Notes

What the hiring side actually sees.

Reported notes on how applications are really read, written from the side of the table most advice never reaches. Every figure is sourced and linked. Nothing is promised that cannot be kept.

SourcedIf we cannot verify a number, we say so and leave it out. Every claim carries a link you can open.
Useful firstA reader who never becomes a client should still finish better informed than they started.
No guaranteesNobody can promise you the offer. We would rather be trusted than believed.
Latest The machine layer · July 2026 · 12 min read

What happens after you press send

Your application is read several times before a person sees it. We mapped the route from the vendors' own manuals, a federal court file and two academic audits: where a machine actually makes the call, how rarely anyone is told, and the short list of things a candidate can still control.

Read the note
From the piece
Applications per open role, 2022 to 2025
116189223244
Greenhouse benchmark data, March 2026: 6,000+ organisations, 640 million applications.
Earlier
The AI screen6 min read

You already tried AI on your resume. Here's why it still isn't landing interviews.

A chatbot can only polish the words you already typed. Your resume was never suffering from bad verbs, it was suffering from missing substance. What changed, why smoother now works against you, and the questions to ask yourself before you let any tool near the page.

Read the note

A note a week, each on something we have actually seen decide an application. If you want to argue with one of them, or you think we got a number wrong, write to us and we will correct it in public.