Applicant tracking: keeping candidates in one place without a platform

Applicant tracking is the unglamorous half of hiring: knowing who applied, which stage each person is at, who is waiting on whom, and what was decided about everybody you did not take. Most small companies run it across an inbox, two spreadsheets and somebody's memory, and the cost appears as candidates who never got an answer and decisions nobody can explain six months later.

What applicant tracking has to hold

• Every applicant, including the ones rejected at first glance — a record of only the shortlist is the one that cannot answer a complaint.
• The stage each person is at, and the date they reached it, which is what makes stalling visible.
• Who is responsible for the next action, because candidates are lost in handovers rather than in decisions.
• The decision and its reason, in a sentence, recorded at the time rather than reconstructed.
• The source — where the application came from — which is the only way to know which channel is worth paying for.
• Consent and retention: how long you will keep the application, and whether they agreed to be considered for future roles.

The parts people forget

Two, and both are legal rather than operational. The first is retention: unsuccessful applications are personal data held with no ongoing purpose, and keeping them indefinitely because deleting feels wasteful is the commonest quiet breach in small-company HR. Decide a period, state it in the acknowledgement, and apply it. The second is consistency of record — if one candidate is rejected with a recorded reason and another with nothing, the pattern across a year is what an employment tribunal reads, and absence of a reason is read unfavourably.

Reply to everybody, including rejections, and make the system make that easy. The commonest complaint about small-company hiring is not being turned down but never hearing anything, and that failure is almost always a tracking failure rather than a decision somebody made.

When a spreadsheet is enough

1. Under about twenty applicants a role, a table with stage, date, owner and decision genuinely works.
2. Keep one row per applicant per role rather than per applicant, since the same person may apply twice for different jobs.
3. Use a fixed set of stage values instead of typed prose, or the pipeline cannot be counted.
4. Put the retention date in a column and filter on it, so deletion happens by routine rather than by decision.
5. Move to a dedicated tool when several people need to act simultaneously, not when the volume grows — concurrency is what breaks a shared sheet.

Measuring the pipeline honestly

Two numbers matter and neither is the number of applicants. The first is time from application to first response, which is what candidates experience and what loses you people to faster employers. The second is where applicants drop out — if two thirds vanish between first interview and second, the problem is the process rather than the market. Counting applications makes a recruitment report look busy; counting those two things tells you what to change.

Ettex Records holds the applicant list with stage, owner, decision and retention date as fields, so the overdue-response list and the deletion list are both filters. The stages themselves belong to the recruitment process, which should be settled before any tracking is set up, since tracking an undefined process produces a record of improvisation.