Good conversion rate: what it is, and for which type of site
There is no such thing as a good conversion rate in absolute terms. A rate can only be judged against three things: the action you count (quote request, sale), what that action is worth in gross margin, and the traffic volume below which the number no longer means anything. Published benchmarks by industry mix definitions that cannot be compared, and mostly serve to sell tools.
- A conversion rate is comparable only to itself, over time. Compared with a vanity industry average, it says nothing, because nobody counts the same action.
- Below a certain volume, the rate moves as much by chance as by your work. Detecting a move from 2% to 2.2% takes more than a hundred thousand visitors per variant.
- The right question is not "what is a good rate", it is "how much is one more percentage point worth, in monthly revenue". Our experts calculate it in ten minutes.
- The gain truly within reach is documented: the Baymard Institute puts at 35.26% the increase obtained by fixing checkout problems that are already known, with no prior testing.
On this page
Conversion rate
The conversion rate is the proportion of visits that end in the action you have decided to count, over a given period. The number of actions, divided by the number of visits. Its difficulty is not the calculation, it is the definition of the action: a completed form, a qualified lead and a sale collected produce three different rates on the same site, and none of the three is wrong.
Why industry benchmarks mislead more than they help
Search for "average conversion rate" and you will find tables by industry, often between 1% and 4%. They are published by software vendors, from their own customers' accounts. They aggregate things that cannot be compared.
Three reasons, in order of importance.
The action counted is not the same. A site that counts newsletter sign-ups shows a rate six times higher than one that counts only closed sales. The second sells better. The first has a nicer vanity number.
The traffic is not the same. A site where half the visits come from brand searches mechanically converts better than a site fed by discovery advertising. That is not a quality of the site, it is a traffic mix.
The denominator is not the same. Sessions, users, unique visits, visitors who reached the funnel: four common bases of calculation, and the gap between the widest and the narrowest is often more than double.
A conversion rate compared with an industry average produces only one useful decision: feeling reassured or feeling worried. Neither one sells anything. The only benchmark that calls for action is your own rate, measured on the same definition of profitability, three months ago.
The threshold below which the number lies
This is the part the guides leave out, and it is the most useful. A conversion rate is a proportion measured on a sample: it therefore carries a margin of error, and that margin tightens slowly.
The order of magnitude is easy to grasp. Detecting a move from 2% to 4% takes little traffic, because the financial effect is huge. Detecting a move from 2% to 2.2% takes more than a hundred thousand visitors per variant, because the effect is the size of the noise. Yet the real gains of an optimization project look far more like the second case than the first.
The practical consequence: below a few hundred visits a month on the page concerned, your rate goes up and down from one month to the next without anything having changed. Reading it as a signal means making decisions on chance. And launching an A/B test at that volume means buying a conclusion the data does not support.
If your monthly conversion rate varies by more than half from one month to the next with no change on the site, you do not have a conversion problem: you do not yet have enough volume to measure your sales.
That does not mean there is nothing to do at low volume. Fixes that require no statistical proof remain within reach: a form that fails on mobile, a price discovered on the last screen, a button that is invisible on a phone. Those can be seen with the naked eye, and our strategists settle them in a day.
How to calculate yours, and above all what it is worth in margin
Calculating the rate takes one minute. The calculation that changes a business decision takes ten.
| Step | The calculation | Example |
|---|---|---|
| 1 | Set the action that counts, only one | A quote requested |
| 2 | Actions divided by visits, same period | 60 ÷ 3,000 = 2% |
| 3 | Value of a sale, multiplied by the close rate (CRM) | $8,000 × 25% = $2,000 |
| 4 | What one more percentage point is worth in margin | 30 actions × $2,000 = $60,000 per month |
The fourth line is the one that decides. Until it is written down, no optimization budget can be justified, and no cost per lead (CPA) is too expensive or low enough. Once it is written, the comparison becomes simple: what the project costs, against what the percentage point brings in.
Note that the rate itself appears only at step 2, and that it only serves to feed step 4. It is a measuring instrument, not a vanity target.
What is truly achievable
The Baymard Institute has been testing the checkout funnels of large companies for more than ten years. Limiting itself to the usability problems it has documented as solvable, it puts at 35.26% the conversion rate increase within reach of a large online commerce site, with no prior testing. These are not hypotheses: they are known fixes, on sites already considered well made.
The same institute measures an average cart abandonment rate of 70.22%, calculated across 50 studies, remarkably stable for more than a decade. The repairable share of those departures comes from friction, not from a lack of interest: fees discovered late, a mandatory account, a form that is too long.
Two honest caveats. These figures come from large American and European sites, and an order of magnitude is not a promise that applies as is. And a 35% increase on a 2% rate gives 2.7%, not 37%: the percentage applies to the rate, never to the revenue.
The three questions to settle before aiming for a rate
A rate target set without these answers produces either a chase after vanity numbers, or a project dropped after three months.
- Which single action are we counting, and since when have we measured it the same way?
- How much is one more percentage point worth, in real monthly revenue?
- Does the page concerned receive enough visits for a gap to be readable in less than a quarter?
Turning a site into an asset that sells rather than a showcase is at the heart of the Improve your site's conversion goal, organized around your revenue objective.
Want the project rather than the theory? See conversion rate optimization and the traffic threshold below which our experts will tell you to do nothing.
Frequently asked questions about conversion rate
What is a good conversion rate?
One that improves on your own rate from the previous quarter, measured on the same action and the same basis of calculation. Industry averages aggregate definitions that cannot be compared: a site that counts newsletter sign-ups and a site that counts signed quotes have no reason to show the same number, and it is often the second that generates the most profit.
How do you calculate a conversion rate?
The number of actions counted, divided by the number of visits over the same period. What changes everything is the definition of the action and the basis of the denominator: sessions, users or visitors who entered the funnel give results that can vary by a factor of two on the same data.
From how many visits does the rate become reliable?
It depends on the gap you are trying to see. A doubling can be detected with little traffic. A relative gain of ten percent, which is the order of magnitude of a real expert project, takes tens of thousands of visits per variant. In practice, below a few hundred visits a month on the page concerned, the number varies mostly by chance.
Why is my conversion rate falling while sales are rising?
Almost always because the traffic mix has changed. Adding a discovery campaign brings in visitors earlier in their decision: the denominator grows faster than the numerator, the rate falls and revenue rises. That is why our strategists judge a channel on its contribution to margin, not on its rate.
Do you need A/B tests to improve your conversion?
Not first. Usability problems that are already documented can be fixed without a test, and that is where the bulk of the financial gain lies. An A/B test serves to settle between two defensible options when volume allows it; it never serves to discover what is wrong, and it never says why a version loses sales.
- Baymard Institute, Cart abandonment rate statistics, average of 70.22% calculated across 50 studies, accessed August 2026.
- Baymard Institute, research on checkout usability, 35.26% increase measured on large American and European sites, accessed August 2026.
- Statistical power orders of magnitude: standard sample size calculation for a comparison of two proportions, at 95% confidence and 80% power.
Gabriel almost always takes your first call and carries out your audit. He builds the strategy starting from your growth goal: where to put your budget, which market to test and how to connect each lead to a real sale in your CRM. He mainly leads engagements for three goals: Optimize the profitability of your digital campaigns, Develop a new market, and Generate demand and growth. With Geneviève, he also works on organic search (SEO), AI visibility (GEO) and conversion rate optimization (CRO). The sales a Google Ads or Meta Ads campaign brings in depend on the page that receives the click. He writes mainly about marketing strategy, paid advertising and measurement.
About Falia →
SEO
Paid advertising
Conversion
AI visibility
Web design