Conversion rate (CVR) is the percentage of visitors who do the thing you wanted — usually, buy something. If 100 people click your ad and 3 of them order, your conversion rate is 3%. It's a metric — a number — not a dimension (a category you slice numbers by, like "landing page" or "device").
Picture a physical shop. A hundred people walk in over an afternoon. Some browse, some pick things up and put them back, one asks where the toilets are. Three actually buy something. Your conversion rate is 3% — and everything about your pricing, your layout, your stock, and your checkout queue is hiding inside that number.
The formula (with actual numbers)
CVR = (Conversions ÷ Clicks) × 100
400 ad clicks, 12 purchases → 12 ÷ 400 × 100 = 3% conversion rate.
One honest complication: ad platforms divide by clicks, while GA4 divides by sessions — and those are not the same thing, so the two numbers will never quite agree. This is normal, not broken. Pick one definition per report and note which it is, or you'll spend a meeting arguing about a discrepancy that's purely definitional. (We have witnessed this meeting. It was long.)
Why should I care about conversion rate?
Because CVR is the single biggest lever connecting traffic to revenue. Doubling your traffic is expensive — you pay for every extra click. Doubling your conversion rate makes every click you already pay for worth twice as much, and it halves your cost per sale without touching a single bid. Same shop, same foot traffic, twice the sales. It's the closest thing digital marketing has to free money, which is exactly why it's never easy.
What's a good conversion rate?
Ranges we see, offered with the standard "your mileage will vary" disclaimer:
- E-commerce, site-wide: commonly 1–3%
- Paid search traffic to purchase: often 2–5% — these people were actively looking
- Lead-generation forms: can reach 5–15% (a form is a smaller ask than a credit card)
- Remarketing audiences: convert at multiples of cold traffic — they already know you
The honest caveat: CVR depends on price point, category, brand strength, and how warm the traffic is. A €900 sofa converts slower than a €19 phone case, and neither shop is doing anything wrong.
The classic mistakes
Judging the blended average. Site-wide CVR mixes high-converting brand traffic with low-converting cold prospecting. If the mix shifts, blended CVR moves with no real change in performance underneath. Segment before judging — always.
Assuming a sudden drop means customers changed their minds. In our experience, an overnight CVR collapse is far more often a tracking problem than a human one — a tag broke, a consent banner update, a checkout change that orphaned the purchase event. Check tag health before redesigning anything. Humanity rarely changes its shopping behaviour on a Tuesday.
Never looking at CVR per page. This is where the money hides. Pages that attract solid traffic but convert poorly are quiet revenue leaks — and one of the most common, costly versions is paid traffic flowing to out-of-stock products. A classic find in audits: ads cheerfully sending visitors to pages where the buy button says "unavailable", costing a few hundred euros a month while every top-level number looks fine. (Finding exactly these pages is what our Conversion Gap analysis was built for, for what it's worth.)
How do I improve my conversion rate?
Send people to the page they expected — category ads to category pages, not the homepage. Speed the site up, especially on mobile, where most of your traffic probably is and most of your patience-testing definitely is. Cut checkout steps and offer the payment and delivery options your customers actually use. Add trust signals: reviews, a clear returns policy, recognizable payment logos. And test on your highest-traffic pages first — a 10% improvement on your busiest page beats a 50% improvement on a page nobody visits.
Related metrics worth knowing: CTR (who clicked the ad — the step before CVR), CPA (cost per sale: CPA = CPC ÷ CVR, so better CVR mathematically cuts it), ROAS (the revenue view), and AOV (how much each conversion is worth).
Key Idea: Doubling your conversion rate halves your cost per sale without spending one extra euro on ads. It's the cheapest growth most shops are sitting on.
This week's homework: find your five most-visited product pages and check two things on each: is the product in stock, and does the page load fast on a phone? Fix whatever fails. That's the whole assignment.
If you'd like the traffic-but-no-sales pages found for you automatically — crawl plus analytics, ranked by lost revenue — that's our Conversion Gap analysis at airdan.ai. No pressure; the homework above is free.
FAQ
What is a good conversion rate for e-commerce? Site-wide, 1–3% is a common range; paid-search traffic to purchase often lands at 2–5%. Price point and category move these a lot, so your own trend matters more than any benchmark.
Why did my conversion rate suddenly drop? Sudden overnight drops are usually tracking breakage — a broken tag, a consent banner change, or a checkout update — rather than real behaviour change. Verify tracking first.
What's the difference between CVR and CTR? CTR measures who clicked your ad after seeing it; CVR measures who converted after clicking. They describe two different doors in the same corridor.
How do I calculate conversion rate? Divide conversions by clicks (or sessions, in GA4) and multiply by 100. Note that click-based and session-based versions won't match — pick one and label it.