Should trial users count in your metrics?

From the SoleOS answers series — written about our own product space; grounded in published definitions and documented behavior, never invented numbers.

Trial users should never be counted in MRR — a trial is not revenue, and folding trial value into your monthly recurring revenue line makes the business look bigger than it actually is. You can and should track trial count or "trial MRR" as a separate leading indicator, but keep it out of the real number. What actually matters is trial-to-paid conversion, not how many trials you're currently running.

Why trials shouldn't count toward MRR

MRR is supposed to be revenue you can count on next month. A trial user — card on file or not — hasn't committed to anything yet. They can cancel for free, and a meaningful share of them will. Fold trial value into MRR and you're reporting revenue that doesn't exist: every month, some of it evaporates as trials expire without converting, and the number you reported has to be quietly walked back.

This matters most when MRR is something you report — to yourself in a monthly review, to a co-founder, to anyone deciding whether the business is working. If trial value is baked in, a marketing push that fills the funnel with trials looks identical, on a dashboard, to actual revenue growth. It isn't. For the underlying mechanics of what belongs on that line and what doesn't, see ARR vs MRR for a small SaaS.

None of this means trial activity is unimportant. It means it belongs in its own bucket, not blended into revenue.

Track trial MRR and trial count — just label them separately

If you want visibility into what's in the pipeline, that's legitimate. Two things worth watching:

• Trial count — how many people are currently mid-trial.
• Trial MRR (or "potential MRR") — what MRR would become if every current trial converted at full price.

Both are useful leading indicators of what's coming, not of what's landed. The rule is simply to never let them touch the actual MRR figure. Label the chart, the column, the message — anything that could get misread as revenue. "Trial MRR: not yet recurring" costs a few words and prevents a stakeholder from mistaking pipeline for revenue three weeks later.

Trials do belong in activation and engagement metrics

MRR isn't the only kind of metric that matters. If you're measuring product usage — logins, key actions, feature adoption, day-1 or day-7 retention — trial users belong in that data, and arguably matter more there than paying customers do. That's your earliest signal of whether the product is doing its job, before any money is on the table.

If a trial user never opens the app after signing up, or opens it once and disappears, that's a real, actionable finding whether or not they ever pay. Excluding trials from engagement metrics would blind you to exactly the population you most need to understand right now: the one deciding whether to become a customer. For a shorter list of which metrics are worth watching in the first place, see first metrics for a new app.

The metric that actually matters: trial-to-paid conversion

Raw trial count describes top-of-funnel interest, not business health. A launch, a marketing push, or a spike in signups can send trial numbers up while the business itself stays flat or gets worse — if none of those trials convert, all you've added is support load, not revenue.

Trial-to-paid conversion rate is the number that says something. Of the trials that reached a decision point — converted or expired, not just the ones still in flight — what share became paying customers? That's worth tracking over time, worth segmenting by acquisition channel, and worth trying to move with onboarding changes. A view that separates "trials running," "trials converted," and "trials that lapsed" gives you that ratio without smuggling any of the three into MRR — the kind of split we built SoleOS's metrics view around, since it's the same reconciliation problem across every product in a portfolio.

Two things worth watching here, rather than a benchmark number to chase — pull your own and compare against your own history:

• Conversion rate over time — is it improving, flat, or slipping as onboarding or trial length changes?
• Conversion rate by source — do trials from organic search convert differently than ones from a launch spike or paid ads? A channel that brings in trials that never convert is a channel to reconsider, not celebrate just because it inflates the trial count.

Free trial vs. freemium — they get treated differently

The two get conflated constantly, and the accounting difference matters.

Free trial — time-limited access, often a week or two, after which the user pays or loses access. It may or may not require a card upfront. Every trial user is, by definition, temporary: they convert or churn out on a known clock. That's why trial MRR is a forecast, not a fact — a calendar resolves it one way or the other.

Freemium — a permanent free tier that sits alongside paid tiers, with no expiration. Freemium users aren't "pending" a decision the way trial users are; a large share will use the free tier indefinitely and never convert, by design. That changes the metric that matters: freemium is measured by free-to-paid conversion over the entire free-user base — typically a much lower rate, over a much longer and less-defined window — rather than trial-to-paid conversion against a fixed deadline.

This also changes what "trial count" even means. A freemium product doesn't have a trial count; it has a free-user count, which carries a different implied ceiling on how much of it will ever convert. If you're still deciding between the two models, that choice shapes the entire pricing page and paywall around it — see how to price a micro-SaaS for the tradeoffs.

A rule of thumb for what goes where

• MRR / ARR — paying customers only. No trials, no freemium free-tier users, ever.
• Trial MRR / potential MRR — trials only, clearly labeled as not-yet-recurring, kept separate from actual MRR.
• Active users / engagement — trials and paying customers both. You want the full activation picture.
• Trial-to-paid conversion — trials that reached a decision point (converted or lapsed), not everyone currently mid-trial.
• Free-to-paid conversion (freemium) — free-tier users only, understood as a longer, lower-rate funnel than a trial.

Keep those five separated and most "wait, is our MRR actually going up?" arguments resolve themselves before they start.

SoleOS published this. It's a portfolio dashboard for founders running more than one small product, and getting metric definitions like this right by default is most of what it's for. If you run a single product off a spreadsheet you already trust, and nobody but you ever looks at the numbers, you probably don't need a tool for this — the rule of thumb above is enough on its own.

Frequently asked questions

Should I count a trial as a "customer" in my customer count?

No, not until they convert. If you need a number for trials in progress, call it that — "active trials" or "trial users" — rather than folding it into a customer count that then gets compared against MRR. The honest label matters more than the number itself.

What if the trial requires a credit card upfront?

Requiring a card tends to change your trial-to-paid conversion rate — it filters for higher intent and usually produces a cleaner number — but it doesn't change the accounting. A card-required trial user is still not a paying customer until the trial converts and the first charge actually succeeds. Don't count an authorized card as revenue.

How do I measure trial-to-paid conversion while trials are still in progress?

Measure it against trials that have reached a resolution — converted or expired — not against everyone currently mid-trial, or you'll understate the rate every time you look, since some of those in-flight trials haven't had the chance to convert yet. For a real-time view, track "converted so far" and "still active" as separate figures rather than blending them into one ratio too early.

Does a trial that later churns still count as a conversion?

Yes. Conversion and churn are two separate events. A trial that converts to paid and cancels a month later was a successful conversion followed by later churn. Blending the two into a single metric hides which problem you actually have — people not converting, or converting and then leaving — and those need different fixes.

Should freemium free-tier users show up anywhere in revenue reporting?

Not in MRR, and generally not in ARR either — they aren't paying and a large share never will. They belong in usage and engagement metrics, and in free-to-paid conversion tracking, which is a distinct metric with a different, usually much lower and slower, baseline than trial-to-paid conversion.