Philippines staffing guide · Published
Write an identity verification script for an outsourced privacy request desk
A privacy request starts a legal clock, and the identity check decides whether the desk can act on it. A written script keeps the request visible while the verification stays proportionate to the data involved.

The short answer
Separate intake from identity verification, use a method the client approved for each request type, and record failed checks as carefully as successful ones. The desk can prepare the decision, but the client's privacy owner should decide whether an unverified requester receives any data.
Start with the request, not the requester
A privacy request can arrive by email, web form, phone call, or a message to a support agent who does not recognize it. The intake step should capture the request itself before asking who sent it.
Write down what the person wants, which data or account it concerns, and the date it arrived. A request that is logged late is still a request that arrived on time.
Do not let the verification step erase the intake record. If identity checks fail, the desk still needs the original request, the contact details used, and the reason the check stopped.
A closed ticket with no trace of the request is the worst outcome for the client. Keep intake and verification as separate fields so one can fail without destroying the other.
Choose verification that fits the risk
The right check depends on what the request can reveal or change. A copy of a utility bill may be enough to confirm an address, while access to a full account history may need a callback to the number already on file.
Ask the client to approve the method for each request type in advance. Improvising verification under deadline pressure is how desks collect too much or too little evidence.
Avoid treating a matching name as proof of identity. Names repeat, email addresses are shared, and a caller can know the last four digits of an account without being its owner.
Use a factor the requester is unlikely to obtain from a breach, such as a live callback or a verified portal login. Write the accepted methods into the script so the decision does not vary by coordinator.
Write the script in plain language
The script should explain why the desk is asking for information, not just what it needs. A short sentence about protecting the account reduces suspicion and gives the requester a reason to cooperate.
Avoid internal jargon such as verification workflow or data subject rights intake. The person on the other end may be stressed, and plain language is a control, not a courtesy.
Keep each question narrow enough to answer without guesswork. Ask for one document or one confirmation at a time, and say what the requester should do if they cannot provide it. Give an alternative path instead of ending the call.
A script that only works for people with perfect records will fail the people who need help most. Read the script aloud with a colleague and remove any phrase that sounds like an accusation.
Record failed and partial checks
A failed verification is useful evidence, not an embarrassment to hide. Record which method was tried, what matched, and what did not. Note whether the requester was told about the alternative path and whether they chose to use it.
These details help the client decide whether to retry, escalate, or close the request. A short note about what the requester was told prevents a repeat call with no context.
Partial matches deserve their own status. A requester who confirms an address but not a date of birth is neither verified nor a fraud case. Give the desk a middle category with a clear next step and an owner.
Binary pass or fail forces coordinators to guess, and guessing is where inconsistent outcomes begin. Name the owner of that middle category so it does not sit between teams.
Route suspicious requests to a human
Some requests should stop at the desk and go to a person with authority. A requester who pressures the agent, supplies inconsistent details, or asks for another person's records needs a supervised decision.
Write the triggers in the script and name the escalation contact for each shift. The desk should not accuse anyone, and it should preserve the details for the handoff.
The handoff should include the request, the contact history, the verification results, and a short note about the concern. Do not editorialize beyond what was observed. The client's privacy or legal owner can decide whether to investigate, ask for more evidence, or close the request.
A calm, complete handoff is faster than a dramatic one. Include the date of the last contact so the client can judge the response window.
Review the script against real cases
Once a month, pull a sample of completed and abandoned requests and read the notes as a reviewer would. Look for missing intake fields, unexplained verification choices, and handoffs with no response. Ask the coordinators which questions caused the most confusion.
The script should change when the cases show a pattern, not when someone has a preference. Bring two examples to the review and ask whether the script would produce the same answer today.
Track time to acknowledge and time to decide separately, because a fast acknowledgment can hide a slow decision. Report both to the client with the number of requests waiting on identity evidence. Set a review date for the next script update and record what changed.
A script that is never revised is a script nobody is learning from. Report the revision to the client so the change is not mistaken for a new procedure.
The review should also test the fallback path. Pick a request where the first verification method failed and read how the desk handled the alternative. If the note says the requester could not be verified without saying what was offered, the fallback exists only on paper.
Write the missing step into the script and coach the next case with it. A fallback that nobody has used is still an assumption.
Questions to copy for the sales call
- "What exactly is the requester asking for, and when did the request arrive?"
- "Which verification method has the client approved for this request type?"
- "What happens if the requester cannot complete the preferred method?"
- "Who owns the decision when a request looks suspicious or inconsistent?"
Sources
- FTC: Privacy and security: Business guidance on handling personal information and responding to privacy risk.
- NIST Privacy Framework: A voluntary framework for identifying and managing privacy risk across data processing.
- ISO/IEC 27701: Requirements and guidance for privacy information management systems.
Common questions
Can the desk skip verification for a small request?
Only when the client's policy allows a lower-risk method for that request type. The decision should be written down and reviewed, not made because the requester seems friendly.
Should the desk collect identity documents it does not need?
No. Collect only what the approved method requires, and delete or restrict the copies under the client's retention rule. Extra documents create risk without improving the decision.
What if the requester is someone else acting for the data subject?
Follow the client's authority rules and verify both the requester and the permission to act. Record the evidence and route the approval to the client's privacy owner.
Does a failed check mean the request is fraudulent?
Not by itself. Record the partial result, offer the alternative path, and let the client decide whether to investigate or close the request.