How to Raise Your Prices Without Losing the Customers You Have

Published July 28, 2026

Part of: Digital Products — our full guide on this topic.

Disclosure: Some links below are affiliate links. If you sign up through them we may earn a commission at no extra cost to you. We only recommend tools we'd genuinely suggest to a friend. See our full disclosure.

This site tells you to raise your prices roughly twenty times.

Price a digital product says most creators wait far too long. Price a membership says the same and adds that you should grandfather everyone already inside. Getting your first client on Fiverr says the low opening rate is a temporary investment. Newsletter sponsorship, coaching, launching a course, membership sites — all of them, in a line or two, tell you to charge more later.

Not one of them tells you how.

That is not an oversight so much as a symptom. Setting a price is a decision you make alone. Raising one is a decision other people find out about — and that second thing is a completely different job, which is why it keeps getting compressed into an aside at the end of an article about something else.

Here is the whole job, properly.

The quick version

• A one-time price and a recurring price are two different problems. Only the second one involves people who are already paying.
• “Grandfather your existing customers” is only meaningful for subscriptions. For a one-time product, past buyers already own it at what they paid.
• The commonest failure is changing the number and nothing else — the sales page still argues for the old price.
• Every grandfathered rise adds a price tier you maintain forever. Promise “for life” only if you can still say it in five years.
• Check what your platform can actually do before you announce it, not after.
• Put the number and the date in the first two lines of the announcement. Do not bury the lede and do not apologise.
• Tell only the people it affects. Grandfathered members do not need an alarming email about a change that skips them.
• Find every place the old price is written, including the ones you cannot search — images, video, other people’s pages.
• You can lower a price back. You cannot un-tell people.

One-time and recurring are not the same job

Almost all the confusion about raising prices comes from treating these as one topic. They are not.

A one-time price only applies to sales you have not made yet. Nobody who already bought is affected by a change — they paid once, they own the thing, and the number on your checkout page has nothing to do with them any more. There is no obligation to announce anything, nobody is worse off, and the only thing genuinely at risk is your conversion rate on future visitors. This is the easy case, and people still get it wrong, for a reason covered in the next section.

A recurring price is an ongoing arrangement with people who are already paying, and changing it changes something they are living inside. That is a different act with different obligations, including — depending on where you and your customers are — notice requirements that are not up to you. Do not guess at those; check what applies to you, and check what your payment platform requires as well, since some have their own rules about how subscription price changes must be communicated.

Services sit between the two. A new rate applies to work you have not agreed to yet, so structurally it behaves like the one-time case, but the person you are quoting it to is usually an existing relationship rather than a stranger — which is why raising a rate with a long-standing client feels harder than raising it on your public rate card. If you work from a written scope, the rise belongs at a natural boundary: the next project, the next retainer period, the renewal of the contract — not mid-engagement.

Worth being honest about where that difficulty actually comes from, because it is rarely about the client’s reaction. The rate that is hardest to raise is almost always attached to the largest share of your income, and the reluctance is a measurement of dependence rather than of politeness. If one client is most of what you earn, read client concentration risk alongside this — the rise is still the right move, and it doubles as the cheapest information you will get about how solid that relationship really is, obtained at a moment you chose rather than one they chose.

Everything below splits along that line.

The occasions that hold up

A price rise lands well or badly largely depending on whether something real changed. Four occasions that carry their own explanation:

1. The product genuinely got better. This is the strongest one you will ever get, and updating a digital product makes the case in full: the thing is honestly worth more than it was, so you can say so without spin. The pairing that makes it work is giving the update free to past buyers — it costs you nothing you were going to collect, and it means the improvement arrives as good news to the people most likely to recommend you.
2. You are consistently at capacity. For services and coaching, a full calendar is not a scheduling problem, it is the market answering a question you did not ask. If every enquiry converts and you are turning work away, the price is below what people will pay. The one check worth doing first is whether the fullness is structural or just a bad month — what to do when you have more work than you can take walks through that distinction, because repricing off the back of one collision month solves a problem you do not have.
3. Your costs changed. Tools, fees, the platform cut. This is a real reason and a checkable one — and if you have never actually audited what your stack costs, do that before you use it as a reason, because it is embarrassing to raise a price over costs you could have removed.
4. The price was simply wrong. You set it before you had any evidence, on the basis of nerves. Now you have sold some and know more. Fixing an early guess is a perfectly honest reason, and it does not require a story.

And one occasion that does not hold up: you need money this month.

It is the worst reason not because it is dishonourable but because it does not work. A price rise only touches sales you have not made yet, it typically dents conversion in the short term, and its effect on revenue is slow and hard to read. If the actual problem is cash in the next four weeks, a time-limited promotion raises money faster and commits you to nothing permanent. Reaching for a price rise there gets you the slowest possible answer to the most urgent possible question.

Raising a one-time price

This is the easy case, and it is the one people fumble, because of a single mistaken assumption: that a price is a field in a checkout.

A price is not a field. It is a claim your sales page has to earn. Every element on that page — the promise, the proof, the detail, what is included, who it is for — was assembled to make a specific number feel reasonable. Change the number and leave the argument alone and you have not raised the price of the product; you have lowered the perceived value of the page. That is why a rise that should have gone unnoticed sometimes reads to visitors as worse value than before, even to people who never saw the old price.

So the order is: improve the argument, then change the number. Practically that means the sales page should say more than it did before the new price goes live — the extra module that got added, the results people have had since, the reviews you did not have last year, a clearer statement of exactly what the buyer walks away with. If nothing on the page can honestly say more than it said before, that is worth sitting with, because it is the question of whether this is the right moment.

Two things you do not owe anyone here:

• An announcement. You can simply change it. Nobody is worse off. If you have an audience that would find a quiet change odd, tell them; if you do not, silence is not deception.
• The old price for latecomers. If someone emails after the change saying they meant to buy last week, honouring it is a kindness and entirely your call. It is not an obligation, and doing it as a policy quietly means you did not raise the price.

The one announcement worth making is a closing window — the price goes up on the first, this is the last week at the current price. It is genuinely effective and, as running a sale puts it, honest only if you actually raise the price and keep it raised. The deadline is real or it is not. There is no third option, and the audience finds out which one it was.

Raising a recurring price

Here you are changing something people are already inside, so the decision has a shape the one-time case does not. Three honest options:

New customers only. The price rises for people who join from now on; everyone already subscribed stays where they are. This is the lowest-risk route by a distance. It costs you nothing you are currently collecting, it gives you no announcement to dread, and it turns your existing members’ rate into a reason to stay rather than cancel and rejoin later. It is the default recommendation in pricing a membership, and for most solo businesses it is simply correct.

Everyone, with notice. Occasionally necessary — your costs genuinely moved, or the gap between what old and new members pay has become indefensible. It is the riskiest option and it is the one that needs the announcement in the next section done properly.

Close the old plan and offer a move. The current plan stops accepting new members and existing members keep it as long as they stay; the new plan sits alongside it, usually with something the old one does not have. This gives people a reason to move up rather than an instruction, and it is the gentlest way to eventually retire a legacy rate.

The cost of grandfathering that nobody mentions

Holding existing members at their old price is good advice. It is also not free, and the bill arrives years later.

The same trap catches freelance retainers, in a sharper form. A monthly amount from a long-standing client stops being a price you quote and becomes a figure that simply arrives, and at that point raising it starts to feel like asking for a pay rise rather than repricing a service — which is a fair signal that the relationship itself has drifted, because you have acquired an employee’s psychology without an employee’s annual review.

Every rise you grandfather creates a price tier you maintain forever. Two the first time. Three the second. And because the tiers do not move, the oldest one drifts further below the current price with every increase, until eventually you have members paying an amount that no longer resembles what the thing costs. At that point you have exactly two options, and one of them is breaking a promise you made in writing. There is a third moment the word has to survive, and it is the one nobody pictures while writing a sales page: the day the thing ends. A “for life” rate on a membership you later close is a promise with no term left to run out, and it is the hardest case in closing a membership precisely because there is nothing to pro-rate.

Worth knowing that this trap is not specific to price. Every version of anything you agree to hold for the people who were already there — an old format, a retired feature, a tier that no longer appears on your sales page — is a version you maintain on exactly the same terms, and changing something people already paid for is where that decision gets made for everything that is not a number.

So before you write the word forever, apply one test: can you still say this sentence out loud in five years? If the honest answer is “probably, unless the business changes a lot”, promise something you can actually keep instead. Both of these are honest, checkable, and self-limiting:

• “This price for the next twelve months.”
• “This price for as long as you stay subscribed without a break.”

The second one is particularly good, because it gives the member a genuine reason not to cancel during a quiet month and it gives you a natural end to the obligation without you having to end it.

Check the platform before you promise

There is a purely mechanical trap here. Platforms differ in how they handle a price change on an existing plan: some hold current subscribers on the price they signed up at and apply the new one only to new sign-ups, some apply the change to everyone on that plan at their next renewal, and some require you to build a separate plan to achieve the first behaviour.

Find out which one yours does before you announce anything. Do not assume, and do not take a general answer from an article — including this one — as a statement about your specific tool and plan. This is a five-minute check in your platform’s documentation or a message to their support, and the alternative is discovering the answer when a customer’s card is charged an amount you promised it would not be. The comparisons of the main selling platforms are the place to start if you are still choosing; if you have already chosen, read that platform’s own docs on subscription price changes.

The announcement

A price rise is the only thing you will ever announce to people who are worse off for hearing it. Every other announcement — a new product, an update, a launch — is good news you are trying to get read. This one is not, and writing it like a launch email is how it goes wrong.

Four things, in this order:

1. The number and the date, in the first two lines. Do not open with a paragraph about the journey. People scanning for what changed and when will find it or they will imagine something worse.
2. What stays the same for them. For most readers this is the actual news. If you are holding existing members at their current rate, this sentence is the entire email.
3. The reason, once, briefly. One or two sentences. You are explaining, not defending.
4. How to leave, stated plainly. Making the exit easy is both the decent thing and the practical one — a customer who leaves cleanly may come back, and handling cancellations well is the difference between a departure and a complaint.

Three things to leave out:

• The apology. “I’m so sorry to have to do this” invites the obvious reply: then don’t. You are allowed to charge what the thing is worth.
• The long justification. Length reads as guilt. A four-paragraph explanation makes a reasonable decision look like one you do not believe in.
• Manufactured urgency. If there is a real window before the new price, say so once. Do not dress a price rise as a limited-time opportunity; people can tell, and it costs you the goodwill the honest version would have kept.

And one distribution rule that matters more than the wording: send it only to the people it affects. If you have grandfathered your existing members, they should get a short, calm note that the price is rising for new members and nothing about their plan changes — or, depending on your relationship with them, nothing at all. Sending the full announcement to everyone means a large number of people who were not affected spend a few minutes wondering whether they are, and some of them will use the moment to reconsider a subscription they had not been thinking about. That is churn you generated yourself, for no reason.

The old price is written in more places than you think

This is the part that actually costs people money, and it is invisible until a customer finds it.

The price is not in one place. It is in the field that takes the payment, and then it is in every place you have ever written it down. Go and find them:

• The sales page and any pricing table, including comparison tables that name other products alongside yours.
• The checkout or payment link itself — the number that actually gets charged. Verify this one by looking at it, not by assuming the change saved.
• Your email sequences. The welcome sequence, the launch sequence, the abandoned-cart flow, the free course that ends in a pitch. These run unattended for years and are the single most likely place a dead price survives — which is the whole argument in auditing the automations that are still sending.
• The lead magnet. A PDF, checklist or template that names a price is a file already sitting on other people’s computers. You cannot fix the copies, but you can fix the one being downloaded today.
• Your own older articles and pages. Anything on your site that mentions what you charge.
• The FAQ. Almost always contains a price and almost never gets re-read.
• Screenshots and video. A price inside an image or a recording cannot be found by searching your site’s text. Make a list of these deliberately, because no search will surface them for you.
• Other people’s pages. Marketplace listings, directory entries, your affiliates’ promotional material, a bio link, a guest post, an interview. You do not control these, but you can ask — and an affiliate quoting a commission as a fixed cash amount rather than a percentage now has a wrong number too.

A stale lower price sitting somewhere public is either a promise you have to honour or an argument you have to have. Both cost more than the fifteen minutes of searching would have.

Then rot-proof it, using the same rule that keeps email sequences from decaying: write the price in as few places as possible and point at it everywhere else. A page that needs to persuade can usually say “starts at” and link to the page that carries the real number, or simply say what you get and let the checkout state the price. Every place you write the number is a place you will have to remember next time — and next time is exactly when you will forget.

Knowing whether it worked

Two honest warnings before any of the signals below.

Wait longer than feels comfortable. A solo business rarely has the volume for a week of data to mean anything. Traffic moves for reasons that have nothing to do with price, and a run of quiet days after a change will read as causation whether or not it is.

Attribution is weak at small numbers. You will rarely get a clean answer, and no amount of staring at the dashboard will produce one. The practical consequence is not despair — it is change one thing at a time, so that whatever you do learn is attached to something.

With that said, what to actually look at:

• Revenue over a comparable period, not the number of sales. Fewer sales at a higher price is frequently the same money or more, with less support work attached. Counting sales alone will tell you the rise failed when it did not.
• For a subscription: cancellations across the first billing cycle after the change, and again at the next renewal. Some churn is expected and does not mean the decision was wrong. The second checkpoint matters because a monthly plan reacts quickly and an annual one takes a year to tell you anything.
• The shape of the questions you get. This is the most useful signal and the least measurable. If objections shift from is this for me to is this worth it, the price has moved ahead of the argument on the page — which is a conversion problem, not a pricing one, and the fix is the page rather than the number.
• Refund requests. A rise in refunds after a price change is worth taking seriously, because a refund means someone bought at the new price and then concluded it was not worth it. That is a clearer signal than not buying at all.

If you need to reverse it

You can lower a price back to where it was. You cannot un-tell people.

That asymmetry is the whole difficulty of a reversal. Everyone who paid the higher price now knows they paid more than the current one, and the fact that you were within your rights does not change how it reads. Two routes:

• Revert quietly. Only works if the rise was never widely announced and few people bought at the higher price. Legitimate, if genuinely quiet.
• Refund the difference to the people who paid it. Expensive, and the only version that leaves nobody feeling cheated. If you announced the rise loudly, this is the option that keeps your word intact.

Which is a good argument for raising prices in a step you can defend rather than a jump you might retreat from. The reversal is always more costly than the smaller rise would have been.

Five mistakes

1. Changing the number and nothing else. The page still argues for the old price, so the new one reads as worse value.
2. Announcing to everyone, including the people it does not affect. You have just invited a group of unaffected customers to reconsider their subscription.
3. Promising a price for life before checking whether your platform can hold it and whether you can.
4. A long, apologetic explanation. It makes a defensible decision look like one you do not believe in.
5. Leaving the old price live somewhere public — a sequence, a PDF, a marketplace listing — where a customer will find it and hold you to it.

The pass, start to finish

1. Decide which kind of rise this is — one-time, recurring, or a service rate — because the rest of the steps differ.
2. Name the occasion in one sentence. If the sentence is “I need money this month”, run a promotion instead.
3. Improve the argument before the number. The sales page should say more than it did.
4. Check what your platform actually does to existing subscribers when a plan price changes. Read the docs; do not assume.
5. Choose your grandfathering promise and make sure you can keep it. “For as long as you stay subscribed” beats “forever” in almost every case.
6. Write the announcement — number and date first, what stays the same second, one short reason third, an easy exit fourth. Send it only to the people affected.
7. Hunt down every other place the old price is written, fix the ones you control, ask about the ones you do not, and cut the number of places it appears at all so the next rise is a smaller job.

Then leave it alone long enough to learn something. The single most common mistake left after all the others is judging a price change in the first week, panicking, and reversing a decision that was working.