CRO / Pricing
How to Raise Prices Without Losing Customers
// the fear of raising prices usually costs more than the raise would
By Fracto Solutions
October 28, 2026
The short answer
Raise deliberately: increase in steps, tie the rise to real or visible value, communicate honestly, and test the effect on total gross profit rather than units. Do not assume demand is insensitive, because measured retail elasticities say otherwise. And check your notice obligations, which are legally mandatory for subscriptions in several US states.
How risky is a price increase, really?
Less risky than it feels, and more risky than the usual advice admits. The honest answer is that it depends heavily on your category, and the published evidence does not support a blanket “customers are less price-sensitive than you think”.
Here is what is actually measured. Meta-analyses of retail price elasticity put the mean at -1.76 (Tellis, 1988, 367 estimates across 220-plus brands) and -2.62 (Bijmolt, van Heerde and Pieters, 2005, 1,851 elasticities across 81 studies), with elasticities growing more negative over four decades and online channels typically more elastic than offline. An elasticity below -1 means demand is elastic: a price rise loses proportionally more volume than it gains in price.
That does not make price rises a bad idea. It means the gain is not automatic and the no-volume-loss case has to be found rather than assumed. The classic leverage figure, from Marn and Rosiello in Harvard Business Review in 1992, is that a 1% price improvement raises operating profit 11.1% for the average company against 3.3% for volume, and that figure explicitly assumes no volume loss. Put those two findings side by side and the job is clear: find the products and segments where demand is least elastic, raise there, and measure. The strategy context is in the pricing pillar.
Elasticity also varies enormously by category. Commodity and comparison-shopped goods, high-stockpiling groceries, and anything with identical competing listings are highly elastic. Differentiated, branded, novel or urgently-needed products are much less so. Your own numbers beat any benchmark here.
How should I implement a rise?
In steps, tied to value, and clearly communicated. Increase gradually rather than in one jump, ideally alongside an improvement, and tell customers honestly. For loyal or subscription customers, forewarn them or grandfather existing pricing for a period.
Do
Avoid
Raise in sensible steps
One large, abrupt jump
Tie it to added or real value
A rise with no reason given
Communicate honestly, in advance
Silent changes customers discover
Forewarn or grandfather loyal buyers
Punishing your best customers first
Check your notice obligations first
Assuming notice is only courtesy
We should be clear that the phasing advice is our practice from client work. We found no published study comparing phased against single increases on retention.
Is advance notice legally required?
For subscriptions in several US states, yes, with specific windows. Treating notice as merely courteous is out of date. General information, not legal advice.
California. The amended Automatic Renewal Law (AB 2863), effective 1 July 2025, requires clear and conspicuous notice no less than 7 and no more than 30 days before a changed fee takes effect, in a retainable form, including cancellation instructions.
Arkansas, from 3 August 2025: 7 to 30 business days before an increase.
New York, from 5 November 2025: affirmative consent to the increase, or a 14-day post-charge cancellation window with a pro rata refund.
Colorado (6 August 2025) and Massachusetts (2 September 2025) have their own versions, and Maryland’s takes effect 1 June 2026.
One correction worth making, because a lot of 2024 and early-2025 guidance has it wrong: the FTC’s federal “click to cancel” rule was vacated in its entirety by the Eighth Circuit on 8 July 2025, days before it was due to take effect, on procedural grounds. It is not in force. ROSCA and FTC Act §5 still apply. If you read advice that treats click-to-cancel as live law, it predates that ruling.
In the EU there is no single notice window for a mid-contract increase; it is governed by the unfair-contract-terms regime, which requires a variation clause to be transparent and usually requires the consumer to be able to exit.
How do I communicate a price increase?
Honestly and briefly, with a reason. If costs rose or you have added value, say so. Give notice where you can, and where the law requires it, give the notice the law specifies.
People accept price rises far better when they understand them and are not ambushed. A short, honest note usually lands fine. For subscription and repeat customers, advance notice is both courteous and, in the states above, required, and it protects the retention you have built. Include cancellation instructions where California’s rule applies, since the notice is non-compliant without them.
How do I de-risk it with testing?
Test before you commit everywhere, and watch total gross profit rather than units. Given elasticities around -1.8 to -2.6, a rise that holds profit while losing some volume is a win and a rise that loses both is easy to spot early.
Trial the increase on a subset of products or a share of traffic, measure gross profit rather than order count, and expand once the numbers confirm it. That is standard testing discipline applied to price.
Two constraints most guides omit. New York’s Algorithmic Pricing Disclosure Act took effect on 10 November 2025 and requires the disclosure “THIS PRICE WAS SET BY AN ALGORITHM USING YOUR PERSONAL DATA” wherever price is set algorithmically using personal data such as location, income or shopping history, at $1,000 per violation; a National Retail Federation challenge to it failed. In the EU, the Omnibus Directive has required traders to disclose personalised pricing based on automated decision-making since 28 May 2022, with penalties up to 4% of turnover. A randomized holdout on traffic is generally fine; splitting price by customer attribute is the thing that triggers these. Also check that your Google Shopping feed price still matches your landing page, which price tests routinely break.
We help you raise prices on evidence, not nerve.
We model the margin math, phase the increase, tie it to value, help you communicate it well (and protect loyal customers), and test the impact on total profit, so a price rise lifts margin without the volume loss founders fear.
It’s the same measurement-first approach behind our conversion optimization and analytics work, where we rebuild the buying path and measure the result .
// we would rather show up in your reporting than in your inbox
Frequently asked questions
Will raising prices cost me customers?
Some, and possibly more than you expect. Meta-analyses put mean retail price elasticity between -1.76 and -2.62, meaning demand is usually elastic, and online channels tend to be more price-sensitive than offline. The leverage on profit is real but it assumes you hold volume, so find the least elastic products and segments and measure the result.
How much should I raise prices at once?
In sensible steps rather than one large jump, ideally alongside added or clearly communicated value. We found no published study comparing phased against single increases, so treat this as our practice from client work rather than a measured optimum, and test it.
Do I have to give notice before a subscription price increase?
In several US states, yes. California’s amended Automatic Renewal Law, effective 1 July 2025, requires 7 to 30 days’ notice in a retainable form with cancellation instructions, and Arkansas, New York, Colorado and Massachusetts have their own rules. Note the FTC’s federal click-to-cancel rule was vacated on 8 July 2025 and is not in force.
How do I know a price rise is working?
Test on a subset of products or traffic and measure total gross profit, not units. A rise can reduce volume and still increase profit, which is the outcome you are looking for. Expand once the data confirms it.
Are there rules about A/B testing prices?
Yes, if the split uses personal data. New York’s Algorithmic Pricing Disclosure Act, effective 10 November 2025, requires a specific disclosure where price is set algorithmically using personal data, at $1,000 per violation, and the EU has required personalised-pricing disclosure since May 2022. A randomized holdout on traffic is generally fine.
Sources
Price elasticity: Tellis, “The Price Elasticity of Selective Demand”, Journal of Marketing Research 25(4), 1988: mean -1.76 across 367 estimates covering more than 220 brands and markets. Bijmolt, van Heerde and Pieters, Journal of Marketing Research 42(2), 2005: mean -2.62 across 1,851 elasticities in 81 studies, growing more negative over four decades. Online channels tend to be more elastic than offline. Retail demand is therefore usually elastic, which is the opposite of the common claim that shoppers are less price-sensitive than founders fear. Elasticity varies enormously by category, so measure your own. https://doi.org/10.1509/jmkr.42.2.141.62296
Pricing leverage on profit: Marn and Rosiello, “Managing Price, Gaining Profit”, Harvard Business Review 70(5), September to October 1992. For the average company in a Compustat aggregate of 2,463 companies, a 1% improvement raises operating profit by 11.1% on price against 3.3% on volume. The 11.1% figure explicitly assumes no volume loss, and the analysis is 34 years old and based on early-1990s large-cap US industrial cost structures. https://pages.charlotte.edu/wp-content/uploads/sites/868/2014/12/ManagingPrice.pdf
Price-increase notice for subscriptions: California’s amended Automatic Renewal Law (AB 2863, effective 1 July 2025) requires clear and conspicuous notice no less than 7 and no more than 30 days before a changed fee takes effect, in a retainable form and including cancellation instructions. Arkansas (7 to 30 business days, 3 August 2025), New York (affirmative consent or a 14-day post-charge cancellation window with pro rata refund, 5 November 2025), Colorado and Massachusetts have their own versions. Note that the FTC’s federal “click to cancel” rule was vacated in its entirety by the Eighth Circuit on 8 July 2025 and is not in force; ROSCA still applies. Any guidance citing click-to-cancel as live law is out of date.
Stated as Fracto’s position, not research: that increases should be phased rather than made in one jump, tied to visible value, and communicated in advance. We found no published study comparing phased against single increases on retention. The elasticity evidence and the notice requirements above are sourced; the sequencing advice is ours, from client work.