# How to Raise SaaS Prices

> When to raise prices, by how much, who to exempt, and what churn to expect, with data on 7 to 20 percent increases and the notice periods that limit damage.

Source: https://saas-marketing.net/guides/saas-price-increases/
Topic: SaaS Pricing
Type: guide
Published: 2026-09-11
Last updated: 2026-09-11
Publisher: SaaS Marketing (saas-marketing.net)
License: CC BY 4.0. Quote or republish with attribution and a link to https://saas-marketing.net/guides/saas-price-increases/

## Short answer

Raise SaaS prices when win rates are above 40 percent, discount pushback has disappeared, and you have shipped meaningful value since the last change. Most increases land between 7 and 20 percent. Give 60 to 90 days notice for annual contracts, grandfather existing customers for a fixed window rather than forever, and model the breakeven churn first: a 10 percent increase stays revenue-positive until roughly 9 percent of affected revenue leaves.

## Key takeaways

- A 10 percent price increase breaks even at about 9 percent revenue churn, so the realistic downside is smaller than teams fear.
- Win rates above 40 percent with no discount pushback are the clearest signal you are underpriced.
- Permanent grandfathering is how companies end up with eleven price books and a billing system nobody can change.
- A 3 to 5 percent annual uplift clause in enterprise contracts beats a 25 percent jump every four years.
- Give 60 to 90 days notice on annual contracts and check the MSA, since many cap increases at renewal.
- Expect elevated support volume for three weeks and staff for it, or your NPS takes the hit instead of your pricing.

---

Four questions decide a price increase, and only four. Is now the right time. How much. Who gets grandfathered. What will it cost in churn. Everything else is rollout mechanics, which matter but are downstream. Get the four answers on one page before you brief anybody, because the meeting where someone says "what if everyone leaves" needs a number in response, not a shrug.

## Is now the right time? Five trigger signals

If three of these five are true, you are underpriced and the increase is overdue.

**Win rate above 40 percent.** Consistently winning more than two in five qualified deals means you're the obvious choice at your price, which is the definition of leaving money on the table. Below 25 percent, price is not your problem and raising it will make things worse.

**Discount pushback has disappeared.** When your reps stop hearing "that's more than we budgeted" and start closing at list, the market has repriced you upward and your price book hasn't caught up. Pull the last 100 closed-won deals and look at average discount. If it's fallen below 8 percent, that's your signal.

**You've paid down feature debt.** You shipped things customers asked for. A price increase with a genuine value story attached is a different conversation from one without. If the last twelve months produced nothing a customer would name, wait a quarter and ship something first.

**Input costs moved.** Infrastructure, particularly inference costs for any product with an AI component, has changed real margins for a lot of SaaS in the last two years. That is a defensible reason customers understand.

**The category repriced.** If comparable products have moved and you haven't, you're now the cheap option, which does more damage to your positioning than the revenue does to your P&L.

Never raise prices in the same quarter as a visible outage, a feature removal, or a support quality collapse. And do not do it while a funded competitor is running an explicit displacement campaign against you on price, because you will hand them the proof point. Wait two quarters. The revenue is still there.

## How much? Sizing against elasticity and contract terms

Most increases that work land between 7 and 20 percent. Here's how to pick within that band.

Three modifiers on those numbers. Enterprise churns less because switching costs are higher and the contract runs to a renewal date anyway. Seat-based products churn less than usage-based, because the increase is visible per seat rather than compounding with consumption. And products with data lock-in, integrations or workflow embedding churn dramatically less than point tools, which is why Salesforce can move price and a standalone scheduling tool cannot.

Above 25 percent, stop calling it a price increase. That's a repackaging, and it needs new tiers, new feature allocation and a new value narrative, not an email about pricing. The elasticity mechanics are set out in [price elasticity in SaaS](/glossary/price-elasticity/), and if you're restructuring tiers at the same time, [price anchoring](/glossary/price-anchoring/) determines whether the new top tier makes the middle one look cheap or the whole thing look greedy.

## The breakeven model: what churn can you actually absorb?

This is the number that ends the argument in the room. It's simple arithmetic and you should have it on a slide.

If you raise prices by P percent, you break even when the percentage of affected revenue that churns equals P divided by (100 plus P), times 100.

| Increase | Breakeven revenue churn | Revenue at 3% churn | Revenue at 8% churn |
| --- | --- | --- | --- |
| 7 percent | 6.5 percent | +3.8 percent | -1.6 percent |
| 10 percent | 9.1 percent | +6.7 percent | +1.2 percent |
| 15 percent | 13.0 percent | +11.5 percent | +5.8 percent |
| 20 percent | 16.7 percent | +16.4 percent | +10.4 percent |

Worked example on a 10 percent increase applied to 6,000,000 dollars of affected ARR. New revenue if nobody leaves is 6,600,000. If 3 percent of the base churns, you keep 97 percent of accounts at the new price: 6,402,000, a gain of 402,000. You'd need to lose 9.1 percent of that revenue before you were worse off than doing nothing.

The number that should worry you is not the arithmetic. It's the second-order effect: churned accounts don't just take their revenue, they take their expansion, their references and occasionally a public complaint. Model the revenue, then ask separately whether the accounts most likely to leave are the ones you most need. If your ten loudest advocates are all on the legacy plan you're about to reprice, that's a different decision. The [price increase impact calculator](/calculators/price-increase/) runs the full model against your own cohort data.

**9.1%** Revenue churn a 10 percent price increase can absorb before it loses money

## Grandfathering: four options, and the one I'd pick

Grandfathering is the lever that converts a churn risk into a goodwill event, and the one that most often gets set wrong in the direction of generosity.

Take the time-boxed twelve month option in most cases. You get the goodwill, customers get real time to budget, and on a known date everyone lands on the same price book. The alternative that seems kinder is permanent grandfathering, and it's the reason companies end up supporting eleven legacy price points, where sales can't quote consistently, support can't answer questions, and any future packaging change requires a migration project that nobody will approve.

If you go permanent, at least make it conditional: grandfathered as long as the customer stays on the same plan, doesn't add seats above a threshold, and doesn't lapse. The moment they change anything, they move to current pricing. Write that into the announcement.

Put a 3 to 5 percent annual uplift clause into new contracts now, indexed to a published inflation measure with a stated cap. Enterprise buyers accept these routinely. Two years from now, your pricing moves automatically and you never run this project again. The companies doing large disruptive increases are usually the ones who skipped this clause five years ago.

## Notice periods, contracts and the legal check nobody runs

Before anything ships, someone has to read the master services agreement. Not skim it.

Many enterprise MSAs specify a minimum notice period for price changes, commonly 60 or 90 days. A good number also restrict increases to renewal, cap them at a stated percentage, or contain a most-favoured-customer clause that means repricing one account triggers obligations to others. Breaching one of those turns a revenue project into a legal exposure, and it will be found, because procurement teams keep their contracts.

Practical notice standards: 90 days for annual enterprise contracts, 60 days for annual SMB, 30 days minimum for monthly. Communicate in writing to the billing contact and the account owner, because the person who reads the email is often not the person who signed.

**Rollout sequence**

The email itself is the most scrutinised thing you'll write that quarter, and there are working versions in [price increase announcement emails](/templates/price-increase-announcement-email/). The full operational sequence, including the exception approval matrix, is in the [price increase rollout playbook](/playbooks/saas-price-increase-rollout/).

'We've made the difficult decision', 'due to rising costs', and 'we hope you understand'. All three make the increase about you. Customers accept price increases when the value story is about them, and they resent apology-shaped announcements because the apology implies you know it isn't justified.

## What to watch after, and the honest failure mode

Three numbers at 30, 60 and 90 days: revenue churn against your model, support volume, and win rate on new business at the new price.

That last one is the one teams forget. Existing customers are sticky, so an increase can look successful for two quarters while quietly suppressing new acquisition. If win rate on new deals falls more than 10 points and stays there, you've priced above the market and no amount of goodwill from the installed base compensates. That's the genuine failure mode, and it's invisible if you only measure churn.

The honest tradeoff is this: a price increase is the highest-margin revenue available to you, and it also spends trust you can't easily rebuild. Do it deliberately, once every 18 to 24 months at a moderate size, with an uplift clause carrying the routine movement in between. Don't do it opportunistically because the quarter is short.

If the answer to "are we underpriced" turns out to be no, the more productive project is usually discounting discipline rather than list price, because most SaaS companies give away more margin at the negotiating table than a price increase would recover. That's covered in [SaaS discounting strategy](/guides/saas-discounting-strategy/). Outcomes from real repricing events are collected in [SaaS price increase outcomes](/research/saas-price-increase-outcomes/), and the wider framework sits in [SaaS pricing strategy](/saas-pricing/).

**Pre-launch checklist**

## What to do next

Pull the last 100 closed-won deals and calculate the average discount. If it's under 8 percent and your win rate is over 40 percent, run the breakeven model this week and take it to your next leadership meeting with a number rather than a feeling.

Then, regardless of whether you raise prices this year, get an annual uplift clause into every new contract from the next one you sign. It's a one-line change that removes this entire project from your future, and it costs nothing to add while the customer is already saying yes. If the increase is going to fund demand generation rather than margin, say so internally, because the case is easier to make when the money has a destination, which is the argument in [SaaS demand generation](/saas-demand-generation/).

## Frequently asked questions

### How much should a SaaS company raise prices by?

Most successful increases sit between 7 and 20 percent. Under 7 percent the operational cost of the change rarely justifies it. Above 20 percent you trigger active evaluation of alternatives rather than passive acceptance, especially in SMB. If you believe you are underpriced by more than 30 percent, do it in two steps a year apart rather than one jump.

### How much churn should I expect from a price increase?

For a 10 percent increase with reasonable notice and grandfathering, plan for 2 to 5 percent of affected revenue to leave in SMB and 1 to 2 percent in enterprise, where switching costs are higher. The breakeven point for a 10 percent increase is around 9 percent revenue churn, so almost any competently executed increase is net positive.

### Should existing customers be grandfathered?

Yes, but for a defined window, not forever. Twelve months or until the next renewal is the standard that works. Permanent grandfathering creates a price book per cohort, and after four increases your billing system, your sales team and your support docs are all describing different products. Time-boxed grandfathering gets you most of the goodwill at a fraction of the cost.

### How much notice do I need to give before raising prices?

Sixty to ninety days for annual contracts, thirty days minimum for monthly. Check the master services agreement first, because many enterprise contracts specify a notice period and some cap the permitted increase or restrict changes to renewal. Breaching that clause turns a pricing exercise into a legal one.

### What is an annual price uplift clause?

A contract term that permits an automatic annual increase, typically 3 to 5 percent or the change in a published inflation index, applied at each renewal. It removes the need for a negotiation every year and makes the increase a contractual fact rather than a request. Enterprise buyers accept them routinely when the cap is stated and reasonable.

### When is the wrong time to raise prices?

During a visible outage or reliability problem, in the same quarter you removed features, when win rates are below 25 percent, or while a well-funded competitor is actively undercutting you in your core segment. Raising prices into any of those signals to customers that you are extracting rather than investing, and the churn will exceed the model.