Freemium conversion rate: what 3-5% really means, and when a card-required trial beats it

A good freemium conversion rate is 3-5% of free signups becoming paying customers within six months, and a great one is 8-12%, according to ChartMogul's 2026 SaaS Conversion Report. The 8% median that gets quoted for all products is a blended number, and a quarter of freemium products sit below 2.5%. By the end of this post you will know which band applies to your model, how to measure your own rate so it compares fairly, and when a card-required trial or a reverse trial is worth testing instead.

Every figure here comes from two benchmark surveys and from vendor documentation, as checked on November 5, 2026. Both surveys are self-reported or survey-based, so treat the bands as a map, not a scoreboard.

What is a good freemium conversion rate? 3-5% is the middle, not the floor

ChartMogul's report, written by Kyle Poyar with ProductLed as a partner, surveyed 200 B2B software products in January 2026. Respondents were asked to "please estimate to your best ability", so these are estimates, not billing data. The report defines free-to-paid conversion as "the percentage of leads or free signups that convert to become a paying customer within six months."

In that report, "good" means the 50th percentile and "great" means the 75th. Here are its bands by acquisition model:

The practical reading: a freemium product converting 4% is ordinary, not failing. A freemium product at 1% is in the bottom quarter, and a no-card trial at 4% sits at the low end of its median band. Comparing either of them with a card-required trial's 30% tells you nothing, because the denominator is a different kind of person.

The 8% free to paid conversion rate median hides a wide spread

ChartMogul's headline figure is that "the median free-to-paid conversion rate across all products is 8%." The next sentence in the report is the one worth remembering: "very few products actually have an 8% conversion rate." The distribution is lumpy, not a bell curve around 8%.

For freemium products, the report says 25% convert below 2.5%, 29% convert between 2.5% and 7.5%, and another 25% convert between 10% and 15%. For free trial products, 20% sit below 2.5%, 30% sit between 2.5% and 7.5%, and 23% convert above 25%. That top group lines up with card-required trials: only 20% of trial products ask for a card, and their median band is 25-35%.

So if you run freemium and you are at 2%, the honest comparison is not "I am a quarter of the median." It is "I am in the bottom quarter of freemium products in this sample." That framing points at different fixes: activation and the paywall, not a pricing page tweak.

The report also notes that "AI-native and AI plus SaaS hybrid products see slightly higher conversion compared to traditional SaaS." The sample is mostly SaaS (55%), with 30% hybrids and only 9% AI-native, so we would not plan an AI product's forecast around a separate AI number.

SaaS conversion rate benchmarks disagree, and the definitions explain why

The other widely cited benchmark is the 2023 survey by Kyle Poyar and Lenny Rachitsky, run with OpenView, Growth Unhinged and Pendo across "1,000+ products" and published in Lenny's Newsletter. The full post is for paid subscribers; the figures below are from the part that is public.

The freemium bands agree on 3-5% as the median. The free trial bands do not: Lenny's 2023 data puts a good trial at 8-12%, while ChartMogul's 2026 no-card trial band is 4-6%. The public part of the Lenny's post does not say whether its trial figures separate card-required from no-card trials, so some of that gap may be a mix of the two. The samples, years and units also differ.

The unit matters more than it looks. Lenny's data is defined "at an account level", where "an account could be made up of a single user or a collection of users." If you count every invited teammate as a signup, your rate will look worse than an account-level benchmark for the same business.

How to calculate free to paid conversion so it compares fairly

Both surveys use a six-month window, and the Lenny's post says "the best practice is to measure free-to-paid conversion on a cohort basis." A cohort is everyone who signed up in the same period, followed forward. Here is the calculation step by step:

1. Pick the unit and keep it: accounts or workspaces if you sell to teams, individual users if you sell to individuals.
2. Group signups into monthly cohorts by signup date.
3. For each cohort, count the accounts that made a first payment within six months of signing up.
4. Divide paying accounts by all accounts in that cohort, including the ones that never activated.
5. Report only cohorts that are at least six months old. Younger cohorts are still converting and will read low.

Take illustrative numbers, not anyone's real data: a March cohort of 1,000 accounts with 30 paying by September has a 3% rate, inside the freemium "good" band. The same month measured as "new payers in March divided by new signups in March" mixes payers who signed up in January with signups who have had a week. That ratio drifts with your growth rate, not your conversion.

The denominator is where most self-reported numbers quietly diverge. Some teams divide by activated users only, which inflates the rate; some include spam and duplicate signups, which deflates it. Whatever you choose, write it down and compare like with like. If you want an activation-adjusted rate as well, report it as a second number, not as your conversion rate.

Watch your tooling defaults too. In Mixpanel funnels, "by default, all customers have 7 days to complete a funnel from the timestamp they perform the Step 1 event," and the maximum window is 366 days. A signup-to-payment funnel left on the 7-day default measures something much narrower than the six-month benchmarks. Mixpanel, Amplitude and PostHog are among the 640 tools in Upfynd's Analytics category as of November 2026; whichever you use, set the window before you compare.

Freemium vs free trial: which converts better per visitor

The conversion rate alone flatters card-required trials, because they shrink the signup pool. ChartMogul's report also gives a per-visitor view, which is the fairer comparison when you are choosing a model. Per 1,000 website visitors:

On these figures, a card-required trial produces roughly twice the paying customers per visitor of freemium, from well under half the signups. A no-card trial produces fewer payers per visitor than freemium in this data, because it halves signups without a large enough lift in conversion.

The strongest case for freemium is what those extra 55 free users do. They can invite teammates, create content others see, or become sales leads; the Lenny's post reports that 24% of freemium companies have sales reach out directly to more than half of sign-ups, against 44% of free-trial companies. If free users spread your product, the per-visitor table undercounts freemium. If they mostly sit idle, it does not.

Two cautions before you switch. These are survey estimates from different products, not one product tested both ways, so a team that already had high-intent buyers may have chosen a card trial for that reason. And the table says nothing about churn after the first payment.

Does requiring a credit card increase trial conversion?

In ChartMogul's sample, yes, by a lot: "free trials that require a credit card see 30% free-to-paid conversion – more than 5x ones that don't require one." Only 20% of free trial products require a card upfront; the other 80% do not. The most common trial length is 14 days (62% of products), followed by 7 days and 30 days at 14% each.

Skip a card-required trial if your product needs weeks of setup before anyone sees value, or if your buyers cannot put a company card in on day one. You will lose most signups and the remainder will not have time to be convinced. It fits best when value shows up inside the trial window and the buyer holds the card.

Billing tools support both shapes. Stripe's documentation on trials says a 0 USD trial offer leaves the subscription in trialling status, which it calls "ideal for no-card-required" flows, and lets you choose what happens when a trial ends with no payment method: create an invoice (the default), cancel, or pause. Its trial_will_end event defaults to 3 days before the trial ends, and Stripe-hosted reminder emails go out 7 days before. If you test a card-required trial on Stripe, send that reminder so the first charge is never a surprise.

Reverse trial: the middle option, with a thin sample

A reverse trial gives new users "temporary access to premium features on a free plan", in ChartMogul's glossary wording, then drops them to the free tier instead of locking them out. You keep freemium's signup volume and its free user base while showing everyone the paid features first.

ChartMogul's bands put reverse trials at 4-6% good and 8-12% great, slightly above regular freemium's 3-5% median. The report is careful about this: the reverse trial sample was small and the difference is not statistically significant. Treat it as a reasonable experiment when you are below the freemium band, not as a proven upgrade.

What to do if you are below your band

Start by recomputing your rate as a six-month cohort with a fixed unit, then place it in the band for your actual model. If freemium cohorts sit under 2.5%, look first at what share of signups reach the moment the product is useful, and at whether the free plan already covers what most people need.

If activation is fine and the free plan is too generous, a reverse trial is the gentler test. If your buyers hold a card and see value inside 14 days, a card-required trial is the bolder one; measure paying customers per 1,000 visitors before and after, not the conversion rate alone. ChartMogul, which published the report, is also listed on Upfynd alongside other revenue analytics tools.

When you have a plan you want people to try, you can list your product on Upfynd for free and see how it ranks on the launch boards.