SafeGraph sells Places, Geometry and Spend transaction data. Its Patterns foot traffic products went to Advan in 2022. Alternatives by job: Foursquare, Overture, Dataplor and Precisely for places; Xtract.io and Echo Analytics for polygons; Advan, Veraset, Unacast or Placer.ai for foot traffic; GSDSI to join POI with mobility, CTV and identity. Test on your own locations.
TL;DR
- Choose GSDSI when you want POI alongside consumer, B2B, identity, CTV, and mobility data in one directly licensed catalog, with schema guidance for custom joins.
- Choose SafeGraph when your need is a specialized, packaged global Places POI product with an established schema, monthly open and close updates, and marketplace delivery.
- In a pilot, validate POI coverage and freshness against your own seed, polygon quality, and how visitation is attributed, not just row counts.
What SafeGraph is, and what it is best for
SafeGraph builds a global database of points of interest and refreshes it every month. Its Places dataset carries the basics a location team needs for each place (name, street address, latitude and longitude, NAICS category, brand and open hours), and SafeGraph says it builds the file with machine learning, web crawling and third-party licensing.
On top of Places, SafeGraph sells Geometry, which adds polygon footprints in standard WKT format and a spatial hierarchy that shows whether a place sits inside another place, shares a building, or has an attached parking lot. Parking lot polygons are a premium product and, per SafeGraph, built for the US. Delivery runs through the data platforms many teams already use, including Snowflake, the AWS Marketplace and Databricks.
That makes SafeGraph a good fit when the job is place truth. If you need a clean, well documented POI table with polygons for site selection, geocoding, mapping or as the base layer for your own visit model, it is a specialist built for exactly that.
Sources: SafeGraph data sourcing process; SafeGraph Geometry documentation; SafeGraph Parking Lots; SafeGraph data partners
Why buyers look for SafeGraph alternatives
The reasons usually come down to scope rather than quality. The ones we hear most often:
- They want foot traffic, not only places. SafeGraph sold its Patterns business (the Weekly, Monthly and Neighborhood Patterns visit products) to Advan Research in November 2022, so visit counts now come from a different vendor.
- POI is one input among several. A trade area study or an attribution model may also need device mobility, consumer attributes, purchase signals or CTV exposure, and buying each from a different vendor means several contracts and several join problems.
- They want direct licensing terms with a provider who will scope the file to the use case, rather than a standard marketplace listing.
- They are testing open data first. Open POI datasets now exist, and some teams want to see how far those get them before paying.
SafeGraph alternatives in 2026, by what they replace
SafeGraph now sells Places and Geometry, plus Spend aggregated transaction data, so the useful question is which part of its lineup you need to replace. Each row summarizes what the provider's public pages say, listed in the sources under this table. Where we could not confirm something, the cell says so. GSDSI sells POI and mobility data, so read this as a vendor's guide: the order is grouped by job, not ranked. For coverage counts, see our POI provider comparison.
| Alternative | Replaces | Polygons | What you get | Best fit (our view) |
|---|---|---|---|---|
| Foursquare Places | Places | Not mentioned on its product page | Commercial POI with rich attributes, by API or flat file | POI for apps and analytics that need many attributes per place |
| Foursquare Open Source Places | Places, free | Not confirmed | An open global POI dataset under Apache 2.0 | A free baseline and a common category taxonomy |
| Overture Maps Places | Places, free | Point geometry; buildings are a separate theme | Open data under CDLA Permissive 2.0 and Apache 2.0; its docs note that records missing key fields are harder to deduplicate | A free baseline you are ready to clean yourself |
| Dataplor | Places | States it offers polygon data | Global places data in a uniform schema with continuous updates | Coverage outside the US |
| Precisely World Points of Interest Premium | Places | Not confirmed | Monthly POI updates keyed to the PreciselyID | Enterprise teams that join POI to addresses and property data |
| Echo Analytics | Places and Geometry | States POIs come with associated shape data | Branded POI, with samples on the Snowflake Marketplace | Brand-level POI that needs shapes |
| Xtract.io | Geometry | States handcrafted brand polygons | POI with handcrafted brand polygons and attributes, with a free sample | Polygons for brand chains |
| Advan Research | Patterns (foot traffic) | States it has manually geofenced over 7 million locations | Monthly and Weekly Patterns+ feeds; Monthly is US only, Weekly adds Canada | The direct successor to SafeGraph Patterns |
| Veraset | Patterns (raw) | Not mentioned on its home page | Data by API, flat file, or delivered into your S3 or GCP warehouse | Data teams that build their own visit models |
| Unacast | Patterns | Accepts a polygon you submit and returns visit counts | Foot traffic insights by on-demand API or bulk data delivery | Foot traffic by API, with a UI if you need one |
| Placer.ai | Patterns, as analytics | Not confirmed | An analytics platform; the API is a paid add-on, and Premium Export delivers CSV or Parquet files | Teams that want dashboards rather than a file |
| GSDSI | Places, Geometry, plus mobility | Polygon boundaries on the POI feed | Directly licensed POI, joinable to mobility, CTV and identity feeds under one set of terms | Teams blending POI with other data in their own warehouse |
Sources: Foursquare Places; Foursquare on FSQ Open Source Places; Overture Maps Places guide; Dataplor global dataset; Precisely World Points of Interest Premium; Echo Analytics places data; Echo Analytics on the Snowflake Marketplace; Xtract.io POI data; Advan Research FAQ; Advan Weekly Patterns+ documentation; Veraset; Unacast foot traffic data; Placer.ai Premium Export documentation; Placer.ai API getting started; Advan announcement on acquiring SafeGraph Patterns
SafeGraph and GSDSI side by side
SafeGraph details below come from SafeGraph documents cited above; GSDSI details come from our own product pages.
| Question | SafeGraph | GSDSI |
|---|---|---|
| Core focus | Global points of interest (Places) and polygon Geometry | Multi-source catalog where POI is one feed among consumer, B2B, identity, CTV and mobility data |
| Refresh | Monthly Places and Geometry updates | Defined per feed and written into the license |
| Polygons | WKT polygons with spatial hierarchy; US parking lots as a premium add-on | Polygon boundaries, NAICS and brand hierarchy on the POI feed |
| Foot traffic | Patterns visit products sold to Advan Research in 2022 | Device-level and aggregate mobility feeds you can join to POI yourself |
| Delivery | Snowflake, AWS Marketplace, Databricks and other partners | S3, SFTP, Snowflake share or API |
| Best for | Teams whose whole requirement is a documented POI base layer | Teams blending POI with other data types in their own warehouse |
Where GSDSI fits
- You need POI polygons, NAICS and brand rollups alongside mobility or CTV outcomes, and you want to join them in your own warehouse.
- You want one set of license terms across several location and audience feeds.
- You need location governance documents (sensitive-location exclusions, retention, deletion) in the contract packet.
- You want a matched-sample pilot against your own store list before any commitment.
Where GSDSI does not fit
Some buyers are better served elsewhere, and it helps to know that early.
- If POI is the entire requirement and you want a self-serve marketplace listing you can subscribe to today, a POI specialist is the simpler buy.
- If you need a packaged foot traffic dashboard with standard trade area reports, a visitation analytics vendor fits better than raw feeds.
- If budget is the main constraint and approximate place data is good enough, start with an open dataset and come back when you hit its limits.
How to evaluate a SafeGraph alternative
Run every candidate through the same checks, on your own locations.
- Coverage and accuracy against your own store or site list, in the geographies you care about, not a headline place count.
- Freshness: how openings, closings and polygon edits are detected, and how fast they reach the file.
- Polygon quality in dense retail strips, malls and mixed-use buildings, where centroids fail.
- Brand and category mapping (NAICS or the vendor taxonomy) and how often it changes.
- Whether you need POI alone or POI joined with mobility, consumer or CTV data you would otherwise buy separately.
- Governance: permitted use, retention, deletion, downstream sharing and sensitive-location handling.
FAQ
- What are the main SafeGraph alternatives?
- It depends which SafeGraph product you are replacing. For places: Foursquare, Dataplor, Precisely, or the free Foursquare Open Source Places and Overture Maps datasets. For polygons: Xtract.io or Echo Analytics. For foot traffic: Advan Research, which acquired the SafeGraph Patterns products in 2022, or Veraset, Unacast and Placer.ai. For POI licensed together with mobility, consumer and CTV data: GSDSI.
- Does SafeGraph still sell foot traffic data?
- SafeGraph sold its Patterns business, which included the Weekly, Monthly and Neighborhood Patterns products, to Advan Research in November 2022. SafeGraph continues to publish Places and Geometry.
- Is GSDSI an alternative to SafeGraph for POI data?
- For buyers who want POI as part of a broader licensed catalog, yes. SafeGraph is a POI specialist; GSDSI licenses POI polygons directly alongside consumer, B2B, identity, CTV and mobility data, with schema guidance for joins.
- Is open POI data good enough?
- Sometimes. Open datasets are a sensible starting point for mapping and prototypes. Check coverage, category mapping and polygon quality on your own locations before you rely on them for site selection or measurement.
- What proof should we request from any POI provider?
- Coverage against your own site list, open and close freshness, polygon edit logs, category mapping, and retention and deletion terms. Pair these tests with the evaluation checklists on /resources/evaluation-checklists.