Pricing and packaging · Answered by Golden Section from more than 400 B2B software companies observed
Raise price where the customer can already see the value, and do it at renewal or through a contracted escalator rather than mid-term. Churn after an increase usually comes from customers who were not getting the return they were sold, so check adoption and realized ROI account by account before anyone sends a letter. Split the base three ways: core accounts with proven value get the increase with a short review of what they gained; accounts with weak adoption get an adoption plan first; and the poor fits, roughly a fifth of any cohort in our experience, get a price that covers what they cost or an orderly exit. For new contracts, write the increase in: an annual escalation clause, negotiated back toward an outside index such as CPI rather than conceded to zero. Next step: rank your customers by realized value before setting the number.
Raise price where value is proven and renewal is near, defer it where adoption is weak until the customer is getting the return, and accept that some poor-fit customers will leave. Build the next increase into every new contract so it arrives without a negotiation.
| Metric | Value | What it means | Source |
|---|---|---|---|
| Price increase to test | 5–15% | change to price, applied to new business first and renewals secondthe play's range; for most founders we would test about 25% on new business first and aim for about 5% a year on the base. Watch close rates and churn for two quarters before going further | Golden Section playbookPre-Sale Value Levers |
| Annual escalation fallback | an outside index, such as nominal healthcare CPI | the counter offered when a customer pushes back on the contract's maximum escalationthe play anchors escalation high and retreats to an objective standard rather than to zero | Golden Section playbookContract Playbook |
| Poor-fit share of a cohort | about 20% | customers whose cost to serve or expectations do not fit the productraise their price or let them go | Golden Section operating viewGolden Section operating view |
| Software price inflation seen by buyers | 12.7%–16.4% (monthly readings, June–September 2026) | Vertice's SaaS inflation ratederived from $75B+ of software spend Vertice processed in 2026; method not published and Q3 figures preliminary | External benchmarkVertice, SaaS Inflation Rate (2026) |
A price change forces a customer to re-decide. One who reached the result you sold renews and usually never mentions it again; one who did not uses the letter as the reason to leave that was already there. So the increase rarely causes churn. It reveals churn the account was already carrying, which is why the churn identification work and the adoption process come before the price, not after it.
The contract decides how much negotiation you face. The contract playbook recommends an annual escalation clause set deliberately high, so that the customer's attention lands there, and countering with an outside standard such as CPI instead of falling back to nothing. Longer terms help too, because the increase becomes scheduled rather than discretionary. Never concede the increase without asking for something back.
The reason to do this at all is value. Price on the existing base is expansion revenue with almost no acquisition cost, and the pre-sale value levers play counts an increase that has held, with retention intact, as proven pricing power. Customers who never see an increase drift below what new customers pay, and that gap becomes a discount the next buyer notices.
All figures are invented. A company at $5M of ARR has 220 customers and has not raised price in three years. Ranking accounts by adoption and realized ROI puts 150 in the core group, 40 with weak adoption and 30 that are poor fits. Core accounts get 8% at renewal with a one-page review of what they gained; four leave over the year, about the usual rate. The weak-adoption group gets an adoption plan and a one-cycle deferral. The poor fits get 15%, and 9 of 30 leave. Core increases add about $310K, the poor fits who stay add about $45K, and the departures cost about $135K, so ARR rises about $220K and the support queue shrinks.
When one customer is a large share of revenue, treat the increase as a negotiation inside a longer contract term rather than a letter. And where an index-linked escalator already runs every year, a further increase needs new value behind it, such as a module or a service tier.
From the Golden Section mistakes list, each paired with the play that prevents it.
An increase is a pitch for continued value, and customers do not accept more spend without being shown what they gained.
Waiving an increase to keep the peace, without anything in return, teaches the base that price is negotiable.
Some customers should leave, and holding them at an uneconomic price costs more than their revenue.
In the order we would run them. Each is on its own page, most with a free Excel template.
Finds the accounts already at risk so the increase does not land on them blind.
Documents what each core customer gained, which is the case for the increase.
Brings weak-adoption accounts to the promised result before they are asked to pay more.
Writes escalation, term length and the fallback position into the contract so future increases are scheduled.
Tracks the increase in run-rate against close rates and churn so it counts as proven pricing power.
Customer plays Price increases succeed or fail on adoption, account management and contract terms, which the customer plays cover end to end.
Yes, where they are getting the value, and at renewal or through a contracted escalator rather than mid-term. Existing customers who never see an increase fall below what new customers pay, and a buyer of the company reads that gap as weak pricing power. Start with the accounts whose realized ROI you can document.
When it holds, it raises net retention, because added price on the existing base counts as expansion. It hurts only if it pushes out accounts that were already at risk, which is why the base is segmented first and weak-adoption accounts get a plan before an increase. Report the poor-fit departures separately so the board sees what was a choice.
Reviewed by Dougal Cameron, CEO & Co-Founder on 2026-10-22. Golden Section observations are labeled separately from external benchmarks and illustrative arithmetic.