How to Raise Prices Without Losing Customers
The arithmetic is generous. It's the execution that goes wrong.
Summary
- At 50% contribution margin, a 10% price increase leaves you better off even if you lose 16.7% of your customers. Most companies lose under 3%.
- Increases fail in execution, not arithmetic: no notice, no choice offered, a front line that hasn't practised the objections, and nobody measuring what was actually collected.
- Start with new customers only, size it at 5 to 10%, give grandfathering an end date, and put an escalation clause in every new contract.
Most Indian companies have never run a deliberate price increase.
Not because they weighed it up and decided against it. Because it was never on the list of things a company does. The price got set in year one, the business got busy, and it stayed there while salaries rose, input costs rose, and the product got considerably better.
With the new financial year starting on 1 April, this is the window in which most Indian businesses either revise their rate cards or quietly decide not to for another year. So here's the arithmetic first, and then the sequence.
Start with the maths, because it's more generous than you think
If you raise your price by p and your is CM, the share of volume you can afford to lose and still be no worse off is:
Affordable volume loss = p ÷ (CM + p)
At 50% contribution margin, a 10% increase leaves you better off even if you lose 16.7% of your customers.
Most companies lose under 3%.
Compare that to the other direction. At the same margin, a 10% discount needs 25% more volume just to break even. Increases tolerate large losses. Discounts demand enormous gains.
That asymmetry is the whole reason price is the strongest lever in pricing, and the reason a company that has never raised prices is leaving the easiest money on the table.
The seven steps
1. Gather the evidence
Before you decide anything, get five things in front of you.
How long since the last change. If it's more than eighteen months, your real price has already fallen.
Your win rate. If you're winning more than about two-thirds of what you quote, you are almost certainly under-priced: you're winning deals you should have had to work for.
Your discount depth, and its spread. Look at the histogram, not the average. If a chunk of your base sits at deep discounts, fix that before raising list prices, or you'll simply widen the gap.
What you've shipped. Every feature, improvement, integration and service added since the last price change. This becomes your communication.
Where your costs have moved. Not to justify the increase to yourself, because cost-plus is a weak argument, but so you know what you're recovering.
2. Decide who it applies to
Three groups, three different levels of difficulty.
New customers only. The safest, and almost always the right first increase for a company that's never done this. No existing relationship at risk. You learn what happens to win rate before you touch anyone who's already paying you.
Existing customers at renewal. The second increase, once you've seen the new-logo data.
Existing customers mid-contract. Only possible if you have an escalation clause in the contract. If you don't, this is the moment to start putting one in every new agreement. A 6% annual uplift in a three-year contract puts year three over 12% above year one, requires no negotiation, and is far easier to agree at signature than to impose later.
Most Indian founders have never asked for one. Start.
3. Size it
5 to 10% is absorbed quietly by most B2B customer bases. This is where you should be unless you have specific evidence to go further.
Above 15% needs a packaging change alongside it: new tiers, a different structure for what you charge per, additional inclusions. Otherwise you'll trigger a procurement review you didn't want. Above 15% on an unchanged product invites the question "why now?" and you need a better answer than "our costs went up."
Different amounts for different groups is legitimate and usually correct. The customers on your deepest discounts have the most room. The ones already paying near list have the least.
4. Decide what happens to existing customers
Grandfathering is a tool, not a kindness, and it needs an end date.
Open-ended grandfathering creates a group of customers on terms nobody can explain, a price list with two versions, and a problem that gets harder every year. Twelve to eighteen months, stated clearly at the time, is the right shape.
And put a floor under it. Any customer already below your contribution margin floor should move regardless of when they signed. That isn't a price increase, it's stopping a loss.
5. Communicate with what they got, not with your costs
This is where most increases succeed or fail, and it comes down to what the letter says.
Lead with what's changed for them. Everything you've shipped since their last price change. If you can't fill a paragraph with that, you have a product problem, not a pricing problem.
Give 60 to 90 days' notice. Enough for their budget process. Not so long that it's forgotten.
Give them something to choose. This is the part almost everyone skips, and it's what turns an announcement into a conversation. Options like: accept the new price now, or lock the current price for two more years by committing today, or move to a different plan that suits you better.
A customer who is given a choice feels agency. A customer given an announcement feels done to.
Same increase, completely different reaction.
Say it plainly and once. An over-apologetic letter signals that you don't believe your own price, and invites negotiation you weren't offering.
6. Arm the front line
Price increases don't fail in the boardroom. They fail on a call with a salesperson who wasn't ready.
Before the letters go out, your team needs the answers to the eight objections you know are coming, a clear escalation path, a set of pre-approved retention offers so nobody is improvising under pressure, and a named person who decides anything unusual.
And they need to have practised it. A team hearing the objection for the first time from a customer will concede.
7. Measure what you actually collected
The step everyone forgets. You announced 10%. Ninety days later, what did you collect? If the answer is 6%, you ran a 6% increase, and the gap is the most useful thing you'll learn from the whole exercise.
Track how many customers accepted at full, how many negotiated and to what level, how many churned, and what your win rate on new deals did. That data is what makes next year's increase better. Without it you'll be guessing again.
What actually causes churn
Worth naming, because founders' fears here are usually misplaced. Customers rarely leave over a modest, well-explained increase. They leave over:
- Surprise. No notice, or an increase they found out about from an invoice.
- Unfairness. Discovering someone comparable pays less. This is why an ungoverned discount history makes every future increase harder.
- Bad timing. Landing during their budget freeze, or a week after a service failure.
- No perceived progress. If the product genuinely hasn't improved, the increase reads as opportunism. And they're not wrong.
Three of those four are execution problems. Only the last one is about the price.
If you're doing this for 1 April
A practical sequence, working backwards.
- Now: gather the evidence. Win rate, discount spread, time since last change, what you've shipped.
- Two to three weeks out: decide the scope and the size. Draft the communication. Prepare the objection handling and get the team to practise it.
- By mid-March: letters out, so customers have notice before the new year starts.
- 1 April: new rates effective for new business. Existing customers at their renewal date.
- End of June: measure what you actually collected.
And whatever else you do this year, put an escalation clause in every new contract you sign. It's the cheapest price increase available to you, and it means next year's conversation is one you don't have to have at all.
Where to start
Work out your own affordable volume loss. Take your contribution margin, take the increase you're considering, and run the number. For most companies it's a bigger figure than they expected, and seeing it is what makes the decision easy.
Our Pricing Maturity Assessment will tell you whether a price increase is your best move right now, or whether fixing your discount discipline first would be worth more.
Frequently asked questions
How much can I raise prices without losing customers?
Run the arithmetic: affordable volume loss = increase ÷ (contribution margin + increase). At 50% contribution margin a 10% increase works even if you lose 16.7% of customers, and most companies lose under 3%. In practice, 5 to 10% is absorbed quietly by most B2B bases.
Should I raise prices for existing customers or only new ones?
New customers first, if you've never done this before. You learn what happens to your win rate without risking existing relationships. Move existing customers at their next renewal, and never mid-contract unless you have an escalation clause.
How much notice should I give before a price increase?
Sixty to ninety days. Enough for the customer's budget process, short enough that it doesn't get forgotten. Lead the communication with what they've gained since their last price change, not with your rising costs.
How useful was this article?
One tap. It tells us what to write more of.
The next one
Get what we publish next, by email.
Working notes on raising, borrowing, protecting, growing and structuring capital in India. One email a week at most, and you can leave any time.
We use your address only to send this. See our privacy policy.
We store your address to send you these emails and nothing else. See our privacy policy.
Related reading
