Five numbers decide whether a Shopify app is a real business: install-to-trial rate, trial-to-paid conversion, monthly churn, ARPU with its expansion trajectory, and net revenue retention. Every other dashboard metric is a diagnostic for one of these five.
- Know these five cold and the business stops being a mystery; miss one and the constraint hides.
- More dashboards feel like more control but actually let the real constraint hide in plain sight.
- A rising install chart can mask a business that churn is quietly shrinking.
Five numbers decide whether a Shopify app is a real business: install-to-trial rate, trial-to-paid conversion, monthly churn, ARPU (average revenue per user) with its expansion trajectory, and net revenue retention. Every other metric in your dashboard is a diagnostic for one of these five. Know these cold and the business stops being a mystery. Miss one and the constraint hides behind a wall of charts you keep checking instead. The wider set of numbers, brand and app, lives in the 2026 Shopify ecosystem benchmark report.
I have sat across from a lot of app founders. They show me twenty slides. A weekly install chart, MRR by plan tier, a cohort retention heatmap nobody has read in six months, a support ticket graph. Then they ask whether the business is healthy. The honest answer is that you can fit the whole question on one screen, and I am going to show you what that screen looks like.
Twenty dashboards
hide the one
line that matters.
The trap with app analytics is that more visibility feels like more control. It is the opposite. When every metric is on the wall, the constraint hides in plain sight. A founder spends a month optimizing install volume while a 6% monthly churn rate quietly empties the bucket behind them. The install chart goes up and to the right. The business is shrinking.
The discipline is to reduce the business to the smallest set of numbers that, taken together, fully describe its shape. For a Shopify app that set is five: how many shops install and start a trial, how many of those convert to paid, how fast the paying base leaks, what each account is worth, and whether the surviving accounts grow over time. That is the funnel and the engine, top to bottom.
The five-number rule
If a number is not one of the five, it is a diagnostic for one of the five, not a peer to it. Sessions, star ratings, and support tickets all matter, but they earn their place by explaining a movement in install rate, conversion, churn, ARPU, or NRR. Keep the hierarchy clean and the constraint stays visible. Sessions fall because onboarding stalls, which shows up in trial-to-paid. Ratings drop before churn rises. Support volume predicts cancellations three weeks out. Every secondary metric is a leading indicator for one of the five primaries. Treat it that way.
What each metric
actually measures,
and what can move it.
Before you benchmark, you need the definitions nailed. Founders disagree on these more often than you would expect, and a loose definition produces a misleading number. Here is what I mean by each of the five, and the levers that actually move them.
1. Install-to-trial rate
This measures the percentage of installs that reach a meaningful first session inside your app, typically defined as completing setup through the point where the core value is visible. An install that bounces at the configuration screen is not a trial start. It counts the shop that got far enough to see what you actually do.
What moves it: your App Store listing and your onboarding entry ramp. The listing sets expectations. If your screenshots show a polished dashboard but the first screen in the app is a five-field configuration form, you will bleed installs immediately. A strong onboarding sequence gets shops to a "working" state in one session without requiring a support ticket. The onboarding benchmarks post walks through what the entry ramp should look like at each stage.
2. Trial-to-paid conversion
This is the percentage of trial starts that convert to a paying plan, measured at the end of the trial window. It is the single most leverage-heavy number for a young app, because it multiplies everything above it in the funnel.
Moving trial-to-paid from 40% to 65% roughly doubles your paid intake from the same install volume, with no additional marketing spend. That math is why I push every early-stage app toward onboarding before acquisition. There is a longer argument in the MVP to $1M ARR post, but the one-sentence version is: you cannot out-acquire a broken conversion step. The free-to-paid conversion guide covers the specific tactics that move this number, and the free trial-to-paid calculator shows the MRR sitting in your own gap.
3. Monthly churn rate
Monthly churn is the percentage of paying accounts (or MRR, depending on whether you weight by account or revenue) that cancel or downgrade to zero in a given month. This is the drain at the bottom of the bucket. Everything you pour in at the top flows through it.
What moves it: not cancellation save flows. Churn is almost always a value-delivery problem wearing a timing costume. The shop cancels in month three because the value they expected never materialized, and the timeline just reflects when their patience ran out. The churn-as-symptom post is the full argument. For the tactical side, the 90-day save playbook covers what you can do once a shop signals they are drifting.
4. ARPU and expansion revenue
Average revenue per user at a point in time, and the trajectory of that number over the customer lifetime. A flat ARPU line on a growing account base is a pricing problem, not a product problem. You have already won the account. They are paying to stay. You have just never given them a reason to pay more.
Expansion revenue comes from usage-based tiers, plan upgrades triggered by hitting limits, and add-on features that solve adjacent problems for your existing base. It is the cheapest growth there is. If your pricing is a flat monthly fee with no usage ceiling and no upgrade path, ARPU cannot rise. The pricing strategy post has the full framework for building a plan structure that allows for real expansion.
5. Net revenue retention (NRR)
NRR measures what happens to the revenue from your existing cohort over time, after accounting for churn, downgrades, upgrades, and expansion. The formula: (starting MRR + expansion MRR - churned MRR - contraction MRR) / starting MRR, expressed as a percentage.
NRR above 100% means your existing base grows on its own. You could stop acquiring new customers today and the revenue number would still go up. That is the threshold where the business shifts from needing to run in place to actually compounding. The strongest apps in the Shopify ecosystem run well above 100%. Below 100% means churn and contraction are outpacing expansion, which is the normal condition for apps that have not yet figured out their retention story.
Five lines.
Three columns.
One screen.
This is the reference. The left column is the metric, the middle is roughly where a healthy app sits as it matures (these are directional ranges from my experience across many app businesses, not published benchmarks), and the right is what the number is actually telling you.
| Metric | What good looks like | What it tells you |
|---|---|---|
Install → trial start % of installs reaching active trial | Most installs reach a real first session. Drops sharply if onboarding stalls at setup. | Listing honesty and setup friction. Low here means the listing oversells or onboarding breaks at step one. Fix the entry ramp before anything else. |
Trial → paid % of trial starts converting to a paid plan | ~60–75% with strong onboarding. Young apps often start lower while finding the right trial structure. | Whether the trial delivers value before it ends. The single most leverage-heavy line for an early-stage app. A 25-point improvement here roughly doubles paid intake from the same installs. |
Monthly churn % of paying accounts or MRR lost per month | ~3–5% while finding fit, trending toward ~1–2% at maturity. Sustained above ~5% is a product problem. | Product-market fit and stickiness. High churn is not a save-flow problem. It is a value-delivery problem with a delayed reveal. |
ARPU and expansion Average revenue per paying account, plus its trend | Rising over the customer lifetime, with a clear trigger for upgrade. Flat ARPU on a growing base is a pricing problem. | Pricing power. Flat ARPU means you are leaving money on usage you already deliver. A usage ceiling or a compelling upgrade feature fixes this faster than a price increase. |
Net revenue retention % of cohort revenue retained plus expansion after one period | Above 100%. The strongest apps in the ecosystem run well above this. Below 100% means churn and contraction outpace expansion. | Whether the existing base grows on its own. Above 100% means you could stop acquiring today and still grow. Below 100% means the bucket has a hole and acquisition is filling it faster than it drains. |
Annual numbers map onto the same shape. Apps with strong retention hold annual revenue churn well under 25%. The monthly ranges above are the same story told at higher resolution. If your monthly churn looks manageable in isolation, calculate the annual equivalent: 3% monthly is roughly 31% annually, which means you are replacing nearly a third of your revenue base every year just to stay flat.
Top to bottom,
the lines are not
equal weight.
Read the chart top to bottom and a sequence appears. Installs and trial starts are the mouth. Trial-to-paid is the gate. Churn is the drain. ARPU and NRR are whether the water you keep rises on its own. The mistake founders make is treating all five as if effort spent on each pays off equally. It does not. The value of improving each line depends entirely on where you are in your lifecycle and what the other lines already look like.
Early on, trial-to-paid is where the leverage lives. The math is unambiguous: if you convert 40% of trials today, getting to 65% is a 62% increase in paid intake from the same install volume. That is equivalent to growing installs by 62%, except it costs you zero in acquisition. You get it by fixing onboarding, by shortening time-to-first-value, and by structuring the trial to end at the right moment rather than an arbitrary day count. The onboarding benchmarks piece has the framework for getting there.
Later, the priority inverts. Once you have real scale, a single percentage point of monthly churn reduction is worth more than almost anything you can do at the top of the funnel, because it compounds against your entire base every single month. If you have 1,000 paying shops, dropping monthly churn from 4% to 3% saves you ten shops a month. That is 120 shops a year that did not require a single install to acquire. This is the part founders feel last and regret most: the compounding cost of churn they could have fixed earlier but managed around instead.
"You cannot out-acquire a broken conversion step, and you cannot out-acquire a leaking base. The chart tells you which one you are fighting this quarter."
Send me your five lines and I will tell you which one is the constraint. The form takes two minutes.
The drain
is louder than
the mouth.
When I open a dashboard, I look at the bottom of the chart first. NRR and churn tell me whether the product earns its keep, before I care how many people are finding it. A beautiful install curve sitting on top of 6% monthly churn is a marketing budget being lit on fire. The installs feel like progress. The business is smaller every month.
The reason churn deserves first read is that it is the most misdiagnosed line of the five. Founders see the number rise and reach for a cancellation save flow or a win-back discount. That treats the symptom. The number is telling you something about value delivery upstream, and the discount just delays the same departure by 30 days. I laid out the full argument in why churn is a symptom and not the problem. It is the post I send more often than any other. The companion to that, for when you already have accounts you need to save, is the 90-day save playbook.
A significant portion of App Store apps have no reviews at all. That is not a chart line, but it is the canary. An app generating real value from paying shops accumulates reviews as a byproduct. A wall of zero reviews next to a respectable install count usually means the value never lands, which will show up in the churn line a few months later. Zero reviews is not a marketing problem. It is a value problem. The metric to fix is not the rating count; it is whatever is causing shops to uninstall without engaging deeply enough to leave a trace.
The other quiet leak is flat ARPU. An app can look clean on churn and conversion and still be a much smaller business than it should be because it never built a reason to pay more. Expansion revenue is the cheapest growth there is: you have already won the account, you have their trust, they are in the workflow. A chart with a flat ARPU line and good retention is a pricing problem wearing a healthy disguise. The fix is usually a usage trigger and a clearly superior plan above it, not a price increase on the existing plan. If you want to work through the mechanics, the pricing strategy post walks through how to structure tiers for real expansion.
The acquisition trap
The pattern I see most often: a founder looks at a rough churn number and decides to grow through acquisition instead of fixing the underlying problem. They figure more installs will hide the leak. For a quarter, it works. The MRR line goes up. Then churn catches up. Now they need even more installs just to hold the number flat. The acquisition budget doubles, then triples. The economics never close.
This is the most expensive mistake in the Shopify app ecosystem, and it is almost entirely avoidable. The chart makes it impossible to hide because churn and NRR are both visible. If you have to ignore two lines to tell yourself the story that acquisition is the lever, the chart is doing its job.
The distribution playbook covers when acquisition spend makes sense and what the qualifying metrics look like. The answer is almost always: later than you think, and only after the bottom of the chart is solid.
One constraint
per quarter.
Name it.
The chart is not a scorecard to admire. It is a tool for picking one fight at a time. Each quarter, find the line that is furthest from its healthy range, weighted by how much moving it would affect the business. Make that the constraint. Everything else holds steady while you work that one number. You can revisit next quarter.
- If install-to-trial is low: the listing and the first screen are misaligned. Fix the listing or fix the entry ramp before touching anything else.
- If trial-to-paid is under 50%: this is the quarter. Shorten time-to-first-value. Remove the steps between install and the moment the app proves itself.
- If monthly churn sits above 5%: stop acquisition spend mentally and ask what value is failing to land. A save flow is not the answer. The problem is upstream.
- If churn and conversion are fine but ARPU is flat: the constraint is pricing and packaging, not the product. Build the usage ceiling and the upgrade trigger.
- If everything reads healthy and NRR is comfortably above 100%: the constraint is now distribution. You have earned the right to spend on acquisition.
That last position is the prize, and almost nobody gets there before doing the work on the lines above it. If you are an app founder wondering whether you have earned the right to grow, the metric that answers the question is NRR. Below 100% and you are filling a bucket with a hole in it. Above 100% and every install you buy compounds.
If you want a second set of eyes on your five lines, that is the kind of thing I cover in a first call. I have seen this pattern in enough app businesses to usually identify the real constraint within the first twenty minutes. The SaaS advisory engagement is designed for exactly this, and the inquiry form is where to start.
Common questions
on app health
metrics.
What is a good trial-to-paid conversion rate for a Shopify app?
A well-onboarded Shopify app should convert 60 to 75 percent of free trial installs into paying accounts. Below 50 percent is a signal that your trial is not delivering value before it ends, whether that is a friction problem in setup or a promise problem in your listing. Apps that nail onboarding get here without a lot of other work. Apps that skip onboarding rarely do.
What monthly churn rate is acceptable?
Early apps commonly run 3 to 5 percent monthly churn while finding product-market fit. A maturing app should trend toward 1 to 2 percent. Sustained churn above 5 percent monthly is a product-market fit problem, not a cancellation-flow problem. No save sequence fixes a product that fails to deliver ongoing value. The sequence is: diagnose the value failure, fix it, then measure whether churn moves. In that order.
What is net revenue retention and why does it matter?
NRR measures whether your existing revenue base grows on its own, after accounting for churn, downgrades, upgrades, and expansion. Above 100 percent means you could stop acquiring new customers today and still grow. Below 100 percent means churn and contraction outpace expansion. The strongest Shopify apps have NRR well above 100 percent, driven by usage-based tiers and genuine expansion triggers, not pricing increases on captive accounts.
Should I focus on acquisition or retention first?
Retention first, almost every time. Acquisition into a leaking base compounds the loss and eventually breaks the unit economics. Fix trial-to-paid conversion and monthly churn before increasing marketing spend. The test is NRR: once it clears 100 percent and is trending up, acquisition spend has a real multiplier. Before that, you are filling a bucket with a hole in it, and more water just means a bigger puddle on the floor.
How do I know which metric to work on first?
Read the chart top to bottom and find the line furthest from its healthy range, weighted by business impact. Trial-to-paid below 50 percent is almost always the first priority for young apps, because the leverage is highest there. For mature apps with scale, a single point of monthly churn reduction beats almost any acquisition push in ROI terms. Name the one constraint, work it for a quarter, then reassess.
Keep the five lines visible and the business stops being a mystery. The posts that go deepest on each metric: churn as a symptom, free-to-paid conversion, pricing for expansion, and onboarding benchmarks. If you want a second set of eyes on which line is your real constraint, send your numbers through the inquiry form and we will figure it out on a first call.
Run your numbers against the chart.
Send me your funnel and I will tell you which line is the one holding the business back. Most founders are wrong about which one it is.
Start a conversation More about Taylor →