Escalating concerns about privacy and security are driving more enterprises to ramp up their requirements for data residency. An enterprise customer can buy regional hosting for its SaaS application but still have employee sign-in records processed abroad. Login events might be sent to a security monitoring service hosted elsewhere, or pulled up by a support engineer working from another country.
Those concerns tend to surface during contract reviews, when the customer asks exactly what a data-residency commitment covers. Requests like these are becoming more common, with 81% of organizations reporting heightened demand for data localization, according to Cisco’s 2026 Data and Privacy Benchmark Study. Much of the discussion begins with application databases and backups. Authentication, however, leaves its own trail of personal data, and buyers need to know where those records end up.
The overlooked data behind B2B logins
Consider a wholesale ordering portal used by retailers and their suppliers. A purchasing employee clicks to sign in, authenticates through their employer’s identity provider, then returns to the portal. The portal receives the employee’s identity and company details so it can display the relevant orders.
Exchanges like this are a core job of B2B IAM in enterprise SaaS. Each login creates its own records: the employee’s email address, IP address, sign-in time, and whether authentication succeeded. The portal may also forward those events to a security monitoring service in another region. Employee data is then processed there, even though the transaction data stays in the customer’s chosen hosting region.
Data residency also matters when support tickets are processed offshore. Say an employee is locked out of their account and asks for help. The ticket may land with a support engineer abroad, and it often includes error logs containing the user’s email, IP address, and failed sign-in attempts. That data has now left its intended region.
Buyers need to verify flows like these rather than accept a general assurance about where data is stored.
When identity data crosses legal boundaries
A promise to keep data local can involve three different questions. Where is information stored or processed? Does a law require the data to stay on servers inside the country? What conditions justify disclosing or transmitting such data abroad? Those questions correspond to data residency, data localization, and international transfer restrictions, respectively. The terms aren’t interchangeable.
Jurisdictions draw their lines differently. The European Commission’s guidance on international transfers sets out mechanisms that include adequacy decisions and appropriate safeguards such as standard contractual clauses. The EU’s General Data Protection Regulation (GDPR) allows certain transfers outside the European Economic Area (EEA), but remote access matters as much as physical location. A database can sit inside the EEA, but if an organization overseas can retrieve personal data from it, that access may itself count as a transfer, depending on the arrangement.
A customer’s contract may also require EU-only processing, which sets a tighter boundary than the law does. A residency statement should therefore explain who can access the data, not just where the servers are located.
Meanwhile, China’s Personal Information Protection Law requires critical information infrastructure operators, and processors handling personal data above certain volume thresholds, to store that data locally. Rules issued in 2024 exempted some cross-border flows from the law’s transfer procedures. A vendor serving global markets may therefore need to determine which requirements apply to each customer and dataset.
The United States follows a more fragmented approach to data residency. For instance, California’s Consumer Privacy Act regulates how covered businesses and their service providers handle personal information. It does not generally require domestic storage, however. At the federal level, the Justice Department’s Data Security Program restricts transactions, including vendor and employment arrangements, that could give designated countries of concern access to bulk U.S. sensitive personal data or government-related data.
For buyers, the practical question is what identity data a vendor’s systems handle and who can reach it from abroad. A generic regional hosting option can’t answer that.
What buyers should verify in B2B IAM
For any vendor that handles your users’ sign-ins, start with one ordinary account. Ask them to walk through what happens when an employee is invited, signs in through the customer’s identity provider, and opens the application. Where is the account profile held? Which service records the IP address and login time? Are those records copied to a monitoring provider in another country? A simple data-flow diagram should answer these, including information the vendor sends to outside services.
This also extends beyond a single organization. A consultant, for instance, may have access to several customer organizations through the same application. When one engagement ends, removing the consultant from that organization’s identity system might not close an active session or revoke permissions stored separately by the SaaS provider.
The vendor should demonstrate how it separates each organization’s records and how quickly access ends for accounts created through automated (SCIM or just-in-time) or manual provisioning.
Support and recovery deserve the same check. The locked-out employee from the support example above is a good test case: ask where that ticket goes and who can open it. Regional hosting for transaction data may not cover authentication logs, support tools, or backup copies held in a disaster-recovery region elsewhere.
Establish residency as a condition of rollout
Every identity-data flow must either meet the agreed location and access restrictions or have a documented, approved exception. Give the vendor a chance to correct failing configurations, then test them again before sign-off.
If the provider can’t explain where its login and support records go, treat that as an open finding, not a pass. Residency that stops at the application database doesn’t cover the people signing in to it.

