Negotiating Salary When You're Hired: The First Anchor Decides Everything

There is a moment in every recruitment process when the conversation changes nature without warning. You were discussing the role; suddenly someone asks, "and on compensation, where do you sit?"

Most candidates treat this as an administrative question to be answered honestly and quickly. It is something else entirely: it is the moment when the number everything else will orbit gets placed on the table.

And "everything else" runs long. A starting salary isn't just the first year's pay: it is the base from which subsequent rises are calculated, almost always as a percentage. An initial gap does not close internally — it widens.

In the article that follows, I explore pay negotiation at hiring and the difficulty of naming a figure in front of a recruiter. If you recognise yourself in this theme, the tests on the platform help you take stock.

The anchoring effect, and why knowing about it isn't enough

Anchoring describes a simple mechanism: once a number is present in a discussion, it steers the estimates that follow, even when everyone knows it is arbitrary.

Tversky and Kahneman demonstrated it in a series of now-classic experiments, and it has since been found in professional negotiation settings. The important part for you isn't the protocols, it's the practical consequence: the first number spoken works for whoever spoke it, and it keeps working even after being disputed.

Knowing the effect exists does not protect you from it. That is what separates it from an ordinary correctable reasoning error: recruiters, who know the phenomenon well, remain subject to it. The only real protection is procedural — decide your number before you walk into the room, and don't let the conversation produce it for you.

"What are your expectations?" is not a request for information

The question looks like an enquiry. It rarely is one.

What happens: in most processes, the company already holds a budgeted range for the role. So the question is not about finding out what you cost: it is about finding out whether you'll come in under it, and about getting you to set the first anchor. What it changes: naming a low figure does not make you more employable. It only signals a low estimate of your own value, and it permanently closes the top of a range you will never be told about. Three usable answers, in order of preference:

Send the question back, once. "You'll have a budgeted range for this role — what is it?" This is legitimate, common, and it leaves the anchor with the other side. In a growing number of contexts, the range is published in the advert anyway.

If pressed, give a prepared range whose bottom is a figure you'd be happy to sign. Never a single number. Building that range is covered in a dedicated article.

Condition it explicitly. "On the scope as described, I sit between X and Y. If the scope moves, that frame moves with it." You set an anchor while keeping a door open.

What to avoid: "I'm flexible", "whatever you think is fair", "salary isn't my priority". These don't signal flexibility, they signal a lack of preparation — and they reliably produce an offer at the bottom of the range.

Your current salary is the wrong base

Many conversations drift towards "and what are you on today?"

That question has a mechanical effect: it turns your past into a ceiling. If your current role is underpaid — common when you haven't renegotiated in years — you carry that undervaluation into the next one.

The relevant reference point isn't what you earn, it's what the role is worth. A simple, non-aggressive reframe: "My current level reflects a different scope. For this role, I sit between X and Y."

Worth noting in passing: several jurisdictions have restricted or banned salary-history questions in recruitment, precisely because of how they reproduce existing gaps. What the rule is in your situation is a matter for a competent source, not an article.

What is negotiable beyond base pay

This is the most under-used part of recruitment, and often where margin remains when base pay is locked by a grid.

One line deserves emphasis: a written review date at six or twelve months, with criteria attached, is frequently easier to obtain than an immediate base increase — and over two years it is often worth more.

What your exchanges with the company have already fixed

This is where preparation stops being about nerve and becomes about reading.

A recruitment process leaves a substantial written trail: the advert, the recruiter's messages, interview summaries, follow-ups, and above all commitment phrasing. "The role will grow into a management position", "we plan a check-in at six months", "the range can move for the right profile." Each of those sentences is a margin left open, with a date and an author.

Those margins are far more usable than any rehearsed line, for three reasons: they come from the employer, they are dated, and they can be quoted without conflict.

That is exactly what ScanMyJob measures: the observable margins inside written exchanges — who proposed what, when, what was accepted, what was left open. The output is not a script but a dated record of facts, each line of which you check yourself. The gallery of examples shows four real cases with their record, which makes the reading concrete.

The difference with standard advice is sharp: instead of preparing a sentence to say, you arrive knowing what has already been said, and by whom.

Why this moment does not come back

Hiring is the only point where two conditions hold at once.

You are in a position of choice. The company has invested in a process, turned other candidates down, and has a dated need. Starting over costs it weeks. That asymmetry disappears the day you sign. You carry no internal constraint yet. Once in post, your salary sits inside a grid, a team, an internal consistency — all perfectly real arguments that will limit any later revaluation.

This is also why the conversation is better handled as problem-solving than as a contest. Pruitt and Rubin's dual concern model distinguishes postures by whether you care about your own outcome, the other party's, or both: the posture producing the best durable agreements is the one that stays firm on substance while actively looking for what suits the other side. In practice: never make the agreement difficult for reasons of form, and offer trade-offs that cost the company nothing.

Before signing: three checks

Everything promised verbally must appear in writing. An unwritten commitment leaves with the person who made it — and recruiters move on. Bonus criteria must be readable by a third party. The test: could an outsider, reading the clause, determine whether the objective was met? If not, it isn't an objective. The response deadline is negotiable, and it helps. Asking for a few days has never cost anyone a job with a serious employer, and it lets you reread with a cool head. What can still be done once the offer is in hand is covered in a separate article.

What this article does not do

It gives you no amount. It describes a mechanism and a method. The number is built from verifiable benchmarks specific to your market. It says nothing about applicable law. Mandatory contract content, probation, minimum rates, the legality of salary-history questions, withdrawal periods: these are legal matters and they vary by country. They go to an official employment information service, an employee representative, or a lawyer. It does not promise that margin exists. Some roles are genuinely constrained by a grid. When that happens, the negotiation moves to non-salary elements, the job title, and the review date — and the unit you discuss in, gross or net, is worth settling clearly before any agreement.