Cut the scope, never the price

There is one rule that survives every version of this argument. If you need to arrive at a lower number, take something out of the job rather than taking money off the price.

The reason is not principle. It is that a discount on identical work tells the buyer what the work is really worth, and they will remember the second number, not the first.

Why price cuts do not stay local

A rate card is a claim about value that you are asking a stranger to accept without evidence. The moment you accept less for the same deliverable, you have supplied the evidence against your own claim, and you supplied it yourself.

Three things follow, none of them immediately.

The buyer's reference price resets. Whatever they paid is now the price, and your card is a document they know you do not mean. The effect is well documented: Kahneman, Knetsch and Thaler (1986) found buyers judge a seller's fairness against a "reference transaction", the last terms they were offered, and treat moves away from it as the thing needing justification. The next job opens with them expecting the discounted figure and you having to explain a rise you did not intend to make.

Discounts travel. A private concession becomes a thing repeat buyers mention to each other, and this trade is more talkative in places you cannot see than earners assume.

And your own resolve degrades. A card you have already broken once is easier to break again, which is how a rate card becomes a fiction over about four months.

The purpose of the card was to make every individual pricing conversation already settled, which is the whole argument for having one. A card you negotiate against is doing none of that work.

Scope reduction, in practice

The alternative is a genuinely smaller job at a genuinely lower price, which leaves both numbers honest.

Four dimensions come off cleanly.

Length or duration. A shorter written assessment or a shorter recording is a recognisable product at a lower tier, and it is the easiest reduction to describe in a sentence.

Depth of structure. Scores on the defined axes with a brief note each, rather than a written justification for every axis. This removes the slowest part of the work, so the discount actually corresponds to less time.

Turnaround. Standard rather than priority, or a stated window rather than a stated day. Buyers who are price sensitive are frequently not time sensitive, and this is the cheapest thing you can give away.

Revisions. One round rather than two, or none rather than one. Say it explicitly, because an unnamed revision policy means the answer is always yes.

Notice that all four make the job faster, which is the test. A reduction that lowers the price without lowering your minutes is a price cut wearing a costume.

When a discount is actually correct

Three cases, and they share a feature: something other than the buyer's persistence has changed.

Repeat buyers. A second job from someone you have already worked with costs almost nothing to win, and that saved acquisition effort is real money you can share. This is a structural discount rather than a concession, and how bundles and repeat pricing are built is the longer version of it.

Volume committed up front. Four jobs agreed at once is a different transaction from four jobs that might happen. The discount is for the commitment, so it lapses if the commitment does.

Genuinely dead periods. An empty week is capacity that expires, and a filled slot at eighty percent beats an empty one at a hundred. This is the case most likely to be abused by an anxious earner, because a quiet fortnight feels like a dead period and usually is not - the distinction is the subject of how to read a quiet stretch, and getting it wrong is the most common self-inflicted pricing wound in the trade.

What is not on the list: a buyer who asked, a buyer who was nice about asking, a buyer who mentioned someone cheaper, and a buyer you would very much like to have.

Making a discount that ends

Any discount without a stated end becomes the price. This is not a tendency; it is a certainty, and the mechanism is that neither party ever raises the subject again.

Three things make one end.

Attach it to a reason that is visible and finite. A quiet January is a reason with a February in it; "you asked" is not.

State the end when you state the discount, in the same sentence, before it has been accepted. Afterwards is a renegotiation.

Put the standard price in writing alongside the discounted one, so the record shows a concession rather than a price. That single line is what makes the return to full price a resumption rather than a rise, and it is exactly the kind of thing worth keeping a record of at the time rather than reconstructing later.

The buyer you are actually pricing for

The unpleasant finding, once you have thirty jobs of records, is that the discount-seeking buyer is not a cheaper version of your normal buyer. They are a different buyer, and on average a more demanding one, for reasons the piece on who the cheapest tier attracts works through properly.

That is why scope reduction protects you twice. It keeps the card intact, and it puts the price-sensitive buyer into a defined smaller job with defined limits, rather than into your full job with an aggrieved sense of what they are owed.

Something worth remembering when a buyer pushes: the reason they are talking to a person at all is that an automated result was not what they wanted. What an automated assessment reliably does is the free option they have already declined, which is a stronger negotiating position than most earners realise they are standing in. Buyers also arrive with a fairly clear sense of what commissioning a person normally costs, and the buyer-side account of choosing and briefing one suggests they are comparing risk rather than hunting for the lowest number. If precision is the specific thing you sell, the discount to avoid is the one that removes it, since a stated method is what makes a figure worth anything and a cheaper tier without one is not a smaller product but a different, worse one.

Rate Cock publishes its judge tiers and what each includes on its judges page, which is a useful example of scope-differentiated pricing written out in public.