Self-assessment
Score yourself
Answer each question honestly. Yes = 2, Partly = 1, No = 0. Keep a running total.
Your score
0/ 24
0 of 12 answered
Not scored yet
The intake
0 / 601 Can a person raise a request on every channel they already use: website, app, call centre, branch, email?
02 Is every incoming request logged and acknowledged automatically?
03 Does each channel have a named owner, so no request lands in a mailbox nobody reads?
The answer
0 / 604 Is identity verification proportionate: enough to stop one person reading another's data, without hoarding identity documents?
05 Can you find one person's data across every system and partner, in days not weeks?
06 Does every request end in a reasoned decision: granted, partially granted, or refused with recorded reasons?
The clock
0 / 607 Is your grievance channel published and easy to find?
08 Is your published response period one you actually meet, with room to spare?
09 Is there an escalation path for requests and grievances that miss their clock?
The proof
0 / 610 When erasure is granted, does it reach archives, backups and processors, not just the app?
11 Do request and grievance volumes and ageing reach the DPO office as a standing report?
12 Have you dry-run each request type end to end, including at least one reasoned refusal?
| Score | Where you stand | What to do next |
|---|---|---|
| 0 to 8 | Exposed | A request today would likely become a grievance, and a grievance a Board case. Fix intake ownership first. |
| 9 to 16 | Halfway | The front door works but the back office leaks. Retrieval across systems and erasure that reaches partners are the usual holes. |
| 17 to 21 | Strong base | Prove it with the dry run at the end of this page, then tighten the clocks. |
| 22 to 24 | Leading | Watch the edges: new channels and new systems joining the estate without joining the workflow. |
Any single "No" is a named gap, whatever the total says. Nothing leaves your browser.
What the rights are
A data principal can ask for a summary of their personal data and how it is being processed, ask for correction and erasure, nominate someone to act for them, and raise a grievance. On grievances the discipline is specific: you must publish how a person raises one, and answer within the period you publish. The published period binds you, and the Board is the next stop for an unanswered grievance.
The framework: four moves
Move 1: Build intake on every channel. Website, app, call centre, branch, email. Log and acknowledge automatically. Every channel has a named owner: a request that arrives where nobody owns it is a missed deadline waiting to happen.
Move 2: Verify proportionately. Check enough identity to stop one person reading another's data. The check should not collect more data than the request itself involves.
Move 3: Retrieve, decide, respond. The request executes against the systems where the data actually lives; this is where the discovery and mapping work pays off. Every decision is reasoned: granted, partially granted, or refused with reasons recorded. A reasoned refusal is a defensible outcome; silence is not. Respond in plain language, inside your stated timeline.
Move 4: Close with evidence, and layer the grievance SLA. Record what was requested, verified, found, decided and done, and when. On top, the grievance machinery: one visible published channel, automatic acknowledgement, escalation for grievances that miss their clock, closure tracking with reasons, and volume and ageing reports to the DPO office. Erasure deserves special honesty: "deleted" in the app while copies persist in the warehouse and at processors is not erasure. The retention and erasure page covers how to reach archives, backups and partners.
Template: the request and grievance log
One row per request or grievance. Copy this and fill it in.
| ID | Type | Channel | Received | Verified | Decision and reasons | Responded | Closed | Days taken |
|---|---|---|---|---|---|---|---|---|
| Example: R-104 | Access summary | App | Granted in full | 6 | ||||
| Example: G-21 | Grievance, marketing after withdrawal | Upheld; suppression list fixed | 3 | |||||
The downloadable file adds the full field list, including the evidence each closure must carry.
What good looks like at the end
You should have: live workflows for each request type with owners and clocks, verification standards per channel, a published grievance channel with its response period, and closure records that prove handling.
The live test: submit one request of each type through your own channels, as a customer would. Track how long each takes, how many people had to be asked, and whether the answer was complete. Then try a withdrawal and check whether processing actually stopped everywhere.
Where enterprises go wrong
- A request form that feeds a mailbox nobody owns.
- Identity checks that demand more data than the request itself involves.
- Erasure answered in the app while copies persist in the warehouse and at processors.
- A published response period chosen for marketing, then missed.
- No evidence of closure, so a handled request looks exactly like an ignored one.
The decisions to make
- Which request types are automated versus manual, and to what depth.
- Verification strength per channel.
- Whether the grievance system is bought, built, or run on existing ticketing.
- How request and grievance SLAs align with the published response period.
Next step
The record of processing and impact assessments, the register these workflows read from. If you need help with any of this, write to us and we will help.