How Much to Ask For: Setting Your Range Before You Open Your Mouth

The most expensive moment in a pay negotiation lasts three seconds. It is the one where someone asks "what did you have in mind?" and you answer without having decided in advance.

What comes out then is almost never the right number. It is produced under pressure, in front of someone, by a person whose main concern is not to sound unreasonable — and it becomes the ceiling of the discussion immediately, because nobody offers more than what was asked.

Setting your range beforehand is less about ambition than method: not letting an important decision get made in a moment of discomfort.

In the article that follows, I explore how a pay request gets put into numbers, and the discomfort surrounding that number. If you recognise yourself in this theme, the tests on the platform help you take stock.

There isn't one number, there are two

This is the most common starting error: hunting for "the" salary matching your job title, as though a single value existed.

Market value is what someone else would pay for your skills today, elsewhere. It depends on the occupation, the level, the location, the sector, the company size. It is reasonably documentable. Value here is something else: what this role, in this organisation, with this actual scope, represents to this employer. It depends on what you concretely carry, what would fall over if you left, the cost of replacing you, and the gap between your job description and your real work.

The two serve different purposes. Market value sets the credibility of your range — it stops you asking for something inaudible. Value here sets your legitimacy — it explains why you, and why now.

A solid request rests on both. One built only on market data gets "that isn't our grid". One built only on personal effort gets "everyone works hard here".

Where to find verifiable benchmarks

The key word is verifiable. A number overheard in conversation does not survive ten seconds in front of an HR team.

Usable sources share one trait: they can be shown, and they are dated.

• Published job adverts for comparable roles, saved as you found them. This is the most concrete source, because it comes from employers themselves. In several countries, publishing a range in the advert has become standard or mandatory.
• Annual sector salary surveys from recruitment firms, noting the year and the geography.
• Public statistical data from national statistics offices — slower, but very robust.
• Internal pay grids where they exist, often accessible through employee representatives.
• Offers you have genuinely received, if you have any.

What is not a benchmark: a national all-sector average, a figure read on a forum, or a colleague's pay when you don't know their seniority or scope.

Value here is already written in your correspondence

This is the point most articles skip, and it is the one that makes the difference in the room.

You don't need to reconstruct from memory what your role has become: it is documented, in your work messages, without anyone having meant to. A scope expansion almost never gets a formal document. It arrives in an ordinary message: "can you pick that account up", "you're their contact now", "just cover it for the moment".

Three sequences can be read from it:

• What you were handed, and on what date — this is the raw material of value here.
• What was promised in return — "we'll look at that", "this will be recognised". These are margins left open, and they carry a date.
• What you already asked for, and how it was received — a refusal explained by the calendar is not handled like a refusal explained by your level.

That is exactly what ScanMyJob produces: a dated record of facts drawn from written exchanges, showing the observable margins rather than supplying a line to recite. The gallery of examples shows four real cases with their record — useful for seeing what this material actually looks like before hunting for your own.

The point is not to accumulate evidence as in a dispute, but to be able to say, without raising your voice: in March I ran A; since September I have run A, B and C, and here are the messages that establish it. That is what makes a number defensible.

Building the range: three points, only one of them spoken

A usable range has three values, of which only one part is announced.

The top must stay defensible. A high anchor works — more on that below — but an implausible one produces the opposite effect: it discredits the whole request. The test is simple: can you cite two sources supporting it? The bottom is where people go wrong most. Many place there the lowest figure they could bear without leaving. That is a mistake, because the bottom is exactly what will be offered. The bottom of the stated range must be a number you would be pleased to sign, not one you could tolerate.

The bottom of your range is not your walk-away

This confusion deserves its own section because it distorts the whole preparation.

What Fisher and Ury call the BATNA — the best alternative to a negotiated agreement — is not an amount. It is a scenario: staying at your current salary with a review date, accepting another offer, moving scope internally, or starting a search. It has a cost, a timescale and a risk level.

Three practical consequences:

• It is known in advance, not during. Someone who doesn't know what they'll do if refused negotiates with the fear of the void, and it shows.
• It is not announced. Waving another offer turns a negotiation into a test of strength and manufactures lasting resentment.
• It doesn't need to be good to be useful. Knowing "if it's no, I ask for a six-month review date" is enough to remove the panic from the moment you hear it.

The zone where agreement is possible — the ZOPA, the space between your real minimum and the employer's real maximum — is never known to both sides. The whole point of preparation is to avoid excluding yourself from it out of excessive caution.

Why state a range rather than a figure

Three reasons, all observable in how conversations actually unfold.

What happens with a single figure: it is heard as a maximum. The discussion can only go down, and every unit gained above it feels like a concession extracted by force. What happens with a range: it puts two numbers into the room. Tversky and Kahneman's work on anchoring shows that the first number stated durably shifts the other party's estimate, including among experienced professionals. A range sets a high anchor while signalling flexibility — which, in Pruitt and Rubin's dual concern model, corresponds to a problem-solving posture rather than a contending one: firm on substance, cooperative in form. What you accept in exchange: you will be offered the bottom. That is near-systematic, and it isn't a failure — it is exactly why the bottom must be a satisfying number.

A range of reasonable width — roughly 8 to 12 % between the two points — stays readable. Too wide, and it signals you haven't done the work.

What counts alongside the number

A range doesn't only cover base pay. Three elements often come in when the salary envelope is stuck:

• Variable pay: the amount, but above all the trigger criteria and who assesses them. A bonus whose criteria aren't written down isn't income, it's an intention.
• Things with a cash value: remote-working days, funded training, equipment, contributions to costs.
• Things with a career value: job title, reporting line, a formalised scope, a written review date.

That last one is often the most profitable when the envelope is closed: a review date in writing turns a refusal into a deadline. And the unit all of this is discussed in — gross or net — is not neutral: a dedicated article covers it.

What this article does not do

It does not give you your number. It gives the method for establishing one from verifiable benchmarks. No article can know your market, your real scope, and the state of your employer. It addresses no question of law. Applicable pay scales, minimum rates, pay-transparency obligations, the form a contractual amendment takes: these are legal questions, varying between countries and sometimes between sectors. They go to an employee representative, an official employment information service, or a lawyer. It does not guarantee the top of your range. Solid preparation clearly improves the starting position, and timing matters just as much. What it does guarantee is that you will not answer an improvised figure to the question that decides everything.