Data Protection and Candidate Records

General orientation, not legal advice; requirements differ substantially by jurisdiction.

Putting the safeguards in “Data Protection and Candidate Records” into practice also requires a documented operating routine. Teams can use this implementation guide to understand where assessment work consumes time and to keep a consistent audit trail, while treating activity data as process evidence rather than as a proxy for candidate ability; proportionality, access controls and human review still come first.

For an independent benchmark, compare the process with ICO employment data guidance; the useful test is whether the local method remains job-related, proportionate and explainable.

A hiring process collects more personal data than most organisations realise, about people with whom they have no ongoing relationship.

What you hold

Application data: contact details, history, sometimes demographic information.

Assessment responses, which for work samples may be substantial.

Scores and assessor notes, including opinions about the person.

Recorded interviews, where used.

References, which contain a third party's opinions.

And rejection reasons.

The basics that apply almost everywhere

Tell people what you collect and why, before you collect it.

Collect only what you need for the decision.

Keep it only as long as you need it, with a stated period.

Secure it.

And be able to respond when somebody asks what you hold.

The access request

Candidates can generally request the personal data you hold about them, including assessor notes and scores.

Which is the practical reason the recording note says to write evidence rather than opinion: you may have to show it.

Prepare for this rather than discovering it. The first request should not be the first time anybody looks for the records.

Retention

Long enough to handle a challenge, which in most systems is months.

Not indefinitely, and not "in case a role comes up" unless the candidate agreed to that specifically.

Talent pools require consent in many jurisdictions and are frequently maintained without it.

Set a period, apply it automatically if you can, and confirm deletion covers your supplier's systems too.

Suppliers

Assessment platforms, video interview tools and background checkers all process candidate data on your behalf.

You remain responsible for what they do.

Know where the data sits, who at the supplier can see it, what they retain, and what happens at the end of the contract.

These are contract questions and they are rarely asked during procurement.

Recorded interviews

A substantial data holding and a sensitive one.

Tell candidates before recording, say who will watch and for how long it is kept, and offer a live alternative.

And delete on schedule, which for video is more important than for text because the volume makes retention expensive and the content is personal.

Monitoring data

Demographic information collected for adverse impact analysis should be separated from the assessment record and not visible to assessors.

Collected voluntarily, with an explanation of why.

And actually used, because collecting it and never analysing it is the worst combination: the intrusion without the benefit.

What to check

Do you have a stated retention period, and is it applied?

Could you respond to an access request within the statutory period?

Is monitoring data separated from assessment data?

And do you know what your assessment supplier retains?

The point

Candidates can generally request the data you hold, including assessor notes.

Write them as though they will be read, because they may be.

Underlying all of this

Almost everything in this collection reduces to one discipline: write down what the job requires, assess that thing directly, record the evidence, and look at your own outcomes afterwards. None of it requires buying anything, and organisations that do those four things consistently outperform ones running longer processes built from instruments chosen before the requirements were known.

The recurring pattern

The recurring failure across every section here is the same: measuring what is convenient rather than what matters, then never checking whether it predicted anything. The check is an afternoon of work once a year, and it is the step that separates a process that improves from one that merely persists.