Free trial vs freemium statistics at a glance
| Category | Statistic | Source |
|---|---|---|
| Filings | Median paid share across Spotify, Slack, Duolingo, Life360 and Dropbox: 9.0%; range 2.6% to 38.6% | Computed from company filings, 2019 to 2026 |
| Adoption | 57% of B2B software products lead with a free trial, 26% with freemium, 7% with a reverse trial | ChartMogul, 2026 |
| Freemium | A good freemium conversion rate is 3% to 5%; a great one is 8% to 12% | ChartMogul, 2026 |
| Freemium | Freemium-to-paid conversion averages 2.6% from organic traffic across agency SaaS clients | First Page Sage, 2025 |
| Opt-in trial | Trials without a card convert 18.2% of trial users to paid (organic traffic) | First Page Sage, 2025 |
| Opt-out trial | Card-required trials convert 48.8% of trial users (organic), 2.7x the opt-in rate | First Page Sage, 2025 (computed) |
| Per visitor | Per 1,000 visitors: 15.5 paying customers from an opt-in trial, 12.2 from opt-out, 3.5 from freemium | Computed from First Page Sage, 2025 |
| Per visitor | Per 1,000 visitors: 5 paying customers from freemium, 3.6 from a free trial, 10.5 from a card-required trial | ChartMogul, 2026 |
| Apps | Hard paywall apps convert 10.7% of downloads to paid by day 35; freemium apps convert 2.1% | RevenueCat, 2026 |
| Retention | One-year retention of yearly subscribers: 27% for hard paywall apps, 28% for freemium apps | RevenueCat, 2026 |
| Trial length | 62% of B2B free trials last 14 days; 14% last 7 days and 14% last 30 days | ChartMogul, 2026 |
| Trial length | App trials of 17 to 32 days convert 42.5% of trial starts; trials of 4 days or less convert 25.5% | RevenueCat, 2026 |
| Experiment | A 7-day trial lifted total paid conversion 20.9% over a 3-day trial across 680,588 users | Frontiers in Psychology, 2025 |
| Spotify | Ad-supported listeners are 63.6% of Spotify's MAU and 9.3% of its revenue | Computed from Spotify Q2 2026 results |
How often do freemium users convert to paid?
Freemium products convert between 2% and 5% of free users to paid in most benchmark data, with ChartMogul calling 3% to 5% good and 8% to 12% great in its 2026 study of 200 B2B software products. The agency data runs at the low end of that range and the survey data at the high end.
| Statistic | Source |
|---|---|
| Among 200 B2B software products surveyed in January 2026, 57% lead with a free trial, 26% with freemium, 7% with a reverse trial, 7% with an interactive demo and 4% with a paid trial. | ChartMogul with ProductLed, February 2026 |
| The median free-to-paid conversion rate across all self-serve products in the study is 8%. A good freemium rate is 3% to 5% and a great one 8% to 12%. | ChartMogul, 2026 |
| Ungated freemium, where users try the product before creating an account, brings 70 signups and 5.6 paying customers per 1,000 visitors. ChartMogul rates 7% to 9% free-to-paid as good for ungated products. | ChartMogul, 2026 |
| The top 20% of self-serve products convert about 10 times better than the bottom 20%. | ChartMogul, 2026 |
| 13.3% of organic visitors sign up for a freemium plan and 2.6% of those free users convert to paid. From paid traffic the figures are 15.9% and 2.8%. The data covers 86 SaaS clients from Q1 2022 to Q3 2025. | First Page Sage, September 2025 |
| Freemium-to-paid conversion by industry ranges from 5.6% in LegalTech and 5.1% in ERP down to 2.1% in EdTech, across 80+ freemium SaaS clients from 2022 to 2026. | First Page Sage, September 2026 |
| Kyle Poyar and Lenny Rachitsky's survey of 1,000+ products put good self-serve freemium conversion at 3% to 5% and good sales-assisted freemium at 5% to 7%. | Lenny's Newsletter, August 2023 |
Two numbers on this page share a value by coincidence: First Page Sage's 2.6% freemium rate and Dropbox's 2.6% paid share further down. They measure different things, one a cohort of free signups and the other a lifetime stock of registrations.
Freemium conversion benchmarks cluster so tightly around 2% to 5% that the number itself tells you almost nothing about your product. The spread that matters is ChartMogul's 10x between the top and bottom fifth. If you're at 2%, ask whether your free plan gives away the thing people would pay for.
A low freemium rate only looks bad next to a free trial rate, so that comparison comes next.
Does a free trial convert better than freemium?
A free trial converts a larger share of its signups than freemium in every dataset here, but freemium brings in more signups, so the free trial vs freemium gap per visitor is much smaller than per signup. In First Page Sage's 2025 data, an opt-in trial converts 18.2% of trial users and freemium converts 2.6% of free users.
| Statistic | Source |
|---|---|
| Per 1,000 website visitors, freemium products get 90 signups and 5 paying customers; free trial products get 45 signups and 3.6 paying customers. | ChartMogul, 2026 |
| A good conversion rate for a free trial without a card is 4% to 6% and a great one 10% to 15%, against 3% to 5% and 8% to 12% for freemium. | ChartMogul, 2026 |
| In First Page Sage's newer freemium report, traditional freemium converts 12.8% of visitors to free users and 3.4% of those to paid; land and expand freemium converts 14.1% and 3.7%. Opt-in trials convert 7.6% of visitors and 17.3% of trial users. | First Page Sage, September 2026 |
| Subscription apps with a hard paywall convert a median 10.7% of downloads to paid within 35 days; freemium apps convert 2.1%. The sample covers 115,000+ apps and $16 billion in revenue. | RevenueCat State of Subscription Apps, March 2026 |
| Median revenue per install is $2.32 for hard paywall apps against $0.27 for freemium apps at day 14, and $3.09 against $0.38 at day 60. | RevenueCat, 2026 |
| One year after purchase, hard paywall apps keep 27% of yearly subscribers and freemium apps keep 28%, a 1-point gap RevenueCat calls noise. | RevenueCat, 2026 |
| Good free trial conversion in Poyar and Rachitsky's survey of 1,000+ products was 8% to 12%, against 3% to 5% for self-serve freemium. | Lenny's Newsletter, August 2023 |
The two B2B datasets disagree on the winner per visitor. ChartMogul's survey has freemium ahead, 5 paying customers per 1,000 visitors against 3.6 for a cardless trial. Computed from First Page Sage's agency data, the opt-in trial wins, 15.5 against 3.5. The populations differ (self-reported survey versus agency clients), so neither result cancels the other.
The app data makes the same point in revenue terms. RevenueCat's hard paywall apps earn about eight times more per install by day 60, yet keep subscribers no longer than freemium apps. So the revenue gap comes from how many people pay; how long they stay is the same. Freemium apps are betting that a larger free base will pay later or earn money through ads; the 60-day window can't show whether that bet pays off.
I'd stop asking which model converts better and ask which one you can afford to serve. RevenueCat's retention figure is the one I keep coming back to: once people pay, freemium subscribers stay as long as paywall subscribers. So the whole difference sits in the first 35 days, and it comes down to who you let in.
Public companies don't publish conversion funnels, but their filings show the end state, which is the next section.
Should a free trial require a credit card?
A card-required (opt-out) free trial converts roughly three times more trial users to paid than a cardless (opt-in) trial, but it also cuts signups sharply. In First Page Sage's 2025 data, opt-out trials convert 48.8% of organic trial users against 18.2% for opt-in trials.
| Statistic | Source |
|---|---|
| 2.5% of organic visitors start an opt-out trial against 8.5% for an opt-in trial. From paid traffic the rates are 2.2% and 7.1%. | First Page Sage, September 2025 |
| Opt-out trials convert 48.8% of organic and 51% of paid trial users to paid; opt-in trials convert 18.2% and 17.4%. | First Page Sage, 2025 |
| Trial-to-paid conversion by industry in the same data ranges from 29.0% in CRM to 18.6% in enterprise software, which also has the lowest visitor-to-trial rate at 5.5%. | First Page Sage, 2025 |
| 20% of B2B free trial products require a card upfront. A good conversion rate for a card-required trial is 25% to 35% and a great one 50% to 60%. | ChartMogul, 2026 |
| Per 1,000 visitors, a card-required trial produces 35 signups and 10.5 paying customers, against 45 signups and 3.6 paying customers for a cardless trial. | ChartMogul, 2026 |
| Across 803 subscription sites from January to August 2022, 87.4% of trials required billing information and converted at 39.0%; trials without billing information converted at 12.0%. | Recurly State of Subscriptions, 2022 |
The Recurly figures are four years old and marked DATED; I keep them because they're the only large consumer dataset that splits the card question within one study. The opt-in versus opt-out figures quoted across the web without a source (60% versus 15% and similar) could not be traced to any named dataset and are left out.
The two B2B datasets also disagree on how much the card costs in signups. First Page Sage's opt-out trials attract 71% fewer trial starts than opt-in trials; ChartMogul's card-required trials attract 22% fewer. That difference decides the per-visitor result. With a 22% signup penalty, a card that triples conversion wins easily, which is what ChartMogul's 10.5 against 3.6 shows. With a 71% penalty, the same tripling roughly breaks even, which is what First Page Sage's data shows. Your own signup penalty is the number to measure before you pick a side.
In the two datasets that report trial-to-paid directly, the card multiplies it by 2.7x to 3.3x and removes somewhere between a fifth and seven tenths of your signups. What changes between datasets is which effect wins. My rule: if a sales team works the trial signups, drop the card and feed them volume. If nobody touches the trial, keep the card, because unworked cardless trials mostly expire.
The other lever is how long the trial runs, and here the app data and the B2B data point in different directions.
How long should a free trial be?
Most B2B free trials last 14 days, the length 62% of free trial products use in ChartMogul's 2026 study, but the strongest evidence says longer trials convert more users over time. RevenueCat finds app trials of 17 to 32 days convert 42.5% of trial starts, against 25.5% for trials of four days or less.
| Statistic | Source |
|---|---|
| 62% of B2B free trial products run a 14-day trial, 14% a 7-day trial and 14% a 30-day trial. | ChartMogul, 2026 |
| Subscription app trials of 17 to 32 days convert 42.5% of trial starts to paid, about 70% better than trials of four days or less at 25.5%. | RevenueCat, March 2026 |
| The share of apps offering trials shorter than four days still rose, from 42.1% in 2025 to 46.5% in 2026. | RevenueCat, 2026 |
| In a randomised experiment with 680,588 new users in 190 countries, who signed up in July and August 2022 and were tracked to July 2024, a 7-day trial raised trial adoption 11.1% and delayed conversions 42.4% over a 3-day trial. Immediate conversion showed no significant change. | Zhang and Duan, Frontiers in Psychology, June 2025 |
| Over the two years, the 7-day group's total subscription rate was 0.445% against 0.368% for the 3-day group, a 20.9% lift. | Zhang and Duan, 2025 |
| Across 803 subscription sites in 2022, trials of 7 days or less converted at 40.4% and trials longer than 61 days at 30.6%; the average trial ran 30 days. | Recurly, 2022 |
The Recurly trial length figure is DATED and describes consumer subscriptions, where very long trials often signal a discounting habit rather than a product that needs time. The Frontiers experiment is the only randomised test on this page, and its product was a freemium image editor, so it tests trial length within a freemium product and says nothing about trial against freemium.
For B2B teams the 14-day default looks like habit more than evidence. None of the B2B datasets here compares conversion by trial length, and the two consumer datasets disagree: Recurly's 2022 data favours short trials, RevenueCat's 2026 data long ones. The honest reading is that trial length should match the time a buyer needs to reach the first result, which for a team tool that needs setup and colleagues is rarely 14 days.
The experiment is the number I trust most, because it's the only one where chance, not the company, decided who got the longer trial. Its finding is subtle: the extra days didn't lift conversion during the trial at all; they lifted it weeks later. If you judge trial length on a 30-day dashboard, you'll pick the short trial and never see what you lost.
Length and card settle the trial side. Whether trial users stay longer than buyers who never trialled is the next question.
Do free trials improve retention and lifetime value?
Free trials improve retention in subscription apps: one-year retention on weekly plans that start with a trial runs 43% higher than on weekly plans without one, according to Adapty's 2026 data on $3 billion in revenue. The long-run retention gap between freemium and paywall models, by contrast, is close to zero.
| Statistic | Source |
|---|---|
| 7 in 10 subscription apps offer a trial, a share unchanged for two years, and 90% of trial starts happen on the day of install. The data covers 16,000+ apps and $3 billion in revenue. | Adapty State of In-App Subscriptions, March 2026 |
| Trial users renew 8% to 60% better at first renewal, depending on plan type, and one-year retention on weekly plans with a trial runs 43% higher than on weekly plans without one. | Adapty, 2026 |
| ChartMogul rates a reverse trial, where users get the paid tier free and drop to a free plan afterwards, as good at 4% to 6% conversion and great at 8% to 12%. | ChartMogul, 2026 |
Adapty's numbers come from consumer apps with weekly and annual plans, so treat them as a direction for B2B teams to test. The reverse trial sits between the two models: 7% of B2B products use it in ChartMogul's sample, with benchmarks close to freemium.
I read the retention data as a filter: a trial removes the people who would have churned at the first bill. That's worth a lot on weekly plans and very little on annual ones. For B2B the reverse trial is the compromise I'd test first, because it keeps the freemium funnel and still shows every user the paid product.
The weekly plan result also explains why so many consumer apps keep trials short. A 3-day trial on a weekly plan tests intent quickly and bills early, while the retention benefit shows up at the first renewal. Annual plans work differently, since the first bill is the big one and a trial mostly delays it.
None of these models run on autopilot at scale, which the sales data makes plain.
How much sales help do free and trial users get?
Most self-serve products add human sales contact for larger accounts: 80% of free trial products and 70% of freemium products bring in a person when an enterprise user signs up, according to ChartMogul's 2026 study. Trial companies lean on sales harder than freemium ones.
| Statistic | Source |
|---|---|
| 80% of free trial products and 70% of freemium products add human touchpoints when an enterprise user starts. | ChartMogul, 2026 |
| About 44% of free trial companies have sales reach out to more than half of their signups, against 24% of freemium companies. | Lenny's Newsletter, August 2023 |
| OpenView's final Product Benchmarks edition, run with Pendo across about 1,000 participants, lists outreach to free sign-ups among the levers used by the fastest-growing companies, scored at +28%. The figure is a correlation, not a tested effect. | OpenView 2023 Product Benchmarks, archived |
| 40% of technology buyers used a free trial or free version during their purchase, and 74% of those called it the most influential resource they used. The survey covered 2,164 buyers in March and April 2024. | TrustRadius, June 2024 |
OpenView closed in 2024, so its 2023 report is the final edition; Lenny's survey and the TrustRadius study are also more than 24 months old and marked DATED. For the demo side of this question, and for how much product-led companies spend on sales and marketing, see our product-led vs sales-led statistics.
A free plan with nobody following up is a sampling programme. The 44% versus 24% split explains part of the trial conversion premium: trial companies call their signups. If you run freemium and compare yourself with trial benchmarks, check whether you're comparing products or sales teams.
How we calculated the original numbers
Four figures on this page don't appear in any of the sources. Here is the arithmetic, so you can check it or swap the inputs.
- 9.0%: the median paid share at five public freemium companies. Paid count ÷ user count, same date or quarter, from each company's filing. Spotify 300M ÷ 777M = 38.6% (30 June 2026). Slack 88,000 ÷ 600,000 = 14.7% (31 January 2019, DATED; both are "more than" figures, so the ratio is approximate). Duolingo 12.7M ÷ 140.6M = 9.0% (Q2 2026; 8.5% a year earlier, 10.9M ÷ 128.3M). Life360 3.2M ÷ 102.4M = 3.1% (30 June 2026). Dropbox 18.19M ÷ 700M = 2.6%, an upper bound because registrations exceed 700M. Life360's numerator is households, not people. Sorted, the middle value is Duolingo's 9.0%. The five companies are hand-picked, not a sample of freemium firms, and each defines users its own way, so read the median as a reference point, not a benchmark. Grindr's self-reported 8.4% for 2025 would sit next to it; I leave it out of the median because the company computes it, not us.
- 2.7x to 3.3x: what the card does to trial-to-paid. First Page Sage agency client data (86 SaaS companies, Q1 2022 to Q3 2025, organic traffic, one table): 48.8% ÷ 18.2% = 2.7x. Recurly (803 subscription sites, January to August 2022, one passage): 39.0% ÷ 12.0% = 3.3x. On signups, First Page Sage's opt-out visitor-to-trial rate is 71% lower (1 − 2.5 ÷ 8.5), and ChartMogul's card-required trial gets 22% fewer signups (1 − 35 ÷ 45). Each ratio stays inside one source; the range spans two populations.
- 15.5 vs 12.2 vs 3.5 paying customers per 1,000 visitors. First Page Sage, organic traffic, September 2025. This rests on one agency's client data: 86 SaaS companies, with rates published but no visitor or trial counts, so it's a modelled funnel that assumes the two rates chain for the same visitors. Opt-in trial: 1,000 × 8.5% × 18.2% = 15.5. Opt-out trial: 1,000 × 2.5% × 48.8% = 12.2. Freemium: 1,000 × 13.3% × 2.6% = 3.5. The opt-in trial yields 27% more paying customers than opt-out per visitor (15.5 ÷ 12.2). First Page Sage's September 2026 freemium report gives a near tie between the trial types: 7.6% × 17.3% = 13.1 against 2.7% × 49.1% = 13.3. ChartMogul's own per-visitor figures point the other way (10.5 card-required against 3.6 cardless), from a different population.
- 63.6% of users, 9.3% of revenue: Spotify's free tier. All four inputs come from Spotify's Q2 2026 results (quarter ended 30 June 2026). Ad-supported MAU ÷ total MAU = 494M ÷ 777M = 63.6%. Ad-supported revenue ÷ total revenue = €446M ÷ (€4,331M + €446M) = 9.3%. Per ad-supported user: €446M ÷ 494M ÷ 3 months = about €0.30 a month, against Premium ARPU of €4.89, a gap of roughly 16x. The per-user figure divides quarterly revenue by a quarter-end user count, so treat it as an approximation.
These are simple ratios from public documents. Every company defines its user base in its own way, and I've kept each definition in the table rather than adjusting it.
What the 2026 numbers say
Freemium loses on conversion rate and often wins on paying customers. Every dataset shows trials converting a larger share of signups, and at least one B2B dataset shows freemium producing more paying customers per visitor. Compare models on the visitor, never on the signup, or you're comparing two different denominators.
The filings stretch the benchmark range far past 2% to 5%. Duolingo at 9.0% and Grindr at 8.4% show that a habit-forming free product can convert at triple the SaaS benchmark. Dropbox at 2.6% or less shows the price of a free plan that's good enough forever. The median of five hand-picked filers, 9.0%, is a reference point rather than a benchmark, but it's harder to game than a vendor survey.
Trial length is the cheapest lever here, and the market is moving the wrong way. RevenueCat finds longer trials convert about 70% better, yet the share of apps with very short trials grew. The one randomised study shows the benefit of a longer trial arrives weeks after it ends. For the channel side of the same funnel, our B2B vs B2C marketing statistics cover acquisition costs by business model, and the full statistics hub has the rest of the series.
The card is a staffing decision in disguise. A card-required trial roughly triples trial-to-paid in both datasets that report the rate directly, and costs between a fifth and seven tenths of signups. Teams with sales capacity can turn a cardless flood into customers; teams without it are better off letting the card do the qualifying. Pick the model your team can follow up on, then measure your own signup penalty, because that single number decides whether the card pays.
FAQ
What is a good freemium conversion rate?
3% to 5% is a good freemium conversion rate and 8% to 12% a great one, according to ChartMogul's 2026 study of 200 B2B software products. First Page Sage's agency data puts the average at 2.6% from organic traffic (2025). Public companies can go far higher: Duolingo converts about 9% of monthly users (Q2 2026 filing).
Does a free trial convert better than freemium?
18.2% of cardless trial users convert to paid against 2.6% of freemium users in First Page Sage's 2025 data. Per visitor the gap narrows, because freemium gets more signups: ChartMogul counts 5 paying customers per 1,000 visitors for freemium and 3.6 for a free trial (2026).
What share of Spotify and Duolingo users pay?
9.0% is the median paid share across five hand-picked public freemium companies, computed from their filings. Spotify had 300 million Premium subscribers against 777 million MAU at 30 June 2026 (38.6%), and Duolingo 12.7 million paid subscribers against 140.6 million MAU in Q2 2026 (9.0%). Each company defines users differently, so the ratios aren't directly comparable.
Should a SaaS free trial require a credit card?
48.8% of card-required trial users convert to paid against 18.2% for cardless trials, according to First Page Sage (2025). The card also cuts visitor-to-trial from 8.5% to 2.5%, so per visitor the two options end up close. Require a card when nobody follows up on trial signups; drop it when a sales team does.
How long should a free trial be?
14 days is the most common B2B free trial length, used by 62% of free trial products (ChartMogul, 2026). Longer can convert better: RevenueCat finds app trials of 17 to 32 days convert 42.5% against 25.5% for trials of four days or less (2026), and a randomised study found a 7-day trial beat a 3-day trial by 20.9% (Frontiers in Psychology, 2025).
What is a reverse trial and how well does it convert?
4% to 6% is a good reverse trial conversion rate and 8% to 12% a great one, according to ChartMogul (2026). A reverse trial gives new users the paid tier free for a set period, then drops them to a free plan instead of locking them out. 7% of the 200 B2B products in ChartMogul's study use it.
Sources
- Kyle Poyar (Growth Unhinged) with ChartMogul and ProductLed, The Conversion Report (February 2026)
- First Page Sage, SaaS Free Trial Conversion Rate Benchmarks (September 2025)
- First Page Sage, SaaS Freemium Conversion Rates: 2026 Report (September 2026)
- Spotify Technology S.A., Q2 2026 results, Form 6-K exhibit 99.1 (July 2026)
- Duolingo, Inc., Form 10-Q for the quarter ended 30 June 2026 (August 2026)
- Dropbox, Inc., Form 10-Q for the quarter ended 30 June 2026 (August 2026)
- Life360, Inc., Form 10-Q for the quarter ended 30 June 2026 (August 2026)
- Grindr Inc., Form 10-K for fiscal year 2025 (March 2026)
- Slack Technologies, Form S-1 (April 2019)
- RevenueCat, State of Subscription Apps 2026 (March 2026)
- RevenueCat, The State of Subscription Apps in 10 minutes (March 2026, updated April 2026)
- Adapty, The trial vs. direct question: what $3B in revenue says (March 2026)
- Recurly, State of Subscriptions: The acquisition chapter (2022)
- Zhang and Duan, Longer or shorter? A randomized field experiment on free trial duration in the Freemium model, Frontiers in Psychology (June 2025)
- Kyle Poyar and Lenny Rachitsky, Lenny's Newsletter, What is good free-to-paid conversion (August 2023)
- OpenView Partners with Pendo, 2023 Product Benchmarks, archived copy (2023, final edition)
- TrustRadius, 2024 B2B Buying Disconnect (June 2024)
Methodology. Numbers were collected in September 2026 and checked against the original publisher's page or SEC filing, not against other statistics roundups. Vendor and agency benchmarks are labelled with their sample and period; First Page Sage figures are agency client data, ChartMogul's are a self-reported survey. Filing ratios use the same date or quarter for numerator and denominator and keep each company's own user definition. Opt-in versus opt-out figures that could not be traced to a named dataset were excluded. Last updated 24 September 2026.