Your ads bring in four dollars for every dollar spent. That sounds like a reason to increase the budget. But if the orders leave only twenty cents of every revenue dollar available to pay for advertising, the campaign is already spending more than it contributes.
The paid marketing metrics that matter most connect ad spend to a defined outcome, the cost of delivering it, and the money left afterward. CTR and CPC help explain traffic. Conversion rate and CPA reveal the cost of an action. ROAS, customer acquisition cost, and contribution margin put those results in business context.
Drawing on the measurement and review approach in Mastering Paid Marketing Metrics, this guide explains the formulas, works through an original campaign comparison, and gives you a review process to use before increasing spend.
In this guide
Define the outcome before choosing the metric
Write down what counts as success for the campaign. For an online store, that may be a completed purchase. For a service business, it may be a qualified inquiry that eventually becomes a paying client.
Keep these actions separate in reporting. An add-to-cart event, a newsletter signup, and a purchase represent different outcomes. A lower cost per conversion means little if the mix has shifted toward easier actions.
Also specify the reporting period, currency, attribution window, and revenue definition. Record whether sales values include tax, discounts, shipping, and refunds. A consistent definition makes comparisons possible.
The core paid marketing metrics and their formulas
Use the same campaign, time period, and event definitions across each calculation. In the examples below, a conversion means a purchase unless stated otherwise.
CTR: how often an impression becomes a click
Click-through rate = clicks ÷ impressions × 100. If an ad receives 2,000 clicks from 100,000 impressions, its CTR is 2%.
CTR helps you examine whether the message attracts attention in that placement. Compare similar formats and audiences. A strong CTR becomes useful to the business when those clicks lead to appropriate next actions.
CPC: what you pay for a click
Average cost per click = ad spend ÷ clicks. Spending $1,000 for 2,000 clicks gives an average CPC of $0.50.
Cheaper clicks can still produce expensive sales if few visitors buy. Read CPC alongside conversion rate and the value of the resulting orders.
Conversion rate: how traffic becomes an outcome
Click-to-purchase conversion rate = purchases ÷ clicks × 100. In this simplified example, 100 purchases from 2,000 clicks gives a 5% rate.
Check the denominator in your reporting tool. A website may calculate conversions per session, while Google Ads uses eligible ad interactions for its conversion-rate reporting. Those percentages are not automatically comparable.
CPA: what a specified action costs
Cost per action = ad spend ÷ attributed actions. With $1,000 in spend and 100 attributed purchases, purchase CPA is $10.
Name the action every time: cost per purchase, cost per lead, or cost per qualified lead. If you sell services, carry the analysis through to accepted leads and customers so an inexpensive form submission does not hide poor lead quality.
ROAS: attributed revenue for each advertising dollar
Revenue ROAS = attributed revenue ÷ ad spend. Revenue of $4,000 against $1,000 in spend gives 4× ROAS, also written as 400%.
Confirm what “value” represents in the platform. Google Ads conversion values can represent revenue or another business value. A manually assigned lead value is an estimate, not cash received.
CAC: the cost of acquiring a new customer
Customer acquisition cost = acquisition-related sales and marketing costs ÷ new customers acquired. Shopify’s customer acquisition guide explains this broader cost view.
Include the relevant creative, agency, staff, and software costs in your chosen scope. For example, $1,500 of acquisition costs divided by 50 new customers gives $30 CAC. Keep the cost scope and customer cohort aligned, allowing for the time it takes to turn a lead into a customer.
Purchase CPA and CAC answer different questions: purchases may include repeat orders, while CAC counts new customers and includes costs beyond media spend.
Why a higher ROAS can leave less money
Consider two hypothetical ecommerce campaigns promoting products with different margins. These figures illustrate the calculations; they are not VISIBLE.BLOG campaign results.
For this comparison, attributed net revenue is after discounts and refunds, excludes sales tax, and uses a consistent attribution basis. The contribution margin before advertising is what remains after product and other variable order costs, including payment fees and fulfillment.
| Metric | Campaign A | Campaign B |
|---|---|---|
| Ad spend | $1,000 | $1,000 |
| Clicks | 2,000 | 1,000 |
| Purchases | 100 | 60 |
| Net revenue | $4,000 | $3,000 |
| Average CPC | $0.50 | $1.00 |
| Purchase CPA | $10.00 | $16.67 |
| Revenue ROAS | 4× | 3× |
| Margin before ads | 20% | 45% |
| Contribution before ads | $800 | $1,350 |
| Contribution after ads | −$200 | $350 |
Campaign A wins on CPC, purchase CPA, and ROAS. Yet its $4,000 of revenue leaves $800 before advertising. Paying $1,000 for those orders produces a $200 shortfall.
Campaign B brings in less revenue and costs more per purchase. Its higher-margin orders leave $1,350 before advertising and $350 afterward. That $350 still needs to help cover fixed overhead and any other costs excluded from the calculation.
ROAS measures a revenue-to-spend relationship. Profit analysis also requires costs, which is why Google’s ROI guidance considers the money spent to produce the result. A campaign leaderboard based on ROAS alone misses this distinction.
Calculate your break-even ROAS before setting a target
When your ROAS numerator is revenue and you know the contribution margin before advertising, you can calculate the point at which that contribution covers ad spend:
Break-even ROAS = 1 ÷ contribution margin rate before ads
At a 20% margin: 1 ÷ 0.20 = 5× ROAS.
At a 45% margin: 1 ÷ 0.45 ≈ 2.22× ROAS.
This is the break-even point for the costs included in that model. Set a higher target if the campaign must contribute toward overhead, additional acquisition costs, and profit. Recalculate when discounts, product mix, fees, or refund rates change.
You can translate the same logic into purchase CPA. If an average order produces $40 of net revenue at a 20% contribution margin before ads, it leaves $8 available for advertising. A $10 purchase CPA exceeds that amount. A target below $8 would leave some contribution for other costs.
For a subscription or repeat-purchase business, evaluate later purchases using contribution from observed customer cohorts and the time needed to recover acquisition costs. Keep an immediate cash result separate from a forecast of future value.
Check the numbers before changing the budget
A precise formula cannot correct an unreliable input. Before making a spending decision, check the following:
- Event accuracy: verify that a purchase is recorded once and carries the correct value and currency. Check for duplicate events and test orders.
- Revenue reconciliation: compare the attributed orders you can identify with store records. Investigate differences instead of expecting platform totals to match automatically.
- Conversion delay: allow enough time for recent traffic to produce purchases. Google’s conversion-lag documentation explains why recent CPA may look higher and ROAS lower while conversions are still arriving.
- Attribution scope: document click and view attribution rules. Multiple platforms can claim credit for the same sale, so adding their reported revenue can double-count it.
- Customer mix: distinguish new-customer acquisition from repeat purchasing and remarketing. These activities solve different business problems.
Attributed revenue is also different from incremental revenue: the latter is the extra revenue caused by advertising. A platform report does not, by itself, show which customers would have purchased anyway. Where practical, use a well-designed holdout or lift test to investigate that question.
Use the pattern of metrics to find the next action
Once the data is credible, inspect where performance changed. These patterns are starting points for investigation, not automatic diagnoses:
- Clicks fall while conversion rate holds: examine delivery, competition, targeting, and creative relevance before rebuilding checkout.
- Clicks arrive but purchase conversion falls: inspect landing-page relevance, mobile usability, pricing, availability, and checkout errors.
- CPA holds but ROAS falls: check average order value, discounts, and product mix.
- ROAS holds but contribution falls: examine margin changes, refunds, and fulfillment or payment costs.
If the gap is between the ad’s promise and the page’s opening, our SEO storytelling guide includes a worked example of making a generic introduction more specific and useful.
A campaign review you can repeat
Set a regular review appointment. A weekly review can organize the work, while the period you analyze should reflect conversion delay, sales cycles, and the amount of data available. A handful of purchases is a weak basis for a large budget change.
- Confirm the inputs. Record spend, the conversion event, attributed revenue, new customers, margin assumptions, and reporting window.
- Compare like with like. Start with your own comparable campaigns and periods. Note changes in audience, offers, channel, or season.
- Check the economic threshold. Compare purchase CPA and ROAS with the contribution those orders can support.
- Choose one testable hypothesis. For example: mobile visitors cannot find the delivery details, which may be reducing checkout completion.
- Document the decision. Record the proposed change, its success metric, its spending limit, and when enough evidence should be available to review it.
When considering a budget increase, ask what the additional spend is likely to produce. Historical average ROAS does not establish the return on the next advertising dollar. Google describes this distinction in its guidance on marginal returns. Test increases within a defined spending limit and watch the additional contribution, not just total revenue.
Build a report that ends with a decision
Start with one campaign. Define its outcome, verify its costs and revenue, calculate the contribution left after ad spend, and write down the next question to investigate. That gives each metric a purpose and turns a dashboard into a working basis for decisions.
7 complementary resources
Mastering Paid Marketing Metrics
Put the numbers to work with the ebook, benchmarking and ROAS guides, two practical checklists, an optimization prompt pack, and the From Clicks to Cashflow audio series. Build a more consistent campaign review process.