A new role opens and the team starts sourcing from scratch. Meanwhile, the ATS contains people who interviewed for a similar job, withdrew because the timing was wrong, or reached an offer before another candidate was hired. Those records deserve a review. They do not deserve an automatic email blast.
ATS database reactivation is a controlled way to revisit that history. The work is finding relevant people, checking whether the data can still be used, and opening a useful conversation. A search result is a starting point. It is not proof that someone wants a job, still has the same skills, or expects to hear from your agency.
Start with the new role, not the oldest list
Write a short role brief before running a search. Include the work the person will do, required experience, location constraints, compensation information you can share, and the recruiter who owns the opening. Separate requirements from preferences. An old title is a weak substitute for evidence about the work.
Next, choose the historical groups worth reviewing. Consider people who reached relevant interview stages, withdrew over timing, or were considered for a role that was cancelled. Read the previous rejection or withdrawal reason. A person who missed a requirement then may have gained it since, but that needs confirmation rather than a guess.
Know what your ATS search actually covers
Greenhouse has different rediscovery workflows. Its older Talent Rediscovery workflow can surface prospects and past applicants. It supports filters such as milestone reached, stage, rejection reason and overall scorecard rating. It also supports searching resumes and notes with required and preferred keywords.
The Past candidates tab in Talent Matching is different. It surfaces applicants only and uses filters rather than assigning AI match scores to past candidates. Users see only past applications they have permission to view. The tab relies on structured historical application data and does not re-parse past resumes. Check the workflow available in your account before deciding that an empty result means the database has nobody suitable.
Use broad searches alongside narrow ones. Review why a known relevant person did or did not appear. Save the filters and the role brief together so a recruiter can explain the shortlist. Do not quietly change the search while continuing to describe the results as the same pool.
Build a review record before sending anything
| Record detail | What to check | What to do next |
|---|---|---|
| Past application | Role, stage, date and outcome reason | Record why this opening is different |
| Skills evidence | Resume text and relevant interview notes | Keep the source date visible |
| Current situation | Availability, location and interest | Ask the person to confirm |
| Contact status | Restrictions, preferences and recent outreach | Suppress or route for review |
| Data use | Purpose, retention review and access | Resolve gaps before activation |
| Ownership | Active processes and responsible recruiter | Assign one conversation owner |
Use separate values for unknown, checked and confirmed. An empty availability field must not become available. A historic salary note must not become the person's current expectation. Keep the original note and add a dated update rather than overwriting the evidence that explained the first search.
Check retention and permitted use first
For teams subject to the UK GDPR, the ICO says personal data must not be kept longer than necessary. It does not set one specific time limit for every type of data. Your team needs to justify retention based on its purposes and review whether information is still needed. Delete or anonymise records when there is no longer a reason to keep them.
The ICO also says a new use must be compatible with the original purpose and have a lawful basis. Reusing records collected under consent has more restrictive compatibility rules. Treat a future-role search as a data-use question, not just a campaign setting. Ask the person responsible for privacy to approve the policy and how it applies to your records. This workflow is operational guidance, not legal advice.
Run reactivation as a small, owned workflow
- Save the role brief and search filters.
- Review historical evidence and select a manageable candidate group.
- Check identity, duplicate records, active applications and recent contact.
- Apply retention and contact restrictions before message preparation.
- Have the recruiter review why each person belongs in the group.
- Send a relevant invitation to reconnect and capture the response.
- Update the existing record with confirmed information and the next action.
The first message should explain the actual role and why you thought it might fit. Refer to prior contact only when the record supports it. Ask whether the person wants to hear more. Give them a clear way to decline future contact. Do not imply that an old interview means they have already passed the current process.
Route replies into distinct outcomes: interested, not now, no longer relevant, wrong contact, and do not contact. An absence of response is not interest. Decide follow-up rules before launch and stop them when a reply or restriction arrives. Make the owner visible so another recruiter does not restart the same conversation.
Give AI a narrow job
An agent can prepare a rediscovery list, summarize relevant past evidence and flag missing information. Require each summary to retain the application or note it came from and the date of that evidence. Ask it to say what is unknown. Do not let fluent wording hide an old source.
Keep the model away from unrestricted exports and unrelated private notes. Use the access scope approved for this workflow. A recruiter should review ambiguous matches and the proposed reason for contact. The agent can draft a factual opening, but it should not invent familiarity, current interests or a conversation that never happened.
Mistakes to avoid in database reactivation
- Treating every rejected candidate as unsuitable for every future role.
- Searching only titles and missing the reason a person reached interviews.
- Assuming a candidate's old location, pay expectation or availability is current.
- Creating a new spreadsheet that drops contact restrictions and reply history.
- Contacting the same person from several recruiters or client searches.
- Keeping records forever because an AI search might use them someday.
Review the first group before expanding. Count records reviewed, candidates approved for contact, replies, confirmed interest and recruiter corrections. Keep each denominator clear. More search results are not a hiring outcome. The useful result is a relevant conversation with current information and an accountable next step.
How Remqa builds ATS database reactivation workflows
Remqa builds AI agents for recruiting teams around the records and review rules your team already uses. A rediscovery workflow can connect search, evidence review, contact checks and ATS updates without making the candidate record a separate side project. Start with one role and a clearly owned pool.
If stale fields or disconnected tools are the main obstacle, review the recruiting stack integration guide. For a wider view of where recruiter work goes, see recruiting hours. Bring an example role and a few permitted records to the discussion so the workflow starts with real decisions.