If you have shopped for location data recently, every vendor describes its data as privacy-safe or privacy-compliant. When you dig in, answers vary wildly. Some mean they delete coordinates after aggregating. Others mean a terms-of-service clause exists somewhere. A few have infrastructure to back the claim. After the FTC's location-data enforcement against X-Mode, InMarket, and Mobilewalla, the bar is operational evidence — not marketing. GSDSI's Privacy Center documents three non-negotiable controls every credible provider should show buyers licensing global mobility or building audience targeting from visitation. Procurement and marketing teams should keep public product claims aligned with tested specs — see AI search readiness for B2B data sites for crawl and schema discipline.
Privacy-safe location data means three testable controls: documented consent chain, opt-out and GPC propagation to your licensed copy, and default sensitive-location exclusion — not a policy adjective without artifacts.
Privacy-safe became a checkbox after FTC location orders — but checkboxes do not survive security review. Buyers need testable controls they can re-run annually: consent chain artifacts, opt-out propagation proof, and sensitive-location QA with zero hits on a supplied test list. Vendors that answer with policy PDFs alone are asking you to carry enforcement risk downstream.
Every device in the dataset should trace to an app where the user actively opted in to location sharing through a consent management platform aligned with the IAB Transparency & Consent Framework. The diagnostic question: can the vendor produce the source-app list and consent-architecture diagram? If not, the consent story is untestable and procurement should pass. Pair with 5 questions before licensing a MAID feed and MAID Feed specs.
Honoring Global Privacy Control signals, processing DSARs within applicable timelines, and maintaining suppression lists that flow through the entire supply chain. If a consumer opts out in one app, that opt-out should follow the device across downstream uses — not only suppress the front-door sale. California AG CCPA guidance makes propagation explicit for processors. GSDSI's Do Not Sell flow is a reference implementation buyers can compare against vendor workflows. Ask for median and tail latency on DSAR processing — marketing ranges hide backlog risk.
Suppression lists should be versioned and auditable. If your licensed copy cannot prove which suppression version was applied to each weekly file, you cannot defend a campaign post-opt-out incident. Engineering should store suppression version IDs alongside each ingest partition.
Credible providers geofence out healthcare facilities, places of worship, domestic violence shelters, military installations, and schools by default. This is enforced in FTC X-Mode / Outlogic and Mobilewalla orders — not optional hygiene. Buyers should run vendor data against a sensitive POI test list and expect zero hits.
If a vendor cannot answer all three with artifacts, privacy-safe is marketing. For the regulatory map underneath, see 2026 state privacy landscape and FTC location enforcement. Keep test artifacts in the vendor file — consent diagrams age, and regulators ask what you knew at purchase time.
International programs need the same three controls with jurisdiction-specific evidence — GDPR lawful basis documentation, UK ICO expectations, and U.S. state laws are not interchangeable checkboxes. Map each geography to collection posture before commingling feeds in one warehouse.
POI catalogs are usually not personal data, but visit analytics join POI to device paths. Scope POI & Geofencing with polygon quality and refresh, then apply the same three controls to the mobility layer. Risk and fraud use cases still need counsel to map permitted use — privacy-safe mobility does not automatically make a score lawful for credit or employment decisions.
Evaluating a new partner? Start at privacy compliance and request diligence materials through contact with your use case and geography attached.
Procurement should reject aggregate-only answers when the use case requires device-level mobility. Aggregation does not retroactively fix non-consented collection upstream. Likewise, a vendor that honors GPC in one app but not across its publisher network fails propagation even if your contract looks strong. Document test procedures in the security appendix so re-attestation repeats the same sensitive-POI exercise annually.
Regulated industries should map privacy-safe mobility to sector rules explicitly — HIPAA for life sciences, FCRA boundaries for credit-adjacent scores, and state health-privacy laws where clinic proximity matters. Audience targeting teams need the same written exclusions analytics teams use, or segments will leak into channels legal never approved.
Boards and general counsel now ask for demonstrable controls, not adjectives. Build a diligence packet you can re-send annually: consent diagram, GPC workflow, sensitive POI test results, subprocessors, and deletion SLAs. When a vendor refreshes its privacy center, diff the changes and re-run the sensitive-location test — policies that tighten upstream may still leave your licensed copy non-compliant if propagation lags.
Finally, separate collection compliance from use compliance. Your permitted use must match what you actually build — fraud scores, audience segments, and market analytics carry different risk profiles. Risk and fraud teams still need counsel sign-off even when mobility is collected with strong controls. Re-run the three-question test after major vendor releases — privacy posture is versioned software, not a static PDF.
Teach stakeholders one sentence: privacy-safe is testable controls, not adjectives. That sentence ends committee debates and keeps RFPs focused on artifacts. Pair it with global mobility diligence whenever visit data enters the warehouse. Re-test after vendor SDK or app-SDK updates — collection posture changes more often than annual contracts renew.
Re-run sensitive-POI tests after vendor SDK or app-network changes — collection posture shifts more often than annual renewals.