What is conversion rate?
Conversion rate is the share of visitors or users who complete a target action, such as a signup, purchase, or upgrade, out of everyone who had the chance to complete it. It is expressed as a percentage and is the core measure of how well a page, flow, or product turns attention into a desired outcome.
What is the conversion rate formula?
The formula is one division: conversions divided by the total that could have converted, times 100. Forty signups from a thousand visitors is a 4 percent conversion rate. The math is trivial, so the whole game lives in the denominator.
Use everyone who actually reached the step you are measuring, not everyone who ever touched the site. If you are measuring checkout-to-paid conversion, the denominator is people who started checkout, not total visitors, most of whom never got near a cart. Mixing the wrong denominator in is the single most common way a conversion number ends up lying to a team.
The same denominator discipline is what makes funnel analysis work: each step has its own clean population, so each step gets its own honest rate.
What is a micro vs a macro conversion?
Not every conversion is the sale. It helps to split them into two sizes: the big outcome you get judged on, and the small steps that lead to it. Both are worth tracking, for different reasons.
When the macro number drops, the micro conversions tell you where. A flat macro rate with a collapsed pricing-page click, for instance, points you straight at the step to fix instead of guessing.
Funnel-step vs end-to-end conversion rate
The same flow has two very different conversion rates depending on where you draw the start and end line, and confusing them is how teams argue past each other.
Funnel-step conversion is the rate between two adjacent steps, checkout to paid, or trial to upgrade. It is a magnifying glass: it tells you exactly which handoff in the flow is leaking, and it is what you optimize one change at a time.
End-to-end conversion is the rate across the whole journey, landing page to paid. It is the yield of the entire flow, and it is the number a business actually lives on. A flow can look healthy at every single step and still convert badly end to end, because small leaks multiply: 80 percent times 80 percent times 80 percent is barely half.
So always say which one you mean. "Our conversion rate is 40 percent" is a step rate or a whole-flow rate, and the difference changes every decision that follows.
What is a good conversion rate?
There is no universal good number, and anyone quoting one out of context is selling something. A conversion rate that is excellent for a high-price enterprise contract would be a disaster for a free-to-signup app, and the same page converts differently for warm search traffic than for a cold ad click.
Conversion rate varies by industry, traffic source, price point, and how much intent your audience arrived with. Those variables swamp any average, which is why we will not print a benchmark number here, it would be a made-up figure dressed up as a target, and it would point you at the wrong thing.
The only benchmark that means anything is your own. Measure your current baseline, ship a change, and see whether the rate moved. Your trend against yourself, over the same traffic and the same audience, is the real scoreboard. Attribution is what lets you compare fairly, by splitting the rate out per source instead of blending warm and cold traffic into one misleading average.
How smolanalytics computes conversion rate
smolanalytics gives you funnels with a conversion rate at every step and end to end, alongside web and product analytics from one snippet. What makes it different is not that it has conversion rate, every tool does, but four choices about how you get the answer:
- 1Every step, plus a verdict. You see the step-by-step and end-to-end rate for each funnel, and beyond that, the verdict flags the step that is bleeding, on the dashboard and in a morning brief, so a dropped conversion reaches you as a decision instead of a chart you had to remember to open.
- 2Ask in plain English. Type "what is my checkout conversion?" or "did signup conversion drop this week?" into a dashboard bar, or into your own Cursor / Claude Code over MCP (47 tools, 13 prompts), using your own AI model so the AI part is free.
- 3Computed, never guessed. Every conversion-rate answer comes from the same deterministic report the dashboard renders, not from an LLM writing numbers. A CI agreement test fails the build if the AI answer ever differs from the dashboard, so the percentage you get is the real one.
- 4One binary. It is a single MIT-licensed Go binary, stdlib only, roughly 7 bytes per event, no Kafka, ClickHouse, or Postgres to run. Self-host it free forever, or use the hosted cloud.
It deliberately does not do session replay, feature flags, experiments, heatmaps, or surveys. It is for teams who want a straight, owned, cheap read on what is converting. See every feature, the related funnel analysis, or how it compares vs Mixpanel.