Conversion Rate Calculator

The formula, worked through

All three tabs above use one fraction, rearranged. Here it is with the default numbers in place, so you can check the calculator against a pencil.

Google uses the same numbers in its own definition: “If you had 50 conversions from 1,000 interactions, your conversion rate would be 5%, since 50 ÷ 1,000 = 5%” (Google Ads Help). The two reverse tabs round up, because you can’t collect part of a conversion: a target of 82.5 means 83.

The first tab also shows a plausible range around your rate. It uses the Wilson score interval, which the NIST/SEMATECH e-Handbook of Statistical Methods sets out for proportions and notes cannot give a negative lower limit. It shows at a glance whether you have enough data. 50 conversions from 1,000 clicks is 5.00%, with a range of 3.81% to 6.53%. 3 conversions from 60 clicks is also 5.00%, but its range runs from 1.71% to 13.70%: the same headline with far less evidence behind it. The range assumes each visitor converts at most once, so it disappears when conversions exceed visitors.

Pick the denominator before you read the number

Every tool agrees on the division. They disagree on what goes underneath it, which is why the AdFlint conversion rate glossary entry says one funnel reports several rates. This is what each source counts, from their own documentation:

GA4 calls conversions key events; how they flow into Google Ads is covered under GA4 conversion tracking. Put your own numbers into the comparison below. The default shows 50 conversions against 1,000 ad clicks, 900 sessions and 700 users.

Three rates (5.00%, 5.56% and 7.14%) from one set of results. Two cautions. The sessions and users rows reuse the same 50, which is only right if every conversion came from a different session and person. GA4 counts sessions or users with at least one key event, so someone who converts twice can add 2 to an ad platform’s count but 1 to GA4’s; treat those rows as an upper bound. And ad clicks and sessions are recorded by different systems, so the two counts can differ. Dividing Google Ads conversions by GA4 sessions gives a number neither tool will ever show you.

Three examples to redo by hand

The rate. A plumber’s search campaign logs 18 calls and form fills from 420 clicks in a month. 18 ÷ 420 = 0.0429, × 100 = 4.29%, or about 1 in 23 clicks. The range is 2.73% to 6.67%, so a slide to 3.5% next month would not prove anything changed.

Conversions needed. A store expects 2,000 visitors during a sale and wants 3% of them to buy: 2,000 × 3 ÷ 100 = 60 orders. If it gets 45, its rate was 45 ÷ 2,000 × 100 = 2.25%.

Visitors needed. A clinic wants 30 booked consultations and converts 2.5% of ad clicks: 30 ÷ 0.025 = 1,200 clicks. If its cost per click is $3.00, that is 1,200 × $3.00 = $3,600. Use your own cost per click, not someone else’s; the ad budget calculator runs this the other way, from a budget to expected conversions.

Conversions and traffic at common rates

The calculator’s three tabs at five example rates, so you can read an answer without typing. Each row assumes 1,000 visitors and a goal of 50 conversions. The 5.00% row is the calculator’s default: 50 conversions from 1,000 clicks.

The visitors column shows what a lower rate costs in traffic: 50 conversions take 1,000 visitors at 5.00% and 5,000 at 1.00%. The range column is the caution. At 1.00% it runs from 0.54% to 1.83%, so its top is about 3.4 times its bottom; at 10.00% it runs from 8.29% to 12.02%, about 1.4 times. A low rate needs more traffic before the number settles.

Published benchmarks, and why they make a poor target

The public figures we could check come from WordStream/LocaliQ. Their 2026 search report covers 13,474 US search campaigns from April 1, 2025 to March 31, 2026 and puts the average conversion rate at 8.18% (WordStream search, 2026). Their 2026 Facebook report covers 452 lead-objective campaigns starting April 1, 2025 and puts the lead campaign rate at 8.54% (WordStream Facebook, 2026). Both say their “averages” are medians. The table lists every category that appears in both reports; last verified 2026-10-05.

Read them with these limits:

• The search figures pool Google Ads and Microsoft Ads campaigns. We found no split between the two in the report.
• The Meta figures are from campaigns using the leads objective. WordStream ties that objective to Lead Ads, a form built into the ad “that viewers can convert off of without ever having to leave Meta apps” (see instant forms), but doesn’t say how many of its campaigns sent people to a website instead. Read them as lead-campaign rates, not website conversion rates.
• Same industry, opposite directions: physicians and surgeons convert at 12.43% on search and 6.41% on Meta leads, real estate at 3.70% and 9.95%. Those are different events, not one channel beating the other.
• The Meta sample is thin: 452 campaigns across 14 categories, each needing only “at minimum 4 unique active campaigns”, against a minimum of 52 in the search report.
• Thin samples swing. The previous Facebook edition (726 campaigns, April 2024 to June 2025) put the lead rate at 7.72% and Education & Instruction at 10.08%; the 2026 edition has 15.87% for that category (WordStream Facebook, 2025).
• There is no e-commerce row. The closest search categories are Shopping, Collectibles & Gifts (4.01%) and Apparel / Fashion & Jewelry (4.50%); the Meta table has neither.

We found no dated, sourced public figure for website conversion rate in general, so this page doesn’t give one. Undated ranges that circulate without a source are left out. Your own history over a comparable period is a better baseline than any row above.

Mistakes that change the number, not the business

• Mixing sources. Google Ads conversions divided by GA4 sessions, or the reverse.
• Counting every conversion, then calling it a share of visitors. With “Every” counting, one click can carry several conversions.
• Mismatched dates. Google’s definition counts interactions “during the same time period”; this month’s sales over last month’s clicks is a different number.
• Blaming the page for a traffic change. New keywords or a broader audience can lower the rate while the page stays the same.
• Comparing a Meta lead-form rate with a website rate. One counts a form filled in without leaving Facebook or Instagram, the other an action on your own site.
• Reading a rate from a handful of conversions. Check the range first. To compare two versions of a page, use the A/B test significance calculator; the A/B testing glossary entry explains the setup.
• Treating a benchmark as a target. A median from other advertisers’ accounts says nothing about your offer or your price.

What conversion rate can’t tell you

It says nothing about money. Cost per conversion is cost per click ÷ conversion rate, so a 10% rate on $40 clicks costs $40 ÷ 0.10 = $400 per conversion, while a 1% rate on $0.50 clicks costs $0.50 ÷ 0.01 = $50. Whether either one makes money depends on what a conversion is worth and on your margin. For a sale, that is what the customer behind it is worth over time, which the customer lifetime value calculator estimates; for a lead, it is that figure times the share of leads that become customers. The CPA calculator and ROAS calculator then set it against what you paid.

This next part works against us. At the same click price, doubling your conversion rate halves your cost per conversion, while doubling your budget at best doubles cost and conversions together and leaves cost per conversion where it was. If your page converts poorly, fix the page before you pay for more traffic or for software to manage it, ours included. The landing page ad-readiness checklist and the landing page experience entry are free places to start.

Related tools: the CPC, CPM & CTR calculator turns spend, impressions and clicks into cost and click metrics in one pass; this page covers only the conversion step and the denominator choice. For the steps before the click, see the CTR calculator and the CPM calculator.