Publishing a salary band is presented as a generosity. It is not. It is an efficiency, and the studios I have done this with recovered the cost within a hiring round.
Before: an average of 2.4 rounds of salary negotiation per offer, an offer acceptance rate of 58 percent, and a median time-to-accept of eleven days. After: 0.3 rounds, 84 percent acceptance, four days. The people who were going to say no said it before the first interview, which is the entire saving.
The cost is real and I want to name it. Two senior candidates looked at the published band, worked out that they were being offered the middle of it, and negotiated to the top. Under the old arrangement we would have anchored them lower and they would have accepted. They were worth the top of the band. We were previously underpaying people who were bad at negotiating, and I am not going to pretend that was a business advantage worth defending.
The rubrics were the harder sell internally. Publishing what you assess means candidates prepare for it, and there is a persistent belief that a prepared candidate is a less informative signal. In practice a prepared candidate is a candidate who wanted the job enough to prepare, and the rubric still distinguishes between them. The unprepared-candidate signal we lost was mostly measuring who had a friend at a studio.
