How to Raise Your Prices Without Losing the Accounts You Actually Want to Keep

The advice everyone gives about raising prices is “give sixty days’ notice and frame it around value delivered”. That advice is not wrong, and it is also not what changes whether the price increase holds. What decides that is the order — which accounts you announce the increase to first, which you announce to last, and how long you wait between each set.

Most agencies and freelancers send the notice to everyone on the same day. That is exactly how you lose the accounts you most want to keep and end up with the accounts most likely to complain, at a fee that is barely different from where you started. This post is about the sequence that avoids that outcome.

Why one-day-to-everyone fails

When you send a price-increase notice to your whole client list on the same day, three things happen inside a week.

One: the most difficult account emails back within an hour. They are the fastest responders because a price increase is exactly the shape of thing they treat as a negotiation, and they know that the sooner they push back the more likely you are to grandfather them at the old rate. Their objection lands before you have any evidence about how the rest of the list will react.

Two: the calmest, best-behaved account replies within about a week with something like “understood, no problem, appreciate the notice”. They have already priced the increase in and moved on. Neither their behaviour nor their fee changes as a result.

Three: three accounts in the middle send back a variant of “let us discuss on the next call” and quietly start looking at alternatives. You will not hear from them for a month, and you will not know why they leave when they do.

The net effect is that the difficult account got exactly what they wanted (grandfathered rate), the good account got the increase you would have got anyway, and the middle accounts left. You have shed revenue from the accounts you most wanted to keep, kept the account you should have raised the fee on the most, and told yourself the price increase “held” on the good clients — which was never the question. The specific psychology at work is well-documented — Tversky and Kahneman’s reference-dependent model of riskless choice formalises why a price change is evaluated against the current fee as a reference point rather than against the fair market rate, and why the loss (paying more) feels roughly twice as heavy as an equivalent gain would feel positive; the tier that resists is the tier that has anchored hardest on the old number.

The reason the same-day announcement produces this is that it collapses information you should be gathering sequentially into a single event you cannot learn from. You do not know how the second announcement should be worded until you have seen how the first announcement landed, and you cannot see the first landing if you have already sent all fifteen at once.

The order that works

The rule that produces a different outcome is: announce first to the accounts whose reaction you most want to learn from, and last to the accounts most likely to negotiate down the fee even if they secretly accept it.

That translates into a specific four-tier sequence. Every account you have goes into exactly one of these tiers, and the sequencing is what does the work.

Tier 1 — the calmest, best-fit accounts. Announce first.

These are the accounts you would happily keep at ANY reasonable fee. Clean scope, low drama, useful feedback, on-time payment. Announce to them first, alone, in a first small batch.

The reason to announce here first is not that they matter most. It is that their reaction is the most honest signal you will get. If two of your three best-fit accounts push back on the new fee, you have priced too high and you should re-word the announcement for tiers 2-4. If all three accept without comment, you have priced correctly for the segment you most want to keep. Either way you learn something before the announcement lands in front of the accounts that would have complained regardless.

When to send: roughly 75 days before the effective date. This is deliberately earlier than the standard “60 days’ notice” guidance — you need the extra 15 days as feedback window before the tier-2 batch goes out.

Tier 2 — the account you would raise the fee on the most, if you could pick one.

There is usually a specific account that is running you closer to break-even than the P&L admits, whose behaviour is the reason you are raising prices at all, and whose retention at the new fee is the entire point of the exercise. That account is tier 2 — one account, alone, with the new fee — and you announce to it after the tier-1 signal has come back and you know your pricing is defensible.

The reason to isolate this account is that its reaction cannot be judged in the context of a larger batch. If you send it alongside twelve others and it pushes back, you cannot tell whether it pushed back because the fee is genuinely wrong for it or because it always pushes back on everything. Sent alone, its response is a clean data point.

When to send: 60 days before the effective date. This account gets the standard notice window; the extra 15 days from tier 1 has been spent calibrating.

The additional preparation: for this one account only, write down before the conversation what “outcome I would accept” is. Is it: the new fee, no concessions? The new fee minus 10% for the next quarter, then the full increase? The new fee with a small deliverable removed? Something else? Decide before the meeting, because in the meeting the account will offer something and you will accept it in real time whether or not it is what you actually want.

Tier 3 — the middle. The steady accounts, in a batch.

These are the accounts that are neither best-fit nor problem — they are steady, professional, transactional. They pay on time, they use what you deliver, they do not create work for you outside scope. Announce to them in a batch, all on the same day, with a stock template refined by the two signals you now have from tiers 1 and 2.

Why the batch is fine here: these accounts do not learn what you sent to whom. They compare their letter to their previous invoices, not to each other’s letters. And their response distribution is boring — a mix of “acknowledged”, “let us discuss on the next call”, and one or two “any flexibility on this?” — which the batch format handles well.

When to send: 45-50 days before the effective date. Slightly compressed window because these accounts move faster; and because you have already learned everything you were going to learn from the earlier batches, there is no reason to wait.

Tier 4 — the difficult accounts. Announce last, and prepare for the exit.

The accounts that have shown the four signals in when to fire a client — scope drift that is defended, an accreting approval chain, briefs that change after delivery, and the pattern of your team quietly avoiding them — are tier 4. Announce to them last, alone or in a pair, with a fee that reflects the actual cost of delivering them at their preferred delivery model.

The point is to give them the option to stay at the real fee, OR to leave. Both are acceptable outcomes. The specific mistake to avoid is discounting the tier-4 fee just because the account is large. The reason it is large is often that it has been priced against how much it looks like it is paying you, not against how much it is actually costing you to deliver — and the price increase is the correction of that gap, not an additional ask.

When to send: 30-45 days before the effective date. The compressed window is deliberate: it removes some of the runway they would otherwise use to negotiate. If they need more time, they will ask, and you will decide whether to grant it based on how the earlier tiers went.

Prepare for the exit before the announcement. Assume one or two of the tier-4 accounts will leave rather than accept the fee. Write down before the announcement how you would fill the revenue gap — retain-and-grow on tier-1 accounts, add a new one from the pipeline, or accept a temporarily lighter month. If you have not planned for the exit, you will make concessions in the meeting that will read to the client as a signal that the fee is negotiable and to your team as a signal that difficult accounts get discounts.

The one-page announcement, actually written down

The wording is the smaller half of the exercise, and half the reason it fails is that people over-engineer it. Here is the shape of the announcement that works for tiers 1-3 (tier 4 needs slightly different language, noted below).

Opening line: the fact, unadorned. “From [date], the retainer fee will move to [new fee] per month.” Do not open with “hope you’re well” or “we wanted to update you” — the client knows the email is about the fee the moment they see the subject line, and prolonging the pre-amble reads as apologising for a decision that should not need an apology.

The reason, one sentence, framed as delivery not cost. “The current fee no longer covers the scope we are delivering; the new fee brings that back into line.” Not “our costs have gone up”. Cost increases are the vendor’s problem; scope-value alignment is the client’s business.

What the client gets, one sentence. Either the scope is unchanged (say so plainly), or something specific is added (say what). Do not add a token deliverable just so there is something to point to — that reads as a bribe.

What happens if they do not respond. State that the new fee takes effect on the date named unless they raise an issue before then. Do not require an active acknowledgement — that creates a queue you will chase and eventually let slide. Passive acceptance is the correct default because it works with the way clients actually process emails.

A specific window to raise concerns. “If you want to discuss, my calendar is here [link].” A calendar link filters the response into a real conversation with people who need one, rather than an email thread with people who are just processing feelings about the number.

That is the entire template. Five sentences. If yours is more than five sentences, you are apologising in the extra ones.

Tier 4 needs one additional sentence

For the tier-4 announcement — the accounts you would happily lose at the wrong fee — add exactly one sentence between “what the client gets” and “what happens if they do not respond”:

“If the new fee does not fit into your budget, we can help you make a clean transition to a competitor over the next 30 days.”

Nothing else. The sentence is important because it says the quiet part out loud: leaving is an acceptable outcome. It is not a threat; it is an invitation. Half the tier-4 accounts you would expect to fight will take the offer and leave with less friction than the fight would have cost. The other half will read the sentence as a signal that the fee is not negotiable, and they will either accept it or negotiate softly for a phased implementation rather than a discount.

Do NOT put this sentence into the tier-1 or tier-2 emails. Announcing an exit path to accounts you want to keep is a message they will remember for years, and they will remember it as “the day you told me it was fine if I left”. Reserve it for tier 4, where the message is exactly the one you mean to send.

The meeting, if it happens

Most price-increase notices do not produce a meeting. The ones that do are the ones that decide whether the account stays at the new fee. The mistake in the meeting is treating it as a negotiation over the fee. The meeting is not about the fee; the fee is a fact. The meeting is about the delivery model — what the client gets for that fee, at what rhythm, with what response expectation.

Three moves in the meeting, in order:

One: re-anchor on what the fee is buying. Not “we deliver a lot of value” (which the client hears as “we are worth more than you think”). Something specific: “The retainer is a monthly deliverable, a strategy touchpoint, and response inside one working day on non-urgent things. That is what the new fee is priced against.”

Two: name what is negotiable and what is not. “The fee is fixed. What we can move is the mix of deliverables inside the fee, or the pace.” This gives the client a real conversation to have without touching the number.

Three: propose one specific concession, from a list of two. Either “we can make the increase phased — the new fee at 50% for the first quarter, then in full”, or “we can hold the current fee for one more quarter and revise at the next natural renewal”. Pick the one that fits the account and offer only that. Never offer both — offering two concessions makes the third one they ask for feel earned.

If the client accepts one of the two, write it into the follow-up email the same day, so the specific arrangement exists in writing rather than in a memory of the meeting. If they push for a third concession, refer back to move two — the fee is fixed, and this is the shape of the flexibility.

The one mistake that undoes all of this

The single most common mistake in price increases — bigger than any wording, bigger than any timing — is grandfathering an account privately without a natural end date.

The scenario is: a tier-1 or tier-3 account pushes back gently, and you agree to “keep them at the old rate for now”. The “for now” has no defined end. Six months later, you are still charging them the old fee, but the account has since accreted small out-of-scope requests that would have justified an even larger increase than the original one, and the “for now” has become “forever” without anyone deciding it should.

The way to grandfather someone without falling into this trap is to name a specific end date and put it in the same email that agrees to the grandfathering. “I’ll hold the current fee through your existing contract term (which ends [date]) and the new fee will apply from [date + 1].” That way “for now” has a boundary, the boundary is in writing, and the next transition is a re-affirmation of a scheduled change rather than a fresh confrontation.

Where this fits with the rest of the work

Price increases are one lever inside the larger question of retainer economics. If your problem is not that the fee is wrong but that the delivery model has quietly outgrown it, the diagnostic and re-scope script are in how a profitable retainer quietly becomes an unprofitable one — often the correct answer is not “raise the fee” but “re-scope the fee”, and both start from the same conversation.

If your problem is that you have not built your pricing model right from the start and every increase feels like a re-negotiation, the underlying decisions are in how to price your services as a freelancer or consultant — particularly the section on why retainers priced for hours reinvent hourly billing.

And if the account you would be raising the fee on is one you have decided you would happily lose, the four signals in when to fire a client tell you whether “raise the fee to what it should be” is the right last step of a relationship you were about to end anyway — the increase is often the honest exit if it lands as a decline.

FAQ

How much notice should I give before a price increase takes effect?

The standard is sixty days, but the four-tier sequence works better on a seventy-five-day window — tier 1 gets seventy-five days, tier 2 gets sixty, tier 3 gets forty-five to fifty, and tier 4 gets thirty to forty-five. The extra fifteen days on the front end is feedback time, not client-facing runway.

Should I raise prices on all clients at the same time or stagger them?

Stagger, in four tiers. Same-day-to-everyone collapses information you should be gathering sequentially into a single event you cannot learn from — you will not know how to word the next batch until you have seen how the first one landed.

What if a client asks to be grandfathered at the old rate?

Agree only with a specific end date written into the same email — “current fee through your existing contract term ending [date], new fee from [date + 1]”. Open-ended grandfathering is the single most common mistake in price increases; “for now” quietly becomes “forever” without anyone deciding it should.

How much should I raise prices by?

The post does not prescribe a percentage because the right number is the one that brings the fee back into line with the scope you are actually delivering, not a benchmark. Write out what the current fee is buying against what the delivery costs you, and price the gap — if the gap is small the increase is small, and if it is large the increase is large.

Should I explain the increase as rising costs or as increased value?

Neither exactly — frame it as scope-value realignment. “The current fee no longer covers the scope we are delivering” is the client’s business; “our costs have gone up” is the vendor’s problem and the client owes you nothing on it.

What if a client wants a meeting to discuss the fee?

Treat the meeting as being about the delivery model, not the fee — the fee is a fact, what is negotiable is the mix of deliverables or the pace. Offer exactly one concession from a prepared list of two (phased increase or a one-quarter hold), never both, and write the arrangement into a follow-up email the same day.

What if a tier-4 client leaves rather than accept the new fee?

Plan for it before you send the announcement — write down how you would fill the revenue gap from tier-1 growth, a pipeline addition, or a temporarily lighter month. Losing a tier-4 account at the wrong fee is an acceptable outcome; making concessions in the meeting because you have not planned for the exit is not.

The one-sentence version

Announce a price increase in four tiers over the same 75-day window rather than to everyone on one day — best-fit accounts first (for the honest signal), the one problem account you most want to keep second (isolated so its reaction is a clean data point), the steady middle in a batch third, and the difficult accounts last, with an explicit exit path in that fourth email only. The order is what decides whether the increase holds; the wording is the smaller half.