Why is my trial-to-paid conversion rate low? Causes and tests

Your trial-to-paid conversion rate is low if it sits under the “good” band in ChartMogul’s 2026 survey of 200 B2B software products: 4–6% for free trials, 25–35% when a card is required upfront. The cause is usually one of four: the wrong denominator, the wrong signups, trials that never activate, or activated trials that stall at the purchase.

• Is it actually low? Compare like with like. The ChartMogul SaaS Conversion Report (self-reported, January 2026) puts the median at 8% across all products and 30% for card-required trials.
• Cause 1, measurement: junk signups in the denominator, or a window that closes on the last trial day.
• Cause 2, traffic: trials from outside your target customer, which convert near zero whatever you do.
• Cause 3, activation: target users never reach first value. In Lenny Rachitsky’s survey of 500+ responses, median activation was 25% (30% for SaaS alone).
• Cause 4, the purchase: activated target users still do not pay: price, plan fit, payment, or a buyer who is not the evaluator.
• The test that separates them: the trial-to-paid cohort split, below.

Why is my trial-to-paid conversion rate low? Check it is low first

Many “low trial-to-paid” problems are two teams computing different fractions. Before diagnosing anything, write down three choices: the denominator (all signups, signups minus junk, or activated trials), the window (conversions during the trial, or within a fixed period after signup), and the entry mechanic (card required or not).

Then compare with the band for your mechanic. ChartMogul’s report, written by its analyst-in-residence Kyle Poyar with ProductLed, gives “good” and “great” bands by model. It is a self-reported survey of 200 B2B products, typically $1–10M ARR, so treat the bands as direction, not a verdict. Note its fraction: leads or free signups that became paying customers within six months, not conversions inside the trial window, so a rate measured on the last trial day will read low against it.

A no-card trial converting at 5% is not low, however disappointing it feels next to a card-required competitor’s 30%.

Trial-to-paid conversion is only low relative to trials with the same entry mechanic; a no-card trial and a card-required trial are different funnels.

Low trial-to-paid: symptom, likely cause, and the test that confirms it

The cut-offs in the test column (a quarter of your best channel, a 60-day look-back) are our reading rules for a first pass, not published thresholds.

A low trial-to-paid rate caused by the denominator or the traffic will not respond to any change inside the product, so those two are tested first.

The trial-to-paid cohort split: a worked diagnosis

The trial-to-paid cohort split is our name for the one test that separates the four causes. Split one month of trials two ways: target customer or not (by your own firmographic rule), and activated or not (by your activation milestone). Then read the rate in each cell. Illustrative inputs; substitute your own.

How to read it:

• Traffic: 600 of 1,000 trials (60%) are non-target and produce 12 of 80 paid accounts. Cutting half of it removes 300 trials and about 6 paid accounts, and the blended rate rises from 8% to 10.6% (74 ÷ 700).
• Activation: only 160 of 400 target trials (40%) activate. Lifting that to 60% moves 80 target trials from the 5% cell to the 35% cell: 28 paid instead of 4, a net gain of 24, from 80 to about 104 paid accounts.
• Purchase: target activated trials convert at 35%. The buying step is working, so a discount would give money away.

If instead the target-activated cell converted at 10%, the diagnosis flips: people get value and still do not buy, which points at price, plan fit, payment or the buying group, the bottom three rows of the table above.

Under the trial-to-paid cohort split, the cell with the largest gap between its rate and what it should be is the cause, and the blended rate hides all of them.

What if activated trials still don’t pay?

This is the cause people usually jump to first and should usually test last, because it is the one that a discount appears to fix. Three tests narrow it down.

1. Look at the pricing page, not the product. If activated users view pricing and leave without starting checkout, the price or plan boundaries are the question. If they start checkout and abandon, it is payment friction.
2. Split by team size. If single-user trials convert and team trials do not, the evaluator is waiting on someone else’s approval.
3. Ask the ones who did not buy. A short question to expired, activated, target trials (“what stopped you?”) produces reasons no dashboard can. Twenty answers is enough to see a pattern.

The specific blocks those conversations surface (stalled imports, integrations the evaluator cannot authorise, unasked security questions, wrong seat counts) and when in the trial to have them are set out in trial-to-paid conversion with outbound conversations.

When activated target trials still do not pay, the constraint is the purchase decision, and asking twenty expired trials what stopped them is the cheapest test there is.

What this diagnosis costs to run yourself

The measurement is a data job, not a strategy exercise. You need a trial table with signup date, source, company domain, an activation flag, trial end and first payment date, joined in one place. For a team with product analytics already instrumented, building the cohort split is a day or two of an analyst’s time; without an activation event defined, add the work of choosing one. Lenny’s survey defines a good activation milestone as one where users who reach it retain at least twice as well as those who don’t.

Where the diagnosis points upstream, the fix sits in a different place: whether interested visitors start at all (how to increase free trial opt-in rate), or whether signups finish creating an account (how to increase SaaS registration rate). Where it points at team-sized trials needing a human conversation, the cost is people’s time per trial, which is the part some SaaS teams hand to an outside B2B SaaS appointment-setting service.

A trial-to-paid diagnosis needs one joined trial table and an activation flag; without the activation flag, the four causes cannot be told apart.

Trial-to-paid conversion rate questions

What is a good trial-to-paid conversion rate?

It depends on whether you ask for a card. ChartMogul’s 2026 SaaS Conversion Report, a self-reported survey of 200 B2B software products, rates 4–6% as good and 10–15% as great for free trials, and 25–35% as good and 50–60% as great when a card is required upfront (ChartMogul SaaS Conversion Report).

Why did my trial-to-paid rate drop suddenly?

A sudden drop with no product change is usually the denominator. A new campaign, a removed card requirement or a looser signup form adds low-intent trials. Recompute the rate on signups with a work email and at least one session; if that rate held steady, the product did not get worse.

Should I shorten my free trial to improve conversion?

Test it rather than assume. First check when paying accounts actually convert relative to signup: if most pay well before the trial ends, a shorter trial may lose little; if many pay after expiry, your window is already too tight to measure. In ChartMogul’s 2026 sample, 14 days was the most common trial length, at 62% of free-trial products.

What is a good activation rate for a SaaS trial?

In Lenny Rachitsky’s survey of more than 500 responses, the average activation rate was 34% and the median 25%; for SaaS products alone the average was 36% and the median 30% (Lenny’s Newsletter, What is a good activation rate). Your milestone should be one where users who hit it retain at least twice as well as those who do not.

Is low trial-to-paid a product problem or a sales problem?

Run the cohort split. If target trials that activate convert well, the product and price are working and the loss is in traffic or onboarding. If target trials that activate still do not buy, the loss is at the purchase: price, plan fit, payment, or a buyer who never saw the trial.