Advertising Metrics Explained: CPM, CPC, CPA, ROAS and Attribution

Why you need to understand the metrics

Agency reports are full of abbreviations. If you cannot read them, you cannot tell whether your money is working, and you depend entirely on the agency's interpretation. You do not need to become an analyst, but a handful of terms cover most conversations. This guide defines them, shows how they are calculated, explains what each leaves out and suggests questions that keep reports honest.

All formulas below are standard definitions. Any figures are invented examples to show the arithmetic, not benchmarks of what is good.

Impressions, reach and frequency

An impression is one display of an ad, whether or not anyone notices it. Reach is the number of different people who saw the ad at least once. Frequency is the average number of times each reached person saw it. If an ad is shown a thousand times to two hundred people, reach is two hundred and average frequency is five.

Limits: impressions do not prove attention. Platforms define a countable impression differently, and reach is often estimated. Ask how each is measured and whether viewability information is available.

CPM: cost per thousand impressions

CPM stands for cost per mille, mille meaning thousand. It equals the total cost divided by impressions, multiplied by one thousand. If a campaign costs five hundred and delivers one hundred thousand impressions, the CPM is five.

Use: comparing the price of reach across channels or publishers, mainly for awareness goals. Limits: cheap impressions are not valuable if the audience is wrong. Combine CPM with measures of audience fit and outcomes.

CTR and CPC

Click-through rate (CTR) is clicks divided by impressions, usually shown as a percentage. Cost per click (CPC) is total cost divided by clicks. If an ad is shown one thousand times and clicked twenty times, CTR is two percent. If those clicks cost forty, CPC is two.

Use: judging how appealing ads are and how expensive traffic is. Limits: a click is not a customer. A high CTR with a confusing landing page wastes money. Always read CTR and CPC next to what visitors do afterwards.

Conversion rate

Conversion rate is the percentage of visitors who complete a goal such as buying, subscribing or requesting a quote. If one thousand visitors produce thirty purchases, conversion rate is three percent. It depends on audience quality, offer, price, trust and page experience, so comparing your rate with another company's can mislead. Track it against your own history and across segments.

CPA: cost per acquisition

CPA, or cost per acquisition, is total cost divided by the number of acquisitions. An acquisition might be a sale, a lead or an app install; define it. If spending three hundred produces ten customers, CPA is thirty. Compare CPA with what a customer is worth. A CPA that looks high may still be profitable if customers return often; a low CPA is not good if the customers are poor quality.

ROAS and ROI

ROAS, return on ad spend, is revenue attributed to ads divided by ad spend. If a campaign spends one hundred and generates four hundred in revenue, ROAS is four, often written as four to one or four hundred percent.

ROAS ignores other costs, such as product cost, shipping, discounts, staff and agency fees. A ROAS of four might be unprofitable if margins are thin. To find your break-even ROAS, divide one by your gross margin expressed as a fraction. With a margin of fifty percent, break-even is two. ROI, return on investment, compares profit (or net gain) with total cost. Be sure you know which one a report uses.

Attribution: who gets credit?

Customers often see several touchpoints before buying: a video, a search, an email. Attribution assigns credit among them. Last-click attribution gives everything to the final click and tends to favor channels near the purchase. Multi-touch models divide credit among several steps. Each model embeds assumptions, and tracking limits imposed by privacy tools and consent rules mean any model sees only part of the picture.

Ask which model and window the report uses, whether the same settings are used month to month, and how platform-reported results compare with your own sales data.

Incrementality: did ads cause it?

Reported conversions include people who would have bought anyway, particularly when ads target existing customers or branded searches. Incrementality asks what extra results the advertising caused. Experiments, such as holding out a random group from seeing ads or pausing in one region, provide stronger evidence than dashboards alone. They take planning, but they are the best protection against paying for credit that advertising did not earn.

Metrics hierarchy: from activity to outcomes

Think in three layers. Activity metrics (impressions, clicks, posts) show that work happened. Efficiency metrics (CPM, CPC, CTR, CPA) show how economical it was. Outcome metrics (revenue, margin, retained customers, qualified leads) show whether the business benefited. Reports should connect all three, with the most space for outcomes.

Questions to ask in every reporting meeting

• Which metric matters most for our goal, and why?
• How is it defined, and from which system does it come?
• What changed since last period, and what caused it?
• What did you test, and what did you learn?
• Which numbers are estimates or modeled?
• What would you do differently next month?
• Can we see the raw platform data ourselves?

Key takeaways

Know the formulas, but also the blind spots. CPM measures price of exposure, CTR and CPC measure interest and traffic cost, CPA and ROAS tie spending to results, and attribution and incrementality decide how much credit advertising deserves. Insist on definitions, consistent settings and a story connecting numbers to business goals.

Frequently asked questions