Conversion Rate Calculator

A site with 40,000 visitors and 900 orders has a conversion rate of 2.25% — conversions ÷ visitors × 100. At an $80 average order value that is $72,000 in revenue, or $1.80 of revenue per visitor. Lifting that rate to 3.00% — a 33.33% relative improvement — produces 300 extra orders, $24,000 more revenue and $10,800 more gross profit at a 45% margin, without buying a single additional visit.

That last clause is the entire argument for conversion work. Every other lever in the funnel costs money to pull. More traffic costs media spend. Higher prices cost you volume. A conversion rate improvement is the one change that makes every number downstream of it better at the same time — it lowers your cost per acquisition, raises the cost per click you can afford to pay, and improves your return on ad spend, all from the same budget.

This guide covers how to calculate conversion rate properly, what a good CVR looks like by industry and device, why chasing the "average" is a trap, the difference between micro and macro conversions, how much traffic you need before a test result means anything, and the fixes that move the number most.

What is a conversion rate?

Conversion rate (CVR) is the share of visitors who complete the action you care about. For a store that is a purchase. For a SaaS company it is a trial signup. For a lead-generation business it is a submitted form.

Conversion rate = (Conversions ÷ Visitors) × 100

Simple arithmetic — but three definitional choices decide whether the number you get is useful or actively misleading.

The denominator. Sessions, users, and visitors are three different counts. One person who visits four times in a week is four sessions and one user. Session-based CVR is always lower than user-based CVR, often by 30-40%. Neither is wrong, but comparing your session-based 2.25% against a competitor's user-based 3.4% tells you nothing except that you used different denominators.

The scope. Site-wide CVR blends branded search — which converts brilliantly, because those people already decided — with cold paid traffic, which does not. A site-wide number moves whenever your traffic mix changes, even if nothing about the site changed. Run a brand campaign and your CVR "improves." Scale prospecting and it "collapses." Neither reflects the quality of your site.

The window. Most analytics tools credit a conversion to the session it happened in, not the session that started the journey. Categories with long consideration cycles therefore systematically under-report their discovery channels and over-report the last click.

Get those three right and CVR becomes the most useful single number on the site. Get them wrong and you will spend a quarter optimizing a metric that was really measuring your traffic mix.

How to calculate conversion rate

Take a real month of data:

Input	Value
Visitors	40,000
Conversions (orders)	900
Average order value	$80.00
Gross margin	45%

Which produces:

Output	Value	How
Conversion rate	2.25%	900 ÷ 40,000 × 100
Revenue	$72,000.00	900 × $80
Revenue per visitor	$1.80	$72,000 ÷ 40,000
Visitors per conversion	44.44	1 ÷ 0.0225

Two of those deserve more attention than the headline rate.

Revenue per visitor (RPV) is conversion rate and average order value fused into one number. It is the better metric to optimize, because it cannot be gamed by the classic trap of "improving" conversion rate through discounting. Slash prices 25% and your CVR will rise while RPV falls. RPV catches that; CVR alone does not.

Visitors per conversion — 44.44 here — is the figure to keep in your head. It reframes an abstract percentage as something physical: roughly one order arrives for every 44 or 45 people who show up. When someone proposes a change that "should only cost us a couple of conversions," this is the number that tells you how many visitors that really is.

The Conversion Rate Calculator computes all of this and prices the lift in the same pass.

What is a good conversion rate?

The honest answer is that it depends almost entirely on what you sell and how much it costs. Here are the ranges most businesses land in:

Sector	Typical range	Rough median
E-commerce, all categories	1.5% – 3.5%	2.2%
Fashion and apparel	1.5% – 2.8%	2.0%
Health and beauty	2.5% – 4.5%	3.3%
Food and beverage	3.0% – 5.5%	4.0%
Consumer electronics	1.0% – 2.0%	1.4%
Home and furniture	0.8% – 1.8%	1.2%
Luxury and high-ticket	0.4% – 1.2%	0.7%
B2B SaaS free trial signup	2% – 5%	3%
B2B demo request form	1% – 3%	2%
Branded search landing page	8% – 20%	12%

At 2.25%, the example site sits mid-pack for general e-commerce — unremarkable for beauty, strong for furniture, and outstanding for anything above a $2,000 price point.

Notice the pattern down the table: conversion rate falls as price and consideration time rise. A $12 impulse purchase and a $4,000 sofa are not competing on the same scale. This is also why a low CVR is not automatically a problem. A furniture retailer at 1.2% with a $1,400 average order value earns $16.80 per visitor. Our example store at 2.25% earns $1.80. The furniture site converts half as often and is worth nearly ten times as much per visit.

Why the "average conversion rate" is a bad target

The industry average is a statistic about other companies. It is a useful orientation and a terrible goal, for four reasons.

It blends incomparable businesses. A published "2.2% e-commerce average" is a median across price points, categories, geographies and traffic mixes that have nothing to do with each other. Half the sites in that dataset would go bankrupt at your unit economics.

It ignores traffic quality. The fastest way to hit any average is to buy narrower, higher-intent traffic. Bid only on your brand name and watch CVR triple while revenue stays flat. You have not improved anything; you have changed who you counted.

It sets a ceiling as well as a floor. Teams that hit the average stop. The competitive advantage lives in the fourth quartile, not the median.

It is not denominated in money. "Average" does not tell you whether the next point of conversion rate is worth a two-week engineering sprint. Your own numbers do.

Replace the benchmark question with a better one: what is the next 0.25 points worth to us, and what would it cost to get? That is answerable, and the answer is in the next section.

What is a conversion rate lift actually worth?

Hold traffic at 40,000 visitors, average order value at $80 and gross margin at 45%, and price each target rate:

Target CVR	Conversions	Extra conversions	Extra revenue	Extra gross profit	Lift needed
2.25% (today)	900	0	$0	$0	0%
2.50%	1,000	100	$8,000	$3,600	11.11%
2.75%	1,100	200	$16,000	$7,200	22.22%
3.00%	1,200	300	$24,000	$10,800	33.33%
3.50%	1,400	500	$40,000	$18,000	55.56%
4.00%	1,600	700	$56,000	$25,200	77.78%

A quarter of one percentage point — from 2.25% to 2.50% — is worth $3,600 of gross profit a month, $43,200 a year. That reframes the build-versus-buy question entirely. A $3,000 checkout redesign that delivers a quarter point pays for itself inside a month and then keeps paying, every month, forever, with no marginal cost.

Two cautions on that table. First, the relative lift column climbs faster than intuition suggests: going from 2.25% to 3.00% sounds like "three quarters of a point" but is a 33.33% relative improvement, which is a serious program of work, not one A/B test. Second, these gains only hold if traffic quality holds. Doubling spend to hit the same conversion rate on colder traffic is a different and much harder problem — model that with the CAC Calculator.

How conversion rate changes your CPA, CPC and ROAS

This is the compounding effect, and it is the reason conversion work outranks almost everything else on the roadmap.

Your cost per acquisition is your cost per click divided by your conversion rate. So at a fixed $1.20 CPC, the same clicks produce wildly different acquisition costs:

Conversion rate	CPA at $1.20 CPC	Max CPC at a $36.00 CPA cap
1.00%	$120.00	$0.36
1.50%	$80.00	$0.54
2.00%	$60.00	$0.72
2.25%	$53.33	$0.81
3.00%	$40.00	$1.08
4.00%	$30.00	$1.44
5.00%	$24.00	$1.80

The right-hand column is the strategic one. Gross profit per order here is 45% of $80 = $36.00, which is the most you can pay to acquire an order and still break even. At 2.25% conversion you can afford $0.81 a click. At 3.00% you can afford $1.08 — a 33% higher bid than every competitor stuck at 2.25%, funded entirely by your own site.

That is how conversion rate becomes a moat. The advertiser who converts better can outbid everyone in the auction, win the better placements, and still make money. Work the chain through with the CPA Calculator, the CPC Calculator, and the Breakeven ROAS Calculator, which turns that $36.00 gross profit into the exact ROAS floor your campaigns must clear.

The same logic runs upstream. CPM buys impressions, CTR turns impressions into clicks, conversion rate turns clicks into customers, and LTV decides how much that customer was worth in the first place. A weak link anywhere breaks the chain, but conversion rate is the only link that improves both your costs and your revenue at once.

Why mobile conversion rates look broken

Mobile is usually the majority of traffic and the minority of revenue. A typical split:

Device	Share of sessions	Sessions	Typical CVR	Conversions
Mobile	60%	24,000	1.90%	456
Desktop	35%	14,000	3.00%	420
Tablet	5%	2,000	2.40%	48
Blended	100%	40,000	2.31%	924

Mobile converts at roughly 63% of the desktop rate, which is normal and mostly structural: smaller screens, more interruption, more browsing-with-intent-to-buy-later, and payment friction that desktop does not have.

It is also the single largest pool of recoverable money on most sites, because it holds the most traffic at the worst rate. Lift mobile alone from 1.90% to 2.40% — half a percentage point, well inside what wallet payments and a shorter form typically deliver — and you gain 24,000 × 0.5% = 120 conversions. Blended CVR moves from 2.31% to 2.61%, worth $9,600 in revenue and $4,320 in gross profit every month. Nothing about desktop changed.

The practical lesson: never optimize a blended conversion rate. Segment by device first, then by channel, then by new versus returning. The blend hides both your worst problem and your best opportunity.

Micro conversions vs macro conversions

The purchase is your macro conversion — the one that pays. But it is a terrible diagnostic, because a single number tells you that something is wrong without telling you where. Micro conversions are the intermediate steps that locate the leak:

Funnel step	Visitors reaching	Step rate	Cumulative rate
Session start	40,000	—	100.0%
Product page view	22,000	55.00%	55.0%
Add to cart	4,400	20.00%	11.0%
Checkout started	2,000	45.45%	5.0%
Order placed	900	45.00%	2.25%

Now the 2.25% has an anatomy. Only 20% of product page viewers add to cart, and only 45% of people who start checkout finish it. That second number is the emergency. These are people who have chosen a product, entered a checkout, and then left — the highest-intent visitors on the site.

Take checkout completion from 45% to 55% and the same 2,000 checkout starts produce 1,100 orders instead of 900. Site CVR goes to 2.75%, worth $16,000 in monthly revenue and $7,200 in gross profit — the 2.75% row of the pricing table above, earned by fixing one screen rather than the whole site.

Worth tracking as micro conversions: product page views, add to cart, checkout start, each checkout step completed, email capture, and account creation. Each one is a rate you can move independently, and the step with the lowest rate against the highest intent is always where you start.

How much traffic do you need to A/B test a change?

This is where most conversion programs quietly fail. At a 2.25% baseline, here is roughly the traffic each variant needs to detect a lift at 95% confidence and 80% power, two-sided:

Relative lift to detect	Visitors per variant	Total test traffic	Time at 40,000/month
5% (2.25% → 2.36%)	278,000	556,000	13.9 months
10% (2.25% → 2.48%)	69,500	139,000	3.5 months
20% (2.25% → 2.70%)	17,400	34,800	26 days
30% (2.25% → 2.93%)	7,700	15,400	12 days
50% (2.25% → 3.38%)	2,800	5,600	4 days

Read that table twice, because it contains the most expensive lesson in conversion optimization: a site with 40,000 monthly visitors cannot reliably detect a 10% lift in under three months. Most A/B tests run for two weeks. Most changes produce lifts well under 20%. The arithmetic simply does not support the practice.

There are three honest responses. Test bigger changes — redesign the checkout, do not move the button four pixels. Test higher up the funnel, where the rates are larger and therefore need less traffic to move measurably; add-to-cart at 20% needs a fraction of the sample that a 2.25% purchase rate does. Or stop testing and start shipping — below roughly 25,000 monthly visitors, spend the time on qualitative research, session recordings, and obviously correct fixes instead of underpowered experiments.

Why calling an A/B test early loses money

Every conversion rate you measure is a sample, and samples wander. In the first days of a test, variant B routinely leads by 30% for reasons that are entirely random. Stop there and you will "discover" a winner that does not exist.

Two specific errors cost the most:

Peeking. Checking a test daily and stopping the moment it crosses significance inflates your false positive rate from the nominal 5% to somewhere between 20% and 30%. You are effectively running the test many times and keeping the run you liked. If you need to monitor continuously, use a sequential testing method designed for it, not a fixed-horizon p-value you glance at every morning.

Ignoring business cycles. Weekday and weekend traffic convert differently, as do paydays and the days around a promotion. Always run in whole weeks, and never conclude a test that spans a sale, a holiday, or a major campaign launch.

The financial damage is not the wasted test. It is that a false winner gets shipped, gets believed, and then anchors the next year of decisions on a result that was noise. A test that returns "no detectable difference" after a full run is a genuinely useful outcome. A test stopped on day three is not an outcome at all.

The highest-leverage conversion fixes

In rough order of return per hour of work, for a typical store:

1. Checkout friction. The single biggest pool. Offer guest checkout, drop every optional field, show total cost including shipping before the final step, and add wallet payments (Apple Pay, Google Pay, Shop Pay). Wallets alone routinely add 10-20% to mobile checkout completion because they remove typing entirely.

2. Unexpected costs. Shipping cost revealed at the last step is the most-cited reason for abandonment in every study ever run. Show it early, or fold it into the price and advertise free shipping — which usually wins even at an identical total. Price that trade-off honestly with the Profit Margin Calculator before you commit.

3. Page speed. Every additional second of load time costs conversions, and the effect is worst on mobile where the traffic is. This is one of the few fixes with a reliable, mechanical payoff.

4. Product page evidence. More and better images, real reviews, clear sizing and specifications, visible return policy. Most add-to-cart failures are unanswered questions, not price objections.

5. Trust signals at the decision point. Security badges, a plainly worded returns promise, and real contact details placed in the checkout, not buried in the footer.

6. Mobile-specific fixes. Correct input types for numeric fields, tap targets that are actually tappable, sticky add-to-cart, and a form that does not require a zoom to read.

Notice what is absent: button colors, hero copy tweaks, and popup timing. Those are the tests that get run because they are cheap, and they are cheap because they change almost nothing.

Common mistakes

Mixing denominators between reports. Sessions in one dashboard, users in another, and a benchmark article using a third. Pick one definition, write it down, and use it everywhere. Most "our conversion rate dropped" panics turn out to be a measurement change.

Improving CVR by discounting. A 25% off sitewide sale will lift conversion rate and shrink gross profit. If you optimize CVR without watching revenue per visitor and margin together, you will discount your way to a beautiful dashboard and a worse business. See how to price products for profit.

Optimizing the blend instead of the segments. A blended 2.31% that hides mobile at 1.90% will send you tuning desktop, where the opportunity is smallest. Segment before you decide anything.

Calling tests early, or running them underpowered. Related but distinct failures: one manufactures winners that are not real, the other guarantees you learn nothing. Both feel productive.

Excluding fees from the value of a lift. Extra conversions arrive with extra platform and payment fees attached. On a marketplace those can be 15% or more of each order, so the profit from a lift is meaningfully smaller than the revenue suggests — check the real number with the Marketplace Fee Comparison or the hidden costs that eat e-commerce profit.

Treating conversion rate as a site metric rather than a channel metric. Email converts at several times the rate of cold social. A shift in channel mix will move site CVR without anyone touching the site. Always read the number next to its traffic source.

Related calculators

• CPA Calculator — what each conversion costs you to acquire, and how CVR moves it
• CPC Calculator — the maximum click price your conversion rate can support
• Breakeven ROAS Calculator — the ROAS floor your unit economics set
• ROAS Calculator — what your campaigns are actually returning against that floor
• CTR Calculator — the step above conversion rate in the same funnel
• LTV Calculator — what a converted customer is worth over their whole life
• Profit Margin Calculator — the gross margin that turns extra revenue into extra profit