ScanSoft UK All articles
Compliance & Capture

A Month Is Enough: Designing Capture So Subject Access Requests Never Feel Like an Emergency

ScanSoft UK
A Month Is Enough: Designing Capture So Subject Access Requests Never Feel Like an Emergency

Photo: records manager searching digital archive index on computer in office, via storables.com

Under UK GDPR, a subject access request starts a clock. The organisation has one month from receipt to respond, extendable by up to two further months where the request is genuinely complex, and the Information Commissioner's Office has consistently taken the view that a request is not complex simply because the records are difficult to find.

That last point is where capture design earns its keep. In most organisations the month is not consumed by legal analysis or redaction. It is consumed by searching. Someone has to establish which systems hold personal data about a named individual, which shared drives were used by a team that disbanded in 2022, and which of eleven boxes in an off-site store corresponds to the reference on a spreadsheet that has not been opened since the last move.

None of that is a legal problem. It is a records problem, and it is designed in — or designed out — long before the request arrives.

Where the month actually goes

The anatomy of a slow response is predictable. A request arrives by email to a shared inbox and sits for two days before anyone recognises it as a statutory request rather than a customer complaint. Identity verification takes a week of gentle chasing. The search itself takes another week, because it is conducted by asking six people whether they have anything. Then the material is reviewed, third-party data is redacted, and the response is assembled in the final three days under visible pressure.

Capture decisions can shorten four of those five stages. They cannot help with identity verification, which is a process question, but they can transform the search, the review and the assembly.

Four capture decisions that buy back time

Searchable text, not just images. A scanned PDF with no text layer is a photograph of a document. An OCR layer is what allows a name to be found across ten thousand files in seconds. The quality bar does not need to be exceptional; it needs to be consistent, and it needs to be verified on a sample rather than assumed.

A naming convention that survives staff turnover. If file names are generated by whoever happened to be at the scanner, the archive becomes a record of individual habits. A short, enforced convention — matter or case reference, document type, date in ISO format — is unglamorous and disproportionately effective.

A records index rather than a records pile. The index is the single document that says what exists, where it is, how long it is kept and who owns it. It does not need to be elaborate. A spreadsheet with five columns, maintained as a condition of the capture workflow rather than as a separate chore, will outperform an ambitious system that nobody updates.

A known handback route for off-site boxes. If records are held by a storage contractor, the organisation should know, without asking, the turnaround time for a retrieval, the reference format required, and whether the contractor can deliver a digital copy directly. A retrieval that takes five working days is entirely acceptable — provided it is discovered in week one rather than week three.

The index that answers a question you did not anticipate

A subject access request rarely names the system. It names a person. The useful index is therefore indexed by person as well as by file, which means that HR files, grievance records, occupational health notes, email exports and correspondence are all reachable from one starting point.

The same discipline helps with one of the more common complications: material relating to third parties. Where records have been captured with consistent structure, the review stage becomes a matter of opening a defined set and applying a redaction routine. Where they have not, the reviewer is reduced to reading whatever turns up and hoping it is complete.

It is also worth remembering that a reasonable and proportionate search is expected to extend to backups and archived material where it is accessible. The organisation does not have to be able to produce everything instantly, but it should be able to explain, credibly, where it looked and why.

The twice-yearly rehearsal drill

The only way to know whether any of this works is to test it. Twice a year, run a drill with the following shape:

  1. Select a name at random from a closed or historic case list — never a live individual, and never a real requester.
  2. Start a timer. Treat the request as received at 09:00.
  3. Record the time taken to identify every location holding material relating to that name. Note every point at which someone has to ask a colleague a question, because each of those is a design failure.
  4. Record the time taken to retrieve off-site material or to request it from the contractor.
  5. Complete a review and redaction sample of twenty pages and record the time.
  6. Produce a one-page summary: total elapsed time, the two slowest steps, and one change to make before the next drill.

The output is not a compliance score. It is a list of specific frictions — a folder with no owner, a naming convention that was abandoned in one department, a contractor reference format that nobody wrote down.

What readiness looks like

An organisation that has done this work twice will recognise a mature position fairly easily. A request arrives and is logged within a day. The search is run as a query rather than a conversation. Off-site material is ordered against a reference that already exists. Redaction is a defined step with a second pair of eyes. The response goes out in week three with time to spare, which is precisely when the awkward cases — the ones with a third-party complication or a genuine extension — can be handled properly.

The month, in other words, is not the problem. It is only tight when the records were never designed to be found.

This article is general guidance on records and capture practice and does not constitute legal advice. Organisations should take advice on their own circumstances, and refer to current ICO guidance on subject access requests.

All Articles

Related Articles

Evidence With a Long Memory: Building Receipt Records That Still Answer Months Later

Evidence With a Long Memory: Building Receipt Records That Still Answer Months Later

The Final Frame: A Controlled Retirement Route for UK Microform Holdings

The Final Frame: A Controlled Retirement Route for UK Microform Holdings

From Dots to Decisions: Defining a Fit-for-Purpose Image Standard for UK Capture

From Dots to Decisions: Defining a Fit-for-Purpose Image Standard for UK Capture