CPM Marketing: What You’re Buying When You Pay Per Thousand

CPM marketing means paying for every thousand impressions your ad receives, regardless of whether anyone clicks, converts, or even notices it. It is the oldest pricing model in digital advertising, and it remains one of the most misunderstood.

Done well, CPM campaigns build brand presence at scale and support demand generation in ways that click-based models cannot. Done badly, they burn budget on impressions that do nothing, delivered to audiences that were never going to buy.

The model itself is not the problem. How most marketers use it is.

What Does CPM Actually Mean?

CPM stands for cost per mille, with mille being Latin for thousand. It refers to the cost of serving your ad one thousand times. If your CPM is £5, you pay £5 for every thousand impressions your ad generates.

It is worth being precise here because there is a common conflation between CPM as a pricing mechanism and CPM as a campaign goal. They are not the same thing. You can buy on a CPM basis while optimising for reach, frequency, brand recall, or even conversions. The CPM rate tells you what you are paying per unit of exposure. It says nothing about whether that exposure is worth buying.

The model predates digital advertising by decades. Print media sold on CPM. Television sold on CPM. When online display advertising emerged in the late 1990s and early 2000s, CPM was the natural pricing framework to import. Online ad spending patterns from that era show just how dominant impression-based buying was before performance models took hold.

CPC, or cost per click, emerged as an alternative partly because it felt more accountable. You only pay when someone engages. But that shift created its own distortions, which I will come back to.

How CPM Fits Into the Paid Advertising Landscape

CPM buying is most commonly associated with display advertising, programmatic, video, and social media placements where reach and visibility are the primary objectives. If you are running a brand awareness campaign on YouTube, a programmatic display campaign across a content network, or a broad Facebook reach campaign, you are almost certainly buying on CPM terms.

If you want to understand how CPM sits alongside other paid channels and buying models, the paid advertising hub covers the broader landscape in detail.

The practical reality is that most digital advertising platforms now blend pricing models depending on campaign objective. Meta will optimise your CPM buying toward conversions if you tell it to. Google Display will shift toward viewable impressions or clicks depending on your settings. The CPM rate you see in a dashboard is often the output of an auction, not a fixed price you negotiate upfront.

That auction dynamic matters because it means your effective CPM is partly a function of how competitive your audience segment is. Targeting CFOs in financial services costs more per thousand than targeting a broad interest category. That is not a bug. It reflects genuine scarcity of the right eyeballs.

When CPM Buying Makes Commercial Sense

There are specific contexts where CPM is the right buying model. There are others where it is the wrong one and where marketers default to it out of habit or because their platform defaults pushed them there.

CPM makes sense when your objective is reach and frequency. If you are launching a new product, entering a new market, or trying to shift brand perception, you need a lot of people to see your message multiple times. Paying per click in that context is inefficient because you are not trying to generate clicks. You are trying to build mental availability.

CPM also makes sense for retargeting. When you are serving ads to people who have already visited your website or engaged with your brand, retargeting campaigns benefit from impression-based buying because you want consistent visibility across a defined audience rather than clicks from an unknown pool. The audience is already warm. Your job is to stay present.

And CPM makes sense at the top of the funnel for demand generation, where the goal is to make people aware that a problem exists and that your brand solves it. Demand generation is a long game, and CPM buying supports it by building reach over time rather than chasing immediate response.

Where CPM goes wrong is when it is used as a proxy for performance. I have seen this repeatedly in agencies: a campaign is bought on CPM, generates millions of impressions, and the report leads with impression volume as though that is the outcome. Impressions are not outcomes. They are opportunities for outcomes. The distinction matters enormously when you are defending budget in a board meeting.

The Difference Between a Low CPM and an Efficient CPM

One of the more persistent myths in digital advertising is that a low CPM is a sign of good buying. It is not. It is often a sign of bad inventory.

When I was running agency operations and reviewing media plans, I would sometimes see programmatic campaigns boasting CPMs of £0.50 or £1. On paper, that looks like extraordinary value. In practice, it usually meant the ads were appearing on low-quality content sites, in non-viewable placements below the fold, or in environments where brand safety was questionable. The cheap impressions were cheap for a reason.

Premium publishers command higher CPMs because their audiences are more valuable and their placements are more visible. A £15 CPM on a respected news site, served above the fold to a verified audience, is almost always better value than a £1 CPM scattered across a long tail of unknown domains.

The metric that matters more than raw CPM is cost per viewable impression, or better still, cost per outcome. If you are running awareness campaigns, track brand search lift or direct traffic. If you are running retargeting on CPM, track conversion rate among the retargeted audience versus a control group. The CPM rate is just an input. What you get out of it is what counts.

Return on ad spend is the commercial lens that should sit behind every CPM decision. Even if you cannot attribute it perfectly, you need a directional read on whether the impressions are contributing to business outcomes.

CPM Versus CPC: The Framing That Misleads Most Marketers

The CPM versus CPC debate is one of those perennial agency conversations that often generates more heat than light. The framing itself is slightly wrong.

CPM and CPC are buying models. They are not campaign strategies. The right question is not “should I use CPM or CPC?” but “what am I trying to achieve, and which buying model gives me the most efficient path to that outcome?”

For direct response campaigns where you want clicks to a landing page, CPC buying is often more efficient because you only pay when someone takes the action you want. The risk is that platforms optimise toward clicks rather than conversions, and cheap clicks from the wrong audience are no better than expensive impressions from the right one. Landing page quality becomes critical in that context because a click that bounces is wasted spend regardless of how low the CPC was.

For upper-funnel campaigns where you want broad exposure, CPM buying gives you more control over reach and frequency. You can cap impressions per user, manage how often your message appears, and build coverage across a target audience more predictably than with click-based buying.

The smartest approach in most cases is to use both, matched to the right funnel stage. CPM at the top to build awareness and warm audiences. CPC or CPA buying lower down to capture the demand that awareness created. Most performance marketers focus almost entirely on the bottom of the funnel and then wonder why their cost per acquisition keeps rising. You cannot harvest demand you never created.

How to Set a CPM Budget That Is Defensible

One of the harder parts of CPM marketing is budget justification. With CPC or CPA campaigns, you can draw a relatively direct line between spend and outcome. With CPM campaigns, the causal chain is longer and the attribution is messier.

That does not mean CPM budgets should be set arbitrarily. It means you need a different framework for thinking about value.

Start with audience size and desired frequency. If your target audience is 500,000 people and you want to reach them with an average frequency of five impressions over a campaign period, you need 2.5 million impressions. At a £10 CPM, that is £25,000. That is your floor. Whether £25,000 is worth spending depends on what those 500,000 people are worth to your business, not on the CPM rate itself.

The second input is competitive context. In categories where brand awareness drives purchase decisions, the cost of not being visible is real even if it is hard to measure. I spent time working across retail and financial services categories where share of voice had a demonstrable relationship with share of market over time. CPM investment was not a nice-to-have. It was a structural requirement for maintaining category presence.

The third input is measurement. Before you commit CPM budget, decide how you will evaluate it. Brand lift surveys, search volume tracking, direct traffic analysis, and incrementality testing are all imperfect but useful. The goal is honest approximation, not false precision. If you cannot define what success looks like before you spend, you will not be able to defend the spend after.

Programmatic CPM: Where the Complexity Lives

Most CPM buying today happens programmatically, through demand-side platforms that bid on inventory in real time. The mechanics are worth understanding because they affect what your CPM actually buys.

In a real-time bidding auction, your DSP submits a bid for each impression opportunity based on the audience data attached to it. The winning bid becomes the price you pay, usually at a second-price auction where you pay just above the second-highest bid. Your average CPM across a campaign is the aggregate of thousands or millions of individual auction outcomes.

This means your CPM is partly a function of how aggressively you are bidding, how competitive your target audience is, and how your creative performs relative to other advertisers in the auction. Platforms that use quality scores or engagement signals in their auction mechanics will reward better creative with lower effective CPMs. That is worth knowing because it means creative quality has a direct impact on buying efficiency, not just on engagement rates.

Display advertising volumes have grown enormously since the early programmatic era, which has made the auction landscape more competitive and driven up CPMs in premium segments. That trend has continued. The implication is that audience precision matters more than it used to. Broad targeting in a competitive auction is expensive. Tight targeting with strong creative is where the efficiency lives.

Private marketplace deals, or PMPs, are an alternative to open auction buying. You negotiate directly with publishers for access to their inventory at agreed CPM rates, with more transparency about where your ads appear. PMPs tend to cost more than open auction but deliver better placement quality and brand safety. For brands where context matters, the premium is usually worth it.

The Creative Problem That CPM Campaigns Expose

CPM campaigns have a way of making weak creative very expensive very quickly. When you are paying for impressions rather than clicks, you are paying for attention whether or not your ad earns it. A dull banner served a million times is a million opportunities wasted.

This is not a minor operational point. It is one of the most common reasons CPM campaigns underperform. The media buying is fine. The targeting is reasonable. The creative is forgettable.

I have judged the Effie Awards, which evaluate marketing effectiveness rather than creative craft, and the pattern that emerges consistently is that campaigns with strong creative generate disproportionate returns relative to their media spend. The creative is not decoration. It is the mechanism through which the media investment actually works.

For CPM campaigns specifically, this means investing in creative that works in the formats and contexts where your ads will appear. A video that requires sound to make sense is a problem on a platform where most users watch with sound off. A display banner with seven words of copy and a small logo is not going to build brand recall. The creative brief needs to account for the viewing conditions, not just the message.

Testing creative variants is standard advice, but it is worth being specific about what to test. In CPM campaigns, the variables that tend to matter most are visual distinctiveness, brand prominence, and message clarity in the first two seconds. Click-through rate is a secondary metric at best. If your CPM campaign is generating clicks, something has probably gone wrong with your targeting or your audience is further down the funnel than you thought.

Measuring CPM Campaigns Without Lying to Yourself

The measurement challenge with CPM marketing is real, and anyone who tells you otherwise is either selling you something or has not run enough campaigns to know better.

Impressions, reach, and frequency are process metrics. They tell you whether the campaign was delivered as planned. They do not tell you whether it worked. The gap between delivery and impact is where most CPM reporting falls short.

The most reliable measurement approaches for CPM campaigns tend to be:

Brand lift studies, which measure changes in awareness, consideration, or purchase intent among exposed versus unexposed audiences. Most major platforms offer these for campaigns above a certain spend threshold. They are not perfect, but they give you a directional read on whether the impressions are doing anything.

Search volume tracking, which looks at changes in branded search queries during and after a CPM campaign. If people are searching for your brand more after seeing your ads, that is a signal worth taking seriously.

Incrementality testing, which compares outcomes between a group that saw your ads and a holdout group that did not. This is the most rigorous approach but requires careful setup and sufficient scale to be statistically meaningful.

The honest version of CPM measurement is that you are working with signals, not certainty. That is fine. Marketing has never offered perfect measurement. What it should offer is enough signal to make informed decisions about whether to continue, adjust, or stop. The goal is not to prove that every impression drove a sale. It is to build a reasonable case that the investment is contributing to commercial outcomes over time.

If you are thinking about how CPM fits into a broader paid strategy, the paid advertising section covers channel selection, measurement frameworks, and budget allocation across the full mix.

What the Display Advertising Rebound Taught Us About CPM Value

There is an interesting historical footnote worth mentioning here. In the early 2010s, display advertising went through a period of significant scepticism. Click-through rates had collapsed from the early days of the web. Advertisers were questioning whether display was worth buying at all. The rise of performance marketing and paid search had made impression-based buying look unsophisticated by comparison.

And then something interesting happened. Brands that had abandoned display started to notice that their search campaigns were becoming more expensive and less efficient. New entrants who had invested in brand awareness through display were seeing lower CPCs and higher conversion rates on search because they had built prior recognition. The awareness investment upstream was making the performance investment downstream work harder.

That dynamic has not changed. The mechanics of how people make purchasing decisions have not fundamentally shifted. Familiarity reduces friction. Brand presence builds trust before the moment of purchase. CPM campaigns, done well, create the conditions in which performance campaigns can operate efficiently.

The mistake is treating CPM and performance as competing priorities rather than complementary ones. They operate on different timescales and serve different functions. The brands that manage both well tend to outperform those that optimise exclusively for short-term conversion metrics. Forrester’s research on B2B marketing has consistently pointed to the gap between what buyers experience before they enter a sales process and what marketing teams actually invest in. CPM is part of closing that gap.

Common CPM Mistakes Worth Avoiding

Having managed hundreds of millions in ad spend across multiple agency roles, the CPM mistakes I see most often are not exotic. They are consistent and avoidable.

Optimising for the lowest CPM rather than the best audience. Cheap impressions from the wrong people are not a bargain. They are a cost with no return. Set your targeting before you set your CPM floor, not the other way around.

Running CPM campaigns without frequency caps. Serving the same ad to the same person twenty times in a week does not build brand affinity. It builds irritation. Most platforms allow you to cap impressions per user per day or per week. Use that control.

Measuring CPM campaigns with direct response metrics. If you judge a brand awareness campaign by its click-through rate, you will always be disappointed and you will always draw the wrong conclusions. Match your measurement to your objective.

Ignoring viewability. An impression that appears below the fold and is never seen is still charged at your CPM rate. Viewability standards exist for a reason. Make sure your campaign settings or your direct publisher deals include viewability thresholds as a baseline requirement.

Separating CPM campaigns from the rest of the marketing plan. Awareness campaigns do not exist in isolation. They feed into consideration, search, and conversion. If your media plan does not account for the relationship between upper-funnel CPM activity and lower-funnel performance, you are managing channels rather than a campaign.

One other thing worth flagging: optimisation is not a one-time event. CPM campaigns need ongoing attention to placement quality, creative fatigue, audience saturation, and bid adjustments. Setting a campaign live and checking it monthly is not a strategy. It is neglect dressed up as efficiency.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions