RehearseaBlog
4 min read

The salary question you are told to negotiate, and the one you are not

Advice about job searches is almost entirely about the number. The thing that decides whether there is a number at all is rarely discussed.

  • job negotiation
  • salary
  • ghost jobs
  • job search

Nearly all the advice about job search is about the number. Negotiate it. Do not name a figure first. Have a range. Know your worth. Every one of those is a conversation about a salary, and they assume the salary is the subject of the process.

Two threads in the research behind this product are about something else entirely, and they are worth putting next to each other because they describe the same mechanism from two sides.

What the people inside the process say

Someone writing from the hiring side of a FAANG-adjacent company put a number in the first paragraph: every posting their team puts up gets roughly two to three thousand applicants within a day or two, and of those maybe three hundred get any consideration at all.

The post is careful about what it is claiming. The author rules out candidate behaviour first, and locates the problem on their own side of the table: recruiters and hiring managers "drilled into your head that you MUST have metrics, impacts and keywords or else your resume is trash", and candidates did what they had to do. The distortion being described is not applicants lying. It is a funnel compressing three thousand people into three hundred, and everyone downstream of it responding to the compression.

That is a system where the number is decided before anybody negotiates anything. Two thousand nine hundred and seventy people are eliminated by a posting and a filter, and the ones who remain are arguing about the last two digits.

The other thread, from the outside

Someone describing ghost jobs names the reasons a company posts a role with no intention of filling it, and gives the one that matters most: their investors believe they are hiring and doing well. The posting is a signal about the company's own reporting, not about a job.

The author's response to it is a rule, and the rule is more useful than the complaint: set a limit of three interviews, and treat anything beyond that as information.

What the two threads have in common

Both describe a process where the decisive fact is not observable by the candidate, and both describe candidates responding rationally to the parts that are.

  • The applicant cannot see that a posting exists to satisfy an investor. They can see that it exists.
  • The candidate cannot see that three thousand applications narrow to three hundred. They can see that they are competing with three thousand.
  • The candidate cannot see the internal candidate. They can see the fifth round.

In every case the rational response is to compete harder on the only axis that is visible, and that is exactly what the advice about metrics and keywords and negotiating hard is telling them to do. The advice is not stupid. It is a set of instructions for winning a competition whose outcome was largely settled before the entry fee was paid.

And that is why the salary advice feels so important. It is the one part of the process the candidate believes is still open, and in a process where most things are not, attention goes there.

What can actually be done

Not much about the funnel. These are the parts that survive contact with a real search:

The three-interview limit is the useful idea in all of this. It is a rule about information rather than about effort: past a certain depth, the process is telling you something, and continuing costs you time you could spend on a process that is actually hiring. It is a heuristic rather than a law, and a candidate with a strong pipeline may reasonably hold a different one. What matters is deciding in advance, because the decision is expensive to make at round five.

Ask whether the role is new or filled. A re-post of the same requisition, unchanged, is the cheapest available signal about whether anyone is being hired. It is also the question recruiters are least likely to answer directly, which is itself information.

Stop treating rejection as a price signal. If a role is posted for an investor's benefit, the salary attached to it was never a negotiation between two parties. That is not a reason to refuse work; it is a reason not to read a number as a verdict.

Keep the record, which is what this product is for. Applications, the stages actually reported, and the outcomes. A candidate with forty of them can see which of these patterns are personal and which are market-shaped, and nobody can see that from one rejection.

The honest limit

These are two forum posts, and one of them is from inside a company whose name it does not give. There is no rate here and no way to check the claim that ghost jobs are common, and this product has started collecting its own data rather than borrowing a figure it would have to invent.

What is defensible from two posts is narrower: the salary is the part of a job search a candidate can see, and it is not the part most likely to decide the outcome. The advice to negotiate hard is not wrong. It is answering a question the process has already mostly answered.

Related