Twenty applications is a morning; sixty is a different job
A panel can read a modest field with care. Past a certain size, fatigue replaces judgement — and the shortlist starts to reward whoever was opened first.
A hiring panel that agrees to “just knock through the applications on Tuesday” often has a number in mind that belongs to last year’s vacancy. Last year there were eighteen files. This year there are sixty-two. The room has not grown. The coffee has not improved. The criteria have usually grown longer.
Twenty written applications, each answering five or six criteria at two thirds of a page, is a morning of honest work for two readers who already know the position description. Sixty applications is not three mornings of the same work. Attention falls. Early files receive marginalia; late files receive a skim of the first criterion and a guess. The guess tends to favour familiar employers and tidy formatting.
Workforce application analytics, as we practise it, begins with an admission about volume. If the field will not fit in the hours the panel actually has, the panel should not pretend it read every statement. Either the field is sampled on a declared rule (which is rarely fair), or the reading is delegated to people who will mark to one frame and return a briefing the convenor can still explain.
Delegation is not abdication. The panel still owns the shortlist. What it cannot own is a fiction that every file received equal care after 9 p.m. When we split a large field into batches, we dual-read a sample at the start of each batch precisely because marker drift is the usual failure. The aim is not a machine ranking. The aim is that application forty-seven is judged by the same ruler as application three.
If your next advertisement is likely to draw a crowd, decide the reading plan before the closing date. Interview calendars that assume a Tuesday skim will punish both the panel and the candidates who wrote carefully at the back of the pile.