Location Data, Explained‍

When we talk about data, the conversation often comes back to quality. It makes sense. We use data to analyze behavior, generate insights, and power products. The input fuels the output, so the input had better be good.

For mobility data, quality means reliable location signals, enough observations to support the analysis, coverage where you need it, and consistency over time. Understanding where the data comes from and how it’s checked also helps you assess whether it’s right for your work.

Of course, the requirements depend on the use case. Measuring visits to a small storefront calls for different data than studying movement across a region. So what should you look for when evaluating mobility data quality?

‍

What Determines Mobility Data Quality?

1. How Accurate Are the Location Signals?

Location accuracy describes how close a reported position is to a device's actual location. Some datasets include a horizontal accuracy field, which estimates uncertainty around the coordinates in meters. That estimate helps you assess a signal, but it doesn't guarantee the point is correct. Reported accuracy can itself be wrong. Some feeds report precise-looking values on fabricated points. More decimal places in a coordinate don't make it more accurate, either. 

Think about two stores next door to each other. A point that places a device on the right block may still be too imprecise to tell you which store it visited. 

Tip: Ask whether accuracy fields are available and how less reliable signals are handled.

2. Are There Enough Observations at the Right Times?

As we explain in How Mobile Location Data Actually Works, location data consists of intermittent observations. There will be gaps.

A device with only one or two pings a day can’t support visit or trip analysis. A single point near a store may not establish a visit. Several observations over a certain period can provide stronger evidence of a visit and help estimate how long a device stayed.

Tip: Look at how observations are distributed across devices and throughout the day. Check when signals were collected, not only when they were delivered.

3. Does the Data Cover the Places and People You Need?

Coverage tells you where and how much data is available. Representativeness tells you how well that sample reflects the population you're studying.

A dataset with millions of devices can still have limited coverage in your market. It can also underrepresent certain groups because participation depends on smartphone ownership, app usage, and location permissions.

Tip: Review coverage for your specific geography and time period. If you're estimating behavior across a population, ask how sampling differences are assessed and whether adjustments are needed. A larger sample won't automatically correct those differences.

4. Can You Make Fair Comparisons Over Time?

If observed store visits fall from one month to the next, did fewer customers visit, or did fewer devices appear in the dataset?

Changes in sources, device participation, or processing can affect the trends you see. Comparing weeks, seasons, or years requires understanding those changes.

Tip: Ask how the provider monitors device counts, coverage, and observation frequency over time, and how it communicates changes that could affect your analysis.

5. What Gets Flagged or Removed?

Location datasets can contain duplicate records, missing fields, invalid coordinates, and suspicious movement patterns. Providers use quality checks to identify and address these issues.

Common checks include:

·       Removing duplicate observations

·       Validating coordinates and timestamps

·       Flagging implausible movement between points

·       Filtering signals that fail defined quality requirements

Tip: Ask which checks are applied and what happens to records that fail. A flag can let your team decide how to handle a signal, while removing it changes the dataset you receive. Stricter filters can improve confidence in the retained data while reducing coverage. Ask how those filters affect coverage for your use case. 

6. Where Does the Data Come From?

Collection methods affect signal accuracy, frequency, and availability. A provider should explain its sourcing approach and how it evaluates suppliers, even if individual relationships are confidential. 

Tip: Ask whether you’re receiving individual location observations, derived visits or trips, or aggregated insights — and what processing has been applied. Ask how consent requirements and opt-outs are handled, and what restrictions apply to sensitive locations or intended uses.

How Do You Test Mobility Data Quality?

Test a sample against the work you actually plan to do. Ask providers for a sample or limited trial that covers the geography, time period, and level of detail you need. Our mobility data buyer's guide has more questions to use when comparing providers.

Focus your evaluation on the requirements of your use case:

·       Retail visitation: Check location accuracy, observation timing, store boundaries, and how visits are defined.

·       DOOH planning: Review coverage around potential placements during the days and hours relevant to the campaign.

·       Transportation research: Examine coverage, gaps between observations, and representation of the populations you're studying.

·       AI and modeling: Review missing values, sampling differences, historical consistency, and whether information is observed or inferred.

How Veraset Approaches Data Quality

Veraset brings together billions of location observations from a diverse network of data partners. Our data spans 300 million devices globally, with 10 billion daily location observations. We evaluate data providers against six criteria:

1. Scale and incrementality: consistent volume that adds devices we don’t already have
2. Coverage: geographic reach in the markets our customers need
3. Quality: accurate, dense, timely location signals
4. Identity: valid, persistent device identifiers
5. Reliability: on-time delivery and stable volume over time
6. Compliance: documented consent, sensitive-location handling, and resale rights. These are hard requirements, so a supplier that misses any of them doesn’t make it in.

Source evaluation is one part of the process. Our ongoing quality process also includes:

1. Quality checks and filtering: We remove duplicate pings and score every observation for coordinate precision and suspicious patterns, such as many unrelated devices reporting the same coordinate. Suppliers are also screened for implausible volumes and timestamps.
2. Monitoring and validation: Every daily run produces automated reports and alerts, so issues surface the same day. We also track each supplier’s delivery consistency and volume over time.
3. Privacy and supplier governance: Data is keyed to a pseudonymous ID, not raw ad IDs. Opt-outs are applied across all feeds. In the US, observations at sensitive locations such as healthcare facilities, schools and places of worship are removed before any product is built.

Get in touch with Veraset to discuss your use case and evaluate a data sample.

For more mobility data insights, subscribe to our monthly newsletter The Signal.

‍

Choosing a mobility data provider isn't just about comparing datasets.

Whether you're building audiences, measuring campaign performance, selecting new retail locations, or powering research and AI models, the quality of your decisions depends on choosing the right data, and the right partner.

Yet many evaluation processes focus on the wrong criteria. Buyers compare scale, coverage, or pricing before they've clearly defined their use case or goals.

After years of working with agencies, retailers, researchers, and technology partners, we've seen the same evaluation mistakes surface time and again.

Here are five of the most common mistakes (and how to avoid them).

Mistake #1: Starting with the Data Instead of the Business Problem

One of the biggest mistakes buyers make is evaluating providers before they've clearly defined what they're trying to accomplish.

Different use cases require different strengths.

For example:

• Audience creation may prioritize scale and consistency.
• Retail site selection may depend on geographic precision and long-term visitation trends.
• Attribution requires reliable methodology and confidence in movement patterns.
• AI applications often demand flexible access to large, well-structured datasets.

Without first defining your objectives, it's easy to optimize for the wrong criteria — choosing a provider based on the largest dataset, the lowest price, or the longest feature list, rather than the one that's best suited to your specific use case.

Before evaluating vendors, ask yourself:

• What business question am I trying to answer or solve?
• Who will use this data, and how?
• What does success look like?
• Which capabilities are essential and which are nice to have?

Once that answer is clear, evaluating providers becomes much more straightforward. Instead of searching for the "best" mobility data provider, you'll be looking for the provider that's best equipped to help you achieve your goals.

Mistake #2: Assuming All Mobility Data Is Created Equal

Not all mobility data is collected, processed, or delivered in the same way, and those differences can have a meaningful impact on the insights you generate.

Providers may offer: 

• Raw location observations
• Processed or modeled datasets
• Aggregated insights
• Specialized products built for specific workflows

None of these approaches is inherently better than another. The right choice depends on what you're trying to accomplish, who will be using the data, and how much flexibility or customization your team requires.

As you evaluate providers, consider questions like:

• How is the data collected and sourced?
• What processing or modeling has been applied?
• How frequently is the data refreshed?
• What quality assurance and validation processes are in place?
• Does the provider offer the data in a format that fits my workflow—whether that's raw data, APIs, or ready-to-use insights?

Understanding these differences will help ensure you're evaluating providers based on what matters most for your business, rather than assuming all mobility data delivers the same value.

Mistake #3: Confusing More Data with Better Data

Bigger is not always better. A larger dataset isn't valuable if it lacks the precision, consistency, or quality required for your application.

If you're building national audiences, training AI models, or conducting broad market analysis, having a large, representative dataset can be a significant advantage. In these scenarios, greater scale often translates into more robust insights and better statistical confidence.

But there are many cases where simply having more data won't improve your results.

For applications like visitation analysis, site selection, attribution, or localized market intelligence, factors such as precision, consistency, and data quality can be just as important — if not more so — than the total number of devices.

Instead of asking, "Who has the biggest dataset?" consider questions like:

• Is the dataset representative of the population I'm trying to understand?
• Is coverage consistent across the geographies that matter most to me?
• How is data quality monitored and maintained?
• How frequently is the data refreshed?
• Does the provider have the scale my specific use case requires?

The best mobility data provider for you is the one with the right combination of scale, quality, and consistency for your business objectives.

Mistake #4: Treating Privacy as a Checkbox

Privacy has become one of the most important considerations when evaluating mobility data, but it's about much more than checking a compliance box.

As regulations continue to evolve and consumer expectations around data transparency increase, organizations need confidence that the data they're using has been collected and managed responsibly. A provider's approach to privacy doesn't just impact legal compliance—it can also influence data quality, business continuity, and long-term trust.

When evaluating mobility data providers, ask questions such as:

• How is user consent obtained and respected?
• How is data collected and sourced?
• What governance and quality controls are in place?
• How does the provider adapt to changing privacy regulations?
• What processes exist to ensure responsible data stewardship over time?

The strongest providers view privacy as an ongoing commitment. They invest in transparent practices, continuously monitor the evolving regulatory landscape, and build privacy considerations into the way they collect, manage, and deliver data.

Ultimately, a thoughtful approach to privacy benefits everyone. It helps organizations reduce risk, maintain confidence in their data strategy, and build solutions that are sustainable as the industry continues to evolve.

Mistake #5: Assuming There's Only One Way to Access Mobility Data

One of the most overlooked aspects of evaluating a mobility data provider is how you'll actually access and use the data.

Organizations have different technical capabilities, workflows, and business needs. A data science team building custom models has very different requirements than a marketing team looking for audience insights or an analyst exploring visitation trends.

That's why the best providers offer multiple ways to work with their data.

For example, at Veraset, customers can choose the approach that best fits their organization:

• Flat Files for organizations that want direct access to raw data for custom analysis, modeling, and integration into their own environments.
• APIs for teams that need real-time or on-demand access to mobility data within existing applications and workflows.
• Self-Service Platform for business users who want to explore mobility insights, visualize trends, and answer questions without relying on engineering resources.

No single delivery method is inherently better than another. The right choice depends on your team's technical expertise, your existing technology stack, and the business problems you're trying to solve.

As you evaluate providers, consider questions like:

• Does this provider support the way my team prefers to work?
• Can different teams access the data in the ways they need?
• Will this delivery model scale as our use cases evolve?
• If our needs change, will we have to switch providers or simply use a different product?

The best mobility data partner isn't just one that delivers high-quality data. It's one that can deliver that data in the format that creates the most value for your organization today and as your business grows.

What Great Evaluations Have in Common

Successful buyers don't begin by comparing vendors; they begin by defining success. They ask thoughtful questions about their business objectives, data quality requirements, privacy expectations, and operational workflows before narrowing their list of providers.

That approach consistently leads to stronger partnerships and better business outcomes.

If you're evaluating mobility data providers, we'd be happy to help you think through the process. Whether you're comparing vendors, validating an existing solution, or exploring a new use case, our team can help you ask the right questions before making a decision.

Get in touch to discuss your use case.

‍

For more mobility data tips and tricks, subscribe to our monthly newsletter The Signal.

‍

As geospatial data and mobility intelligence become increasingly important across industries, organizations need scalable ways to organize, analyze, and visualize location data.

That’s where geospatial indexing systems like H3 and Geohash come in.

Both H3 and Geohash are methods for dividing the world into smaller geographic cells that make location data easier to process. These systems are foundational to modern geospatial analytics, powering everything from mobility intelligence and urban planning to logistics, mapping, advertising, and machine learning.

While H3 and Geohash solve similar problems, they work differently and are often optimized for different use cases. Choosing the right approach depends on how geographic data will be stored, queried, and analyzed.

In this article, we’ll break down:

• What H3 and Geohash are
• How they work
• The key differences between them
• When to use each system
• Why H3 has become increasingly popular for mobility and geospatial analytics

What Is a Spatial Index?

Before comparing H3 and Geohash, it’s important to understand the problem they solve.

Geospatial datasets are massive. Modern mobility datasets can contain billions of GPS signals, visits, trips, or movement events spread across the globe. Efficiently storing, querying, aggregating, and analyzing that data requires a way to organize geographic space into manageable units.

A spatial index is a system that divides the earth into smaller cells or grids so geographic coordinates can be grouped together efficiently. Instead of analyzing millions of raw latitude and longitude points individually, systems can aggregate data into spatial cells for:

• Faster queries
• More scalable analytics
• Improved visualization
• Pattern detection
• Spatial joins
• Geographic aggregation

This becomes especially important in mobility intelligence, where organizations analyze movement patterns, visitation trends, trade areas, dwell behavior, and population flows at scale.

What Is Geohash?

Geohash is a geographic encoding system that converts latitude and longitude coordinates into short alphanumeric strings.

Developed in the early 2000s, Geohash works by recursively dividing the world into rectangular grid cells. Each additional character in a Geohash represents a more precise geographic area.

One of Geohash’s biggest advantages is simplicity. Nearby locations often share similar prefixes, making Geohash useful for:

• Database indexing
• Location search
• Basic proximity queries
• Lightweight geospatial applications

Because Geohash outputs compact strings, it became widely adopted in early mapping and location systems. While Geohash is widely used for indexing and search, other frameworks may be better suited for certain analytical and mobility-focused workflows.

What Is H3?

H3 is a hierarchical hexagonal spatial indexing system originally developed by Uber Technologies for large-scale geospatial analysis.

Instead of rectangular cells, H3 divides the world into hexagons. This difference is important. Hexagons provide more consistent adjacency and distance relationships than squares or rectangles, making them useful for:

• Mobility analysis
• Spatial clustering
• Traffic analysis
• Routing
• Catchment modeling
• Movement visualization
• Machine learning applications

H3 is hierarchical, meaning each hexagon can be subdivided into smaller hexagons at increasing resolutions. This allows organizations to analyze data at multiple geographic scales while maintaining consistency across datasets.

Today, H3 is widely used across geospatial engineering, mobility intelligence, transportation, logistics, and urban analytics, particularly where spatial analysis and aggregation are important.

Geohash vs H3: Key Differences

‍

1. Cell Shape

The most obvious difference is the grid shape itself. Geohash uses rectangular cells, while H3 uses hexagonal cells. This matters because rectangular grids create uneven neighbor relationships. Distances between adjacent cells can vary depending on direction, especially near the poles.

Hexagons create more uniform adjacency:

• Each cell has more consistent neighboring relationships
• Distortion is reduced
• Movement modeling becomes more natural
• Spatial smoothing and clustering improve

For mobility and movement analysis, this often produces cleaner and more accurate results.

2. Spatial Consistency

H3 is often preferred for modeling geographic relationships because hexagons can represent movement and proximity patterns more consistently than rectangular grids.

This becomes especially important in:

• Human mobility analysis
• Trade area modeling
• Traffic flow analysis
• Route optimization
• Urban movement studies

Geohash is commonly used for indexing and lookup workflows, while H3 is frequently chosen for movement analysis and other spatial analytics applications.

3. Hierarchical Structure

Both systems support hierarchical indexing, but H3’s hierarchy is specifically optimized for scalable geospatial analysis. Each H3 resolution level nests consistently into larger parent cells, making aggregation and multi-resolution analysis easier.

For example:

• A city-wide mobility analysis may use larger hexagons
• Neighborhood-level analysis may use smaller hexagons
• Both datasets remain compatible within the same indexing framework

This flexibility is one reason H3 has become increasingly common in modern mobility intelligence platforms.

4. Visualization Quality

H3’s hexagonal structure typically creates more visually balanced maps and heatmaps.

Rectangular grids can introduce directional bias or visual artifacts, particularly when aggregating movement patterns.

Hexagons tend to:

• Reduce edge effects
• Improve spatial smoothing
• Produce more natural-looking density visualizations
• Better represent continuous movement patterns

This is particularly valuable when visualizing visitation trends, audience movement, or transportation flows.

5. Performance and Scalability

Both systems are highly scalable, though they were designed with different priorities in mind. Geohash is often used for efficient indexing and search, while H3 was built for large-scale geospatial analysis.

Organizations working with the following often prefer H3 because of its efficient indexing and analytical flexibility:

• Billions of GPS signals
• Real-time movement analysis
• Large mobility datasets
• Distributed geospatial systems

As geospatial datasets continue growing, scalable spatial indexing frameworks are becoming increasingly important for modern data infrastructure.

When Should You Use Geohash?

Geohash may be a strong fit when:

• Simplicity is important
• Lightweight indexing is sufficient
• Applications rely on string-based lookup systems
• Basic geographic search is the primary use case

It remains widely used across:

• Databases
• Mapping systems
• Search applications
• Lightweight geospatial tooling

Reach for Geohash when: 

• You need a simple, lightweight way to index and search geographic data
• Your workflows rely on database lookups or proximity searches
• Interoperability and broad ecosystem support are priorities
• You want a straightforward hierarchical structure for location data

When Should You Use H3?

H3 is often better suited for:

• Mobility intelligence
• Human movement analysis
• Spatial modeling
• Large-scale analytics
• Trade area analysis
• Transportation analytics
• Audience intelligence
• Machine learning workflows
• Geospatial visualization

Reach for H3 when:

• You're analyzing movement, visitation, or geographic relationships
• You need to aggregate data across multiple geographic scales
• Visualization quality and spatial consistency are important
• You're working with large-scale analytics or mobility datasets

Because H3 handles adjacency and aggregation more naturally, it has become increasingly common in advanced geospatial and mobility applications.

‍

Why H3 Is Commonly Used in Mobility Analytics

Modern mobility datasets are fundamentally about movement:

• Where people travel
• How populations flow between regions
• How visitation changes over time
• How audiences move through physical space

Analyzing these behaviors requires spatial frameworks that can handle:

• Scale
• Resolution changes
• Proximity relationships
• Clustering
• Geographic aggregation

Because of its approach to aggregation, adjacency, and multi-resolution analysis, H3 has become a popular choice for many mobility intelligence and geospatial analytics workflows.

‍

What we use at Veraset

At Veraset, we use both H3 and Geohash because each serves a different purpose within location data workflows. Rather than viewing one as better than the other, we apply each where it provides the most value.

• Geohash supports efficient indexing, spatial matching, and data delivery workflows.
• H3 is provided alongside movement data to support spatial analysis, aggregation, and mobility use cases.
• Different resolutions are used depending on the level of geographic detail required, from broad regional analysis to highly granular movement patterns.
• Together, H3 and Geohash help balance efficient data processing with flexible geospatial analytics.

Final Thoughts

Both H3 and Geohash are powerful spatial indexing systems, and each has valuable use cases.

Geohash remains a lightweight and effective option for many geographic search and indexing workflows.

But as mobility intelligence, geospatial analytics, and large-scale movement analysis continue to evolve, H3 has emerged as a preferred framework for many modern geospatial applications due to its flexibility, scalability, and spatial consistency.

As organizations increasingly rely on location intelligence to power analytics, decision-making, and operational workflows, understanding the differences between these spatial indexing systems becomes increasingly important.

Whether you're building mobility models, analyzing visitation trends, optimizing logistics, or powering geospatial applications, choosing the right spatial framework can significantly improve how effectively geographic data is organized and analyzed.

‍

For more on mobility data, subscribe to our monthly newsletter The Signal.