Click-through rate
Click-through rate is clicks divided by impressions. Of 250,000 impressions, 3,750 clicks is 1.5 percent. It shows whether an ad draws attention.
A high rate on the wrong audience wastes money, so it is a first signal and not a goal. Compare ads shown to the same audience in the same place.
Cost per click and cost per thousand impressions
Cost per click is spend divided by clicks, and cost per thousand impressions is spend divided by impressions times 1,000. On $4,200 of spend, the click cost is $1.12 and the impression cost is $16.80. The first prices attention and the second prices reach.
Which one you care about depends on whether you want visits or awareness.
Conversion rate
Conversion rate is the share of visitors who do what you want, such as buy or sign up. Of 12,000 visitors, 360 converting is 3 percent. Rates differ widely by goal and traffic source, so compare your rate with your own history and with the same kind of goal.
A rise or fall matters most when it is measured over enough visitors.
Acquisition cost needs the conversion definition
Assume $4,200 advertising spend and 360 attributed acquisitions in the same period. Ad-only cost per acquisition is $11.6667. Counting newsletter signups instead of paying customers changes the event and therefore the meaning.
A full customer-acquisition cost can also include sales compensation, tools and other costs excluded from this ad-only measure.
Revenue return does not settle profitability
Assume attributed revenue $20,000 and spend $5,000: ROAS is 4. At an assumed contribution margin of 30% before advertising, revenue contributes $6,000; the advertising leaves $1,000 before remaining fixed costs. Gross margin can omit other variable selling costs, so it is not automatically the right contribution rate.
Attribution and incrementality also differ: recorded attributed sales need not all have been caused by the campaign.
Return on ad spend
Return on ad spend divides revenue attributed to ads by the ad spend. A ratio of 4 means $4 of revenue for each dollar spent. It ignores costs other than the ads, so a high ratio does not guarantee profit.
Pair it with the margin on what is sold to see whether a campaign makes money.
Customer lifetime value and payback
Customer lifetime value estimates what a customer brings over the whole relationship. Comparing it with the cost to acquire a customer shows whether growth pays. Payback time is how long the margin from a customer takes to repay the acquisition cost.
A short payback lets you reinvest sooner.
Testing changes fairly
Changing an ad or a page and seeing a rise does not prove the change caused it. An A/B test splits visitors between two versions and tests whether the gap is bigger than chance. Fix the sample size in advance, run for whole weeks and do not stop the moment the result looks good.
Optimizing the right number
Any metric can be gamed by pushing it and ignoring the others. A low cost per click can come from low-quality traffic, and a high conversion rate can come from a discount that erodes margin. Choose the metric that ties most closely to profit, and watch the others as guardrails.
Assumed inputs and calculated outputs
Marketing Metrics Calculator: input sensitivity
| Impressions | Click-through rate | Cost per click | Cost per 1,000 impressions | Cost per conversion |
|---|---|---|---|---|
| 200,000 | 1.9% | $1.12 | $21 | $28 |
| 225,000 | 1.7% | $1.12 | $19 | $28 |
| 250,000 | 1.5% | $1.12 | $17 | $28 |
| 275,000 | 1.4% | $1.12 | $15 | $28 |
| 300,000 | 1.3% | $1.12 | $14 | $28 |
Open the Marketing Metrics Calculator to use your own numbers →
What goes wrong
Marketing metrics: conversion rate, CTR, CPC, CPA and ROAS: common errors
| The mistake | Why it misleads | Do this instead |
|---|---|---|
| Optimizing click-through rate alone | A high rate on the wrong audience pays for visits that never turn into customers. | Watch cost per acquisition and revenue per visitor as well. |
| Comparing conversion rates across different goals | A signup and a purchase have very different typical rates. | Compare like with like over the same period. |
| Reading a high ROAS as profit | It counts revenue against ad spend only, and ignores the cost of the goods and other costs. | Compare it with the margin to see whether the campaign pays. |
| Stopping a test as soon as it looks significant | Peeking raises the false positive rate above the stated level. | Fix the sample size or the stopping rule in advance. |
Marketing metrics: conversion rate, CTR, CPC, CPA and ROAS: worked calculation
Using the Marketing Metrics Calculator worked example, moving Impressions from 200,000 to 300,000 changes the Click-through rate from 1.3% to 1.9%.
| Input | Value | Definition |
|---|---|---|
| Impressions | 250,000 | Entered example assumption. |
| Clicks | 3,750 | Entered example assumption. |
| Ad spend | $4,200 | Entered example assumption. |
| Conversions | 150 | Entered example assumption. |
| Click-through rate | 1.5% |
|---|---|
| Cost per click | $1.12 |
| Cost per 1,000 impressions | $17 |
| Cost per conversion | $28 |
| Impressions | Click-through rate | Cost per click | Cost per 1,000 impressions | Cost per conversion |
|---|---|---|---|---|
| 200,000 | 1.9% | $1.12 | $21 | $28 |
| 225,000 | 1.7% | $1.12 | $19 | $28 |
| 250,000 | 1.5% | $1.12 | $17 | $28 |
| 275,000 | 1.4% | $1.12 | $15 | $28 |
| 300,000 | 1.3% | $1.12 | $14 | $28 |
Which input moves the click-through rate most in the marketing metrics calculator?
Ranked by how far each moves the click-through rate across the range tested: impressions (0.303%, 20%) and clicks (0.300%, 20%). Ad spend and conversions do not change it.
If I double impressions in the marketing metrics calculator, does the click-through rate double?
Doubling it from 250,000 to 500,000 takes the click-through rate from 1.5% to 0.750%, which is 0.50 times the worked-example figure. So it falls instead of rising. Halving it to 125,000 gives 3.0%.
Marketing metrics: conversion rate, CTR, CPC, CPA and ROAS: points to carry forward
- Marketing metrics are ratios of counts, so keep the period and audience the same on both sides.
- CTR shows attention, CPC and CPM price it, and cost per acquisition ties spend to results.
- Revenue per visitor is the best figure for comparing pages and campaigns.
- Return on ad spend ignores costs other than the ads.
- Test changes with a fixed sample size and do not stop at the first good-looking result.
What is a good click-through rate?
It depends on the channel and audience. Compare ads with each other and with your own past results, not with a single benchmark.
How do I calculate cost per acquisition?
Divide the total ad spend by the number of conversions. Use the same reporting period and attribution definition for the spend and conversions; otherwise the numerator and denominator describe different activity.
What is the difference between CPC and CPM?
Cost per click prices each click. Cost per mille prices each thousand impressions, so it measures the cost of reach.
How do I calculate conversion rate?
Divide conversions by visitors and multiply by 100.
What does ROAS mean?
Return on ad spend, the revenue attributed to ads divided by the ad spend. A ratio of 4 is $4 of revenue per dollar spent.
How do I know if an A/B test result is real?
Use a significance test on the two rates and a sample size fixed beforehand. The A/B test calculator gives the z-score and the p-value.
How do I compare LTV and CAC?
Divide the customer lifetime value by the acquisition cost. Higher ratios and shorter payback times mean growth is paying for itself.
What is a good ROAS?
It depends on your margin. A ratio of 4 can be profitable for high-margin goods and a loss for low-margin ones, so compare it with the margin on what you sell.
What is customer acquisition cost?
The total sales and marketing spend to win customers divided by the number of customers won in the same period.
How long should a CAC payback period be?
Shorter is better, since it lets you reinvest sooner. Many businesses aim for a payback within a year, though the right length depends on cash and margins.
Why do my click numbers differ between tools?
Tools count clicks and sessions differently, and they filter bots and repeat clicks in their own ways. Use one source consistently for a trend.