privacy framework or specific k-anonymity settings configurable by the agency."

CityData proposes an "API-First" architecture designed to function as a connected "System of Systems" rather than a siloed application. Our open API support and system interoperability ensures seamless interoperability with city’s existing technical ecosystem through robust adherence to open standards and flexible data exchange mechanisms.

Open API Support

• RESTful Architecture: We prioritize the use of standard RESTful APIs for all communication between services. This maximizes flexibility and ensures that external city applications can programmatically trigger simulations or retrieve datasets without manual intervention.

• Web Services Compliance: Our solution aligns with the municipal IT architecture standards for "Compatibility/Interoperability," ensuring efficient data exchange (sending/receiving) using established Web Services standards and Open API frameworks.

• Third-Party Integration: We explicitly provide the ability to integrate third-party APIs and datasets, allowing the Digital Twin to ingest live feeds from external portals or push actionable insights back into operational dashboards. System Interoperability with Municipal Applications We utilize a multi-pronged integration strategy to connect with specific IT systems:

• ArcGIS & Spatial Data: We ensure full compatibility with ESRI ArcGIS (the standard GIS platform for most cities and counties) by supporting the direct ingestion and export of standard spatial formats including GeoJSON, Shapefiles, and KML. This allows the Digital Twin to act as a dynamic layer within the City's existing GIS environment.

• Asset Management (Cartegraph & DOTI Assets): By leveraging IFC (Industry Foundation Classes) and COBie standards, our platform can exchange rich asset data with management systems like Cartegraph . This enables the twin to reflect real-time asset status or maintenance history.

• CASR Portals (Climate Risk, Touchstone): Our architecture utilizes Google BigQuery as a centralized data warehouse that supports Standard SQL queries. This facilitates direct, high-volume data connectors to CASR’s Climate Risk and Touchstone databases for reporting on building emissions and energy efficiency without complex proprietary translation layers.

• Mobility Models (Travel Demand): The CITYSIM engine is designed to ingest standard travel demand model outputs (e.g., OD matrices) and enhance them with real-world mobility data. Conversely, simulation outputs (e.g., VMT, mode share) can be exported in CSV or Parquet formats to feed back into the city’s or county’s long-range planning models . Data Output Flexibility To ensure broad compatibility across all listed systems, we support a comprehensive range of downloadable and API-accessible formats:

• Vector/Spatial: GeoJSON, XML, DWG, PLN .

• Tabular/Analytical: CSV, JSON, Parquet .

• Building/BIM: IFC.

Yes, CityData fully supports this. Because our platform is hosted entirely on Google Cloud, we leverage Firebase Authentication and Google Cloud Identity to seamlessly manage access. Single Sign-On (SSO) and Single Log-Out (SLO) are highly configurable across typical authentication protocols, including but not restricted to: OAuth, OpenID Connect (OIDC) , SAML 2.0, WS-Trust, and WS-Federation.

For example, integrating corporate credentials (e.g., Azure AD / Microsoft Entra ID) for SSO/SLO is fully supported natively through Google Workspace and Cloud Identity, simplifying access and improving security.

Yes, CityData ensures complete encryption. All data communications with our cloud-based platform use RESTful APIs powered by Google Cloud Functions, secured with strong TLS 1.2+ (HTTPS) encryption in transit. Furthermore, Google Cloud encrypts all customer data stored at rest by default using AES-256 encryption within Firestore, BigQuery, and Cloud Storage buckets. Google's cryptographic modules and Key Management Service (KMS) are validated to FIPS 140-2 Level 1 standards or higher.

Upon ingestion, all unique identifiers coming from our first-party curated crowdsourced mobility data are subjected to immediate cryptographic hashing prior to storage or archival in BigQuery and Firestore. CITYDATA utilizes the SHA-256 cryptographic hash function. This one-way transformation ensures that original identifiers cannot be reconstructed from their hashed counterparts, effectively pseudonymizing the data before any spatial enrichment is performed using Google Maps and Google Places APIs.

Yes. CITYFLOW is built in Looker Studio and hosted 100% natively in Google Cloud. The entire application and its underlying data architecture are configured with the appropriate and effective security controls established by Google Cloud’s shared responsibility model.

• We utilize Google Cloud’s Identity and Access Management (IAM) and Virtual Private Cloud (VPC) Service Controls to enforce a strict "Least Privilege" model, ensuring only authorized services and personnel access specific resources.
• Our infrastructure provider, Google Cloud, undergoes rigorous third-party audits and is certified compliant with SOC 2 Type II, ISO 27001, FedRAMP, and PCI-DSS.
• CityData conducts annual internal risk assessments and quarterly vulnerability scans of our application code and APIs (e.g., OWASP Top 10 vulnerabilities).

Yes. We enforce Separation of Duties (SoD) through strict Role-Based Access Control (RBAC) in Google Cloud IAM:

• Developers: Work exclusively in "Development" and "Staging" environments utilizing dummy data; they do not have access to production City data.
• Operations: Only a limited number of vetted Senior Operations Engineers have access to the production environment for troubleshooting purposes. Their access is temporary, strictly audited, and requires specific approval.

Yes. Because our solution utilizes Serverless technologies (Google Cloud Functions, App Engine, Firestore), the underlying operating systems and runtime environments are fully managed and automatically patched by Google. This removes the risk of unpatched servers and ensures we are always running on the latest, most secure versions. We strictly monitor our dependencies and do not utilize any software or systems that have reached their End of Life (EOL).

No. Our primary strategic partner is Google (Google Cloud Platform), which provides the hosting infrastructure and data center services. We do not utilize any other third-party subcontractors for application development, data processing, or system operations. All application logic and support are handled by CityData's direct employees.

Yes. We configure all our Google Cloud resources (Compute Engines, BigQuery datasets, Firestore instances) to reside exclusively in US-based regions. This ensures your data never leaves the physical borders of the United States. Furthermore, we enforce a strict policy prohibiting the storage of City data on local laptops, USB drives, or mobile devices. All data remains within the secure cloud environment and is accessed via secure web portals or virtualized sessions.

CityData maintains a formal Incident Response Plan (IRP) that outlines specific phases: Detection, Analysis, Containment, Eradication, Recovery, and Post-Incident Activity. In the event of a confirmed security breach affecting City data, our policy is to notify the City's designated Point of Contact within 24 hours of discovery.

To date, CityData has not experienced a single security breach or data compromise since our inception.In the unlikely event of a breach, CityData carries a specialized Cyber Liability Insurance policy (limits available on request) to cover data breach response, legal fees, notification expenses, and credit monitoring.

Yes. We adhere to a rigorous data lifecycle management process aligned with NIST 800-88 standards for media sanitization. Upon the conclusion of the project or termination of the contract:

• Logical Deletion: We issue deletion commands to our active databases (Firestore) and file storage (Cloud Storage buckets) within 30 days of the request.
• Cryptographic Erasure: We rely on Google Cloud's standard deletion process, which removes the encryption keys associated with your data, rendering it cryptographically unreadable and effectively destroyed.
• Physical Sanitization: Google's underlying storage media is tracked and physically destroyed (shredded or crushed) when retired.

Yes, CityData has a comprehensive BCDR plan focused on CITYFLOW. Because our environment is hosted entirely within Google Cloud, this plan ensures that operations can continue seamlessly and recover quickly from any disruption or disaster.

• SLA Metrics: We adhere to a Recovery Point Objective (RPO) of 1 hour (maximum data loss in a disaster) and a Recovery Time Objective (RTO) of 4 hours (maximum time to restore service).
• Testing: We perform quarterly disaster recovery drills and have achieved a 100% success rate in restoring services within our SLA limits over the last 12 months.

Yes, data preservation is highly automated within our Google Cloud environment:

• Auto-Backups: CITYFLOW relies on Google Cloud’s automated snapshot and backup schedules. All backups are stored in geo-redundant Cloud Storage buckets and are encrypted at rest using AES-256.
• Archival: Historical data is continuously archived into Google Cloud Storage (Archive class) for highly durable, long-term storage.

Yes. We follow a structured Change Management Policy. For routine updates, we use "rolling deployments" to ensure zero downtime. For major upgrades that require a scheduled maintenance window, we provide the City with at least 72 hours' advance notice via email and in-app banners, detailing the expected duration and impact.

If an unexpected service interruption occurs, we notify the City's technical contacts within 30 minutes, provide hourly updates, and issue a formal Root Cause Analysis (RCA) once resolved. Routine platform maintenance is run during non-business hours to maintain our 99% uptime SLA.

Yes. We conduct comprehensive background checks on all employees and long-term contractors as a condition of employment, including criminal history, education verification, and employment history. We have a zero-tolerance policy for crimes involving dishonesty, fraud, or theft, and do not employ individuals with such convictions. Furthermore, the City has the right at any time to require the immediate removal of any CityData representative deemed detrimental to the working relationship.

Yes. Our platform utilizes Google Cloud Logging and Cloud Monitoring to capture granular telemetry. We capture user login events, failed authentication attempts, API calls, and administrative actions. We can configure automated exports of these logs to a secure storage bucket accessible to the City or provide ad-hoc reports in CSV/JSON format.

Yes. In addition to quarterly internal vulnerability scans of our application code and APIs (which scan for OWASP Top 10 vulnerabilities), CityData engages independent, third-party cybersecurity firms to conduct annual external penetration testing of our platform and cloud infrastructure. Executive summaries or letters of attestation for these tests can be provided to the City upon request under a Non-Disclosure Agreement (NDA).

Yes. All CityData employees and long-term contractors are required to complete comprehensive security awareness and data privacy training during onboarding. Furthermore, mandatory refresher courses are conducted annually. This training covers critical topics including phishing awareness, secure remote access, strict data handling procedures, and incident reporting protocols.

Yes. While CityData strictly prohibits the downloading or storage of City data on local devices, all company-issued laptops and workstations used by our staff are secured and centrally managed via Mobile Device Management (MDM) solutions. Security enforcement includes full-disk encryption, automated OS and application patching, next-generation antivirus (NGAV)/Endpoint Detection and Response (EDR), host-based firewalls, and remote-wipe capabilities in the event a device is lost or stolen.

Yes. CityData maintains a formal Data Classification and Handling Policy. Although we strictly do not collect Personally Identifiable Information (PII) from citizens, all aggregated mobility datasets, platform configurations, and customer-provided "ground truth" data are classified internally as "Confidential." Access, transmission, and storage controls are strictly mapped to this classification level to ensure data is handled with the highest level of care.

Yes. All CityData APIs are secured using Google Cloud API Gateway and Cloud Armor (Web Application Firewall). We enforce strict authentication via Firebase tokens, implement rate limiting to prevent Distributed Denial of Service (DDoS) and brute-force attacks, and continuously monitor API traffic for anomalous behavior.

Yes. Because our data architecture relies entirely on Google Cloud’s serverless databases (BigQuery and Firestore), we benefit from robust, built-in ransomware protections. We utilize immutable, point-in-time snapshots and can enforce Write-Once-Read-Many (WORM) storage configurations for our automated backups in Google Cloud Storage. This ensures that even in the unlikely event of a malicious intrusion, historical backups cannot be altered, encrypted, or deleted by unauthorized actors.

Yes. As part of our annual Business Continuity and Disaster Recovery (BCDR) review process, CityData conducts a formal Business Impact Analysis (BIA). This process identifies our mission-critical cloud systems, evaluates the potential operational impact of various disruption scenarios (e.g., regional cloud outages), and validates that our established Recovery Time Objectives (RTO) and Recovery Point Objectives (RPO) are sufficient to maintain service continuity for our municipal clients.

Yes. While CityData’s platform relies exclusively on Google Cloud Platform (GCP) for hosting and infrastructure, we maintain a formal Third-Party Risk Management (TPRM) policy for any internal corporate tools. Any new software, development tool, or service introduced into our environment is rigorously vetted for SOC 2 Type II compliance, encryption standards, and strict adherence to our internal privacy and security policies before procurement and implementation.