Before you write any billing code, open your analytics and write down how many days a typical new user needs to reach the first moment where the product clearly works for them. That one number decides most of this choice. Use a free trial when that moment arrives within days and the buyer is ready to evaluate now, which describes most B2B SaaS. Use freemium when a permanently free tier creates a network, a habit or a referral loop that feeds paid upgrades.
The wrong pick doesn't only cost conversions. It forces you to redesign onboarding, feature gating and billing later. This guide sits under SaaS Billing and Payments and covers one fork: how people get in before they pay. Every benchmark and Stripe detail below was last checked on 30 September 2026.
What is the difference between a free trial and freemium?
A free trial gives full access for a fixed window and then requires payment to continue, while freemium gives a permanently free, deliberately limited tier and asks for payment only when the user outgrows it.
A trial is a sales motion compressed into the product: it asks for a decision soon. Freemium is a long-term acquisition channel: it is comfortable with a user staying free for a year. A third option, the reverse trial, starts with full access and drops the user to a free tier when the window closes instead of showing a paywall.
| Model | Access at start | What happens at the end | Stripe object needed for free users |
|---|---|---|---|
| Free trial | Full product | Pay or lose access | Yes, a subscription with a trial |
| Freemium | Limited tier | Nothing, until the user upgrades | Not necessarily |
| Reverse trial | Full product | Drop to limited free tier | Usually, until the downgrade |
Consider a hypothetical two-person team selling scheduling software to small clinics. Their buyers compare two or three tools, need a booking calendar plus payments before they can judge anything, and pay per location. A limited free tier would show the calendar but hide the part that convinces them, so a time-boxed trial fits. Now consider a hypothetical note-taking tool where each free user shares pages with colleagues. There, the free account is also distribution, and freemium can make sense. Same billing system, opposite answers, because the value and the audience differ.
What do conversion benchmarks say?
ChartMogul's SaaS Conversion Report, published on 4 February 2026 and covering 200 B2B software products, gives a median free-to-paid conversion rate of 8%. It also publishes "good" and "great" ranges by model.
| Model | "Good" free-to-paid | "Great" free-to-paid |
|---|---|---|
| Freemium | 3%–5% | 8%–12% |
| Free trial, no card | 4%–6% | 10%–15% |
| Free trial, card required | 25%–35% | 50%–60% |
The report adds that, per 1,000 website visitors, card-required trials produced 10.5 paying customers against 5.6 for ungated freemium, despite lower signup rates. Read that carefully. It's a benchmark across 200 products, not a test on yours, and the report doesn't say why the gap exists. My inference is that the model matters less than who you let through the door, but that's an inference.
When should you choose a free trial, freemium or a reverse trial?
Choose by how fast value lands and whether free users have strategic worth: fast value and evaluation-driven buyers favour a trial, while network effects and near-zero cost per free user favour freemium.
When does a free trial fit?
Pick a trial when most of these hold: value lands inside a short window, the buyer is comparing two or three tools, free accounts create no strategic value, and pricing is per seat or per unit so the paid plan is simply the same product billed.
Callidus, the UK aesthetic-clinic SaaS I built on React, Firebase and Stripe Connect, is priced at £15 a month on Starter and £50 on Aesthetics, with a 14-day free trial (figures from the Callidus case study). A clinic can run real appointments and billing through it in days, which fits a time-boxed trial. The case study describes the lifecycle around it: reminder emails three days and one day before expiry, a three-day grace period, then suspension. That follows the trial model, but I have no conversion data to publish, so I'm not claiming the trial converts well.
If you want users to reach their first win before the clock runs out, onboarding has to carry them there. See the B2B SaaS onboarding flow that activates users and, for the card question, how to design a SaaS free trial that converts.
When does freemium fit?
Freemium earns its keep when the free tier is a growth engine. That means free users make the product better for paid users, daily habit deepens switching costs, sharing is built in, and a free account costs you almost nothing to host.
The trap is a free tier that's too good. If it solves the whole problem, nobody upgrades. It should deliver real value and leave a clear ceiling that users hit often. The upgrade prompt carries much of the weight here, so pricing-page craft matters; see the B2B SaaS pricing page that converts.
When does a reverse trial fit?
A reverse trial gives full access with no card, then downgrades to a limited free plan. Users keep their data and a usable product while the premium features switch off, so losing access feels like a step down from something they had.
I couldn't find a primary study that compares reverse trials with the other two models, so I won't claim it beats them. It suits self-serve products with a sharp premium-versus-free contrast. In a sales-led B2B process, a hard trial deadline can be useful pressure on a buying committee, which argues for a plain trial.
How does the choice change your Stripe setup?
Each model wires into Stripe differently. A trial means a subscription with a trial period, freemium usually means your own database decides who is on the free plan, and a reverse trial needs custom downgrade logic in your webhook handler.
Stripe's free trial documentation describes the classic path. You start a trial with trial_period_days or trial_end when you create the subscription, and the trial period must be 730 days or less. A trial doesn't need a payment method, and Stripe warns that skipping one can let spammers create many fake customers, suggesting an account step and a captcha first. When you start without a card, trial_settings.end_behavior.missing_payment_method sets what happens at the end:
| Value | Result when the trial ends with no payment method |
|---|---|
cancel | Subscription cancels immediately |
pause | Subscription pauses and can resume later |
create_invoice | An invoice is created; if unpaid, the subscription goes past_due |
Listen for customer.subscription.trial_will_end so you can warn customers before billing starts. Stripe's trial offers page says this event follows the Dashboard "Trial ending events" setting, which defaults to 3 days before the end. The same page introduces a newer Trial Offer API that requires API version 2026-09-30.endive or later and flexible billing mode, and it says Checkout doesn't support trial offers. Before you build, check which of the two paths your integration can use.
For freemium, Stripe usually enters at upgrade. Your app, not Stripe, enforces free-tier limits, because free users are often just rows with a plan flag. That's a design choice, not a Stripe requirement. For the states a subscription moves through once it exists, read SaaS subscription lifecycle states, and for the wiring itself the Stripe subscription billing setup guide.
BookBed, my Flutter, Firebase and Stripe booking product, shows the gap between wired and proven. Its case study lists subscription infrastructure with Stripe in live mode, webhook handlers and a trial lifecycle at €9 a month for up to twenty units, but notes it isn't yet activated for paid traffic. So I have no trial or freemium results from BookBed, and I won't pretend otherwise.
How do you decide for your own product?
Work through three questions in order, then instrument the result.
- How fast does value land? Days to a clear win favours a trial. Value that builds over weeks of habit favours freemium or a reverse trial.
- Do free users have strategic value? If they bring network effects, content or near-zero cost, freemium can pay for itself. If not, a trial keeps the funnel honest.
- What is the buying motion? Self-serve, bottom-up adoption favours freemium or a reverse trial. Evaluation-driven B2B favours a time-boxed trial.
If the answers don't give a clear winner, run a small test with one model on part of your new signups and compare activation before you rebuild the whole flow. Track signup-to-activation, activation-to-paid, and how often free users hit the ceiling. Those three numbers tell you whether to shorten a trial, tighten a free tier or change models. The decision is reversible, but it's cheaper to come close the first time than to migrate a live billing system.
Your next step: write down your activation event and the median days to reach it, then compare that number with a 14-day window. If most users need longer, what would you change first, the trial length or the onboarding?