The Price Increase Playbook: raise prices on the customers you already have

Almost every business raises prices the easy way: it charges new customers more and leaves everybody else alone. It feels like loyalty. It is actually a slow transfer of your margin to the people who have been with you longest, and it compounds. Five years of 6% cost inflation and no increase means your oldest accounts are paying about 75 cents on the dollar — usually while consuming the most support, the most custom work and the most of your attention.

The reason nobody fixes it is not analysis. It is fear of the email. So this playbook skips the theory and gives you the arithmetic that tells you how much churn you can afford, the sequence that de-risks the rollout, the notice itself, and the four scripts that cover roughly 90% of the pushback you will get.

Budget about three hours for the prep and four weeks for the rollout.

Step 1: Do the break-even math before anything else

The single most useful number in this entire exercise is the churn you can absorb and still come out level. It is one line of arithmetic:

Break-even churn = 1 − (1 ÷ (1 + increase))

Worked example. You have 200 customers paying $200 a month — $40,000 in monthly revenue. You are considering a 12% increase.

• 1 ÷ 1.12 = 0.893, so break-even churn = 10.7%.
• That is about 21 of your 200 customers. If 21 leave, you are exactly where you started on revenue.
• But you are meaningfully ahead on profit, because serving 179 accounts costs less than serving 200. Support tickets, onboarding, invoicing and the emotional overhead all fall with the headcount.

Write that number on a card: “I can lose 21 customers and be fine.” When three angry replies land on day two, that card is the difference between holding your price and folding. Three is not twenty-one.

Two adjustments before you move on. If your gross margin is below about 40%, the break-even churn is lower than the formula suggests, because each lost customer takes contribution with it — run it on gross profit rather than revenue. And if a single account is more than 10% of your revenue, take it out of this exercise entirely and handle it as its own negotiation.

Step 2: Segment the base into four buckets

A flat increase across everyone is the version that generates the most noise for the least money. Sort the list — a spreadsheet is fine — into four groups.

• Underpriced and healthy. Long tenure, low support load, using the product heavily, paying an old rate. This is where most of your money is. Full increase, sometimes more.
• Underpriced and expensive. Paying an old rate and consuming disproportionate support or custom work. Full increase, and quietly hope for the churn. Some of these accounts you would rather lose.
• Correctly priced. Signed recently at current rates. Leave them alone entirely, or apply a small indexed rise. Nothing corrodes trust faster than raising a price you set four months ago.
• Strategic. Reference customers, logos you use in sales, accounts with genuine expansion ahead. Hold them for one cycle and increase them next time, having banked the reference value.

The pattern that usually emerges: 20–30% of accounts carry most of the underpricing. You do not need a universal increase. You need a targeted one.

Step 3: Set the number using three anchors

Pick the increase from evidence, not from what feels survivable at 11pm.

1. The gap anchor. What would this customer pay if they signed today at list? That difference is the ceiling. Closing it in one step is usually too aggressive; closing half of it is normal.
2. The cost anchor. What has actually gone up since their rate was set — wages, hosting, model inference, insurance, materials? This is the number you can defend out loud, and it is the only one the customer will consider reasonable.
3. The value anchor. What did you ship since they signed? List it. If the list is thin, the increase should be smaller and the notice should lean on cost, not value. Do not claim improvements you cannot name.

For most businesses the answer lands between 8% and 15%. Below 5% you have annoyed people for rounding error — either skip it or index it automatically. Above 20% you are not doing a price increase, you are doing a repositioning, and it needs a conversation rather than an email.

Step 4: Sequence the rollout — never send it to everyone at once

Run it in three waves, a week apart. This is the step people skip and the one that saves the whole exercise.

• Wave 1 — the 10% test. Twenty accounts drawn from across the buckets, weighted toward the ones you are least afraid of losing. Send, then wait five business days.
• Read the result. Under 10% pushback: proceed unchanged. 10–25%: proceed, but tighten the notice wording where the objections clustered. Over 25%: stop. Your number is wrong or your timing is, and you have learned it for the price of twenty emails instead of two hundred.
• Wave 2 — the bulk. Everything except strategic accounts and your largest three customers.
• Wave 3 — the hand-carried ones. Largest accounts and anyone with a real relationship. These get a call or a personal note before the formal notice arrives. Never let a big customer learn about a price change from an automated email.

Step 5: The notice

Short, dated, unapologetic, no wall of gratitude. Apology invites negotiation; length invites suspicion. Adapt this:

The rules underneath it: give 60 days’ notice for monthly billing and a full cycle for annual. State the old number and the new number in currency, not percentages — “$200 to $224” reads smaller than “up 12%” even though it is identical. Name a real date. Send it from a person, not billing@. And never bundle a price rise with a feature removal, a plan migration or a terms change; each of those is its own conversation, and stacking them turns a routine increase into a reason to shop around.

Step 6: Four scripts for what comes back

“This is a lot. Can you hold my rate?”
“I can hold it through [date] if you move to annual billing today. That is the version I can do — it costs us less to serve, so I can pass that back. What I can’t do is keep the monthly rate where it is, because it hasn’t moved since 2023.”

“We’ll have to look at alternatives.”
“That’s fair, and I’d do the same. Before you spend the time: the things that would take you a while to rebuild are [X] and [Y]. If you do want to look, I’ll give you an export in whatever format helps. And if you come back, your rate is the new one, not a penalty.” — Then stop talking. Offering a discount into this silence is how a routine increase becomes a permanent one-off exception.

“What am I actually getting for this?”
“Since your rate was set: [three specific things]. If none of those are useful to you, that is worth a proper conversation — you might be on the wrong plan, and there may be a cheaper one that fits better.” Downgrading a mismatched customer is a win, not a loss.

“We can’t afford it right now.”
“Understood. I can push your start date to [60 days later], once. After that the new rate applies.” A dated deferral costs you two months. An open-ended exception costs you forever, and it will be quoted back to you at the next increase.

Step 7: The save ladder — in this order, never skipped

Decide the concessions in advance, in strict sequence, and do not jump to the bottom because someone sounded upset.

1. Deferral — same increase, later start date.
2. Annual prepay at the old rate for one term.
3. Phased increase — half now, half in six months.
4. Downgrade to a smaller plan at the new rates.
5. Full hold — reserved for accounts you would genuinely chase to win back. Cap this at 5% of the base and write down who used it.

The failure modes

• Folding on day two. The people who reply fast are the people who always reply fast. The silent majority pays. Look at your card.
• Discounting in the notice. Any “but let me know if this is a problem” in the original email converts the increase into an opening bid. Take it out.
• Explaining too much. Three sentences of reason is credible. Six paragraphs reads as guilt, and guilt reads as negotiable.
• Raising the price without fixing the known problem. If there is an open complaint on an account, resolve it first. Charging more for something visibly broken is the one version that generates genuine anger.
• Doing it once and never again. The whole reason this is painful is that it has been years. Put a recurring 12-month review in the calendar now, while the spreadsheet is open. Annual 5–7% adjustments are boring. Boring is the goal.

Measuring it honestly

Count at 90 days, not at 7. The number that matters is net revenue: total billing after the increase and after every departure and concession. Track four things — percentage of accounts that pushed back, percentage that actually left, revenue retained, and how many landed on each rung of the save ladder. That last one tells you whether your number was too high or, far more often, too low.

The uncomfortable finding most people hit: churn comes in well under break-even, and the accounts that leave are disproportionately the ones that were costing the most to serve. Which means the increase you were afraid of was smaller than the one you should have run.

The four-week calendar

• Week 1. Break-even math, segment the base, set the number, draft the notice.
• Week 2. Wave 1 to twenty accounts. Log every response verbatim.
• Week 3. Read the result, adjust wording, send Wave 2. Book the calls for Wave 3.
• Week 4. Wave 3 by phone. Schedule the 90-day measurement and the next annual review.

The customers you already have are the cheapest revenue you will ever touch. Charging them what the work is worth is not a betrayal of the relationship — it is the thing that keeps you around long enough to have one.