A keyword search reads the job title somebody typed into your ATS years ago. It cannot read the CV underneath it, and it cannot tell you whose ticket has expired. Run one here, over a sample database, then read the same records properly and see the difference.
01 · Your database
Pick the size of your database. Everything below is arithmetic you can check, not a projection.
Where the 80–90% comes from: Will Grady’s own figures from running an energy desk, and confirmed against two agencies’ real database sizes. It is not a published benchmark, and we do not present it as one. 1.42p a CV is measured from our own billing, not estimated.
02 · The search you run today
This is a sample pool of 24,800 records — Commissioning & start-up, UKCS — records touched in the last 3 years. Search it the way a consultant would.
Press search. It behaves exactly like the search in your ATS: it matches the words in the job-title field, and nothing else.
That is the whole result. No order, no evidence, and no way to tell from the list which of them could mobilise next month. So look at it two ways.
03 · The same records, read
A pool brief, not a client vacancy. No end client is named and no decision is made about anyone: the output is a ranked pool your consultants work from.
The pool brief
What a score is not. It is a reading of one CV against one brief, stamped with the date of that CV. It does not decide anything about anybody, and no candidate is rejected by it — your consultants work the list. Scoring is name-blind and age-blind, and the fairness testing behind that is published.
04 · What comes back
Nothing to install, nothing for IT to build, and no connection to your ATS is required for this — you export the records, we read them and hand them back.
Every row carries the record number it came from, so a consultant bulk-tags the pool inside your system in one go. No re-keying, no new database.
Each line of the brief marked evidenced, partly or not on the CV, with the sentence from the CV that shows it. Your consultant can check our reading in seconds.
What the pool holds, where it is thin, and which tickets have lapsed across the whole discipline — the re-engagement list writes itself.
We scan and purge. The records come back to you and do not stay with us, so the data-retention conversation is a short one.
On the lawful basis, because it is the first thing a good recruiter asks. The purpose does not change — they gave you their CV so you would find them work, and this finds them work. No decision is made about anyone, you remain the controller throughout, and a pool brief never names an end client. Every score is date-stamped to the CV it read.