CPM Marketing: What You’re Paying For and Whether It’s Worth It

CPM marketing means paying per thousand impressions, not per click or conversion. You buy visibility at scale, and the commercial logic only holds if that visibility is reaching the right people at the right moment in their decision-making process. When it works, it builds the kind of awareness that makes every downstream channel perform better. When it doesn’t, it’s budget disappearing into the void with a dashboard full of green numbers to show for it.

I’ve managed media budgets across dozens of categories over the years, and CPM has always been one of those topics where the gap between how it’s sold and how it actually performs is wider than most people admit. It’s worth being honest about both sides of that.

What Does CPM Actually Mean?

CPM stands for cost per mille, where mille is Latin for thousand. If you’re paying a £5 CPM, you’re paying £5 for every 1,000 times your ad is served. That’s the unit of trade in most display advertising, programmatic buying, video pre-roll, social media awareness campaigns, and a significant chunk of digital out-of-home.

The model has its roots in traditional media. TV, radio, and print have always been sold on reach and frequency, with CPM as the standard comparison metric. When digital advertising arrived, it inherited the same model for display inventory, even as it introduced new performance-based alternatives like CPC and CPA. Today, CPM buying coexists with those models across most major platforms, and understanding when to use which is one of the more commercially important decisions a media buyer makes.

The core appeal of CPM is predictability. You know what reach costs before you spend. The core risk is that reach without relevance is waste, and the model doesn’t automatically protect you from that.

CPM vs CPC vs CPA: Choosing the Right Buying Model

These three models represent different bets on where value is created in the advertising process.

CPC, cost per click, shifts risk toward the platform. You only pay when someone engages. That sounds attractive, but platforms optimise toward clicks, which aren’t always the same as commercial intent. You can drive a lot of clicks from curious people who never buy anything.

CPA, cost per acquisition, pushes risk further still. You pay only when a defined action occurs. This is the most commercially aligned model on paper, but it requires volume and data before platforms can optimise effectively, and it can create perverse incentives where the algorithm chases easy conversions rather than valuable ones.

CPM sits at the top of that risk spectrum for the advertiser. You’re paying for exposure regardless of what happens next. That’s appropriate for brand building, where the goal is mental availability over time, not immediate response. It’s less appropriate when you need direct commercial return from every pound spent.

The honest answer is that most sophisticated campaigns use all three, depending on where in the funnel you’re operating. CPM for awareness and reach, CPC for consideration and traffic, CPA for conversion and retention. Treating them as competitors rather than complements is a mistake I’ve seen made repeatedly, usually by teams that have been pushed into a single-metric accountability framework that doesn’t reflect how buying decisions actually work.

If you want a broader view of how CPM fits into the full picture of paid media, the paid advertising hub covers the landscape in more depth, from programmatic to paid search to social.

What Drives CPM Rates?

CPM rates vary enormously depending on several factors, and understanding them matters if you’re going to negotiate intelligently or interpret your media plan critically.

Audience quality is the biggest driver. A CPM targeting senior financial decision-makers in a specific industry will cost multiples of a broad consumer CPM, because that audience is harder to reach and more commercially valuable to advertisers competing for their attention. When I was working with B2B clients at agency level, we’d regularly see CPMs five to ten times higher than equivalent consumer campaigns, and they were often the better investment because the audience precision was so much tighter.

Placement context matters too. Premium editorial environments, high-viewability positions, and formats that command attention (full-page takeovers, pre-roll video with high completion rates) all carry higher CPMs than remnant inventory. The question isn’t whether the CPM is high. It’s whether the context justifies it.

Seasonality creates significant swings. Q4 in consumer categories, back-to-school periods, and major sporting events all drive up auction prices as more advertisers compete for the same inventory. If you’re planning CPM campaigns, building in seasonal cost assumptions is basic commercial hygiene that gets skipped more often than it should.

Format is another variable. Video CPMs are typically higher than display CPMs, which are higher than native content placements. Connected TV inventory commands a premium over mobile web. These aren’t arbitrary, they reflect differences in attention, completion rates, and the difficulty of reaching audiences in those environments.

The Viewability Problem Nobody Talks About Honestly

Here’s something that took me longer than I’d like to admit to fully internalise: a served impression and a viewed impression are not the same thing, and the industry standard definition of viewability is not particularly demanding.

The Media Rating Council’s standard for a viewable display impression is 50% of the ad in view for at least one second. For video, it’s 50% in view for two seconds. Those thresholds exist because the industry needed a workable standard, not because one second of half-visible exposure reliably creates brand memory.

When you’re buying CPM, you’re often buying against a viewability target, say 70% viewable impressions, which means up to 30% of your impressions may not meet even that modest standard. In practice, the gap between reported impressions and genuinely seen, registered impressions is wider than most media plans acknowledge.

This isn’t a reason to avoid CPM buying. It’s a reason to ask harder questions about where your impressions are being served, what viewability standards your contracts specify, and whether your creative is designed to work in the environments where it’s actually appearing. A lot of display creative is built for a desktop context and served predominantly on mobile, which is a mismatch that compounds the viewability issue.

I’ve sat in enough media reviews to know that these numbers often get reported as successes when they’re really just activity. Impressions delivered, viewability percentage met, brand safety targets achieved. All of that can be true and the campaign can still have done nothing commercially useful.

Where CPM Marketing Genuinely Earns Its Return

There are specific use cases where CPM buying is not just defensible but clearly the right approach.

New product launches and category entry are the clearest cases. When nobody knows you exist, you can’t rely on search demand that hasn’t formed yet. You need to create awareness before you can capture intent. CPM campaigns that build reach efficiently across a defined audience are the right tool for that job, and the commercial logic holds even without immediate conversion data to point to.

Retargeting is where CPM often delivers its most measurable return. You’re serving impressions to people who have already visited your site, engaged with your content, or shown some signal of interest. The impression has context. The audience is warm. Retargeting campaigns consistently outperform cold audience display because the work of establishing relevance has already been partially done. If you’re running CPM buying and you haven’t separated your retargeting audiences from your prospecting audiences in your reporting, you’re almost certainly misreading your results.

Competitive defence is another legitimate use case. If a competitor is aggressively building share of voice in your category, a CPM campaign that maintains your presence in front of your existing customer base has real commercial value, even if it’s hard to attribute directly. The counterfactual, what happens to retention if you go dark, is often more expensive than the campaign itself.

And then there’s the role CPM plays in supporting demand generation more broadly. Demand generation is about creating the conditions in which people want to buy from you, not just capturing the people who already do. CPM campaigns that build category awareness and brand salience are part of that infrastructure, even when they don’t show up cleanly in last-click attribution.

How to Calculate Whether Your CPM Is Working

The measurement challenge with CPM is real, but it’s not a reason to abandon accountability. It’s a reason to use the right tools.

Start with the basics. If you’re running a CPM campaign and you have no hypothesis about what commercial outcome it should support, you’ve already made the fundamental mistake. Define what you expect to move: brand recall, search volume for your brand terms, direct traffic, consideration scores in your category. Then measure those things before, during, and after the campaign.

Brand lift studies, available through most major platforms, give you a controlled read on whether your campaign is moving awareness and consideration metrics. They’re not perfect, but they’re honest. They tell you whether the people who saw your ads are more likely to recall your brand or consider purchasing than those who didn’t. That’s a meaningful signal.

Search volume trends for your brand terms are a useful proxy. If you’re running significant CPM activity and branded search isn’t moving at all, that’s worth investigating. It doesn’t prove the campaign failed, but it’s a data point that should prompt questions.

Return on ad spend is the metric that in the end matters for any paid media investment. Understanding how ROAS is calculated and where CPM campaigns fit into that calculation is essential context for anyone managing media budgets. CPM campaigns rarely show strong direct ROAS because they’re not designed for immediate conversion. But they should show up in the assisted conversion data and in the performance lift of channels that benefit from the awareness they create.

Incrementality testing is the most rigorous approach. Run your campaign to one geographic or demographic segment and hold out another as a control. Measure the difference in commercial outcomes. It’s more operationally complex than standard campaign reporting, but it gives you a defensible answer to the question of whether the spend is working.

Early in my career, when I was first getting to grips with digital advertising, I was too quick to trust the platform dashboards. They’re designed to show your campaign performing well, and they’re good at it. The discipline of building your own measurement framework, separate from what the platform reports, is something I’d recommend to anyone managing meaningful CPM budgets.

Programmatic CPM: Scale, Precision, and the Complexity Tax

Most CPM buying today happens programmatically, through real-time auction systems that match advertiser demand with publisher supply in milliseconds. The appeal is obvious: you can reach specific audience segments across thousands of publishers at scale, with targeting parameters that direct media buys can’t match.

The complexity is equally obvious, and it comes with a cost that doesn’t always appear in the headline CPM. The programmatic supply chain involves demand-side platforms, supply-side platforms, data management platforms, ad exchanges, and verification vendors, each taking a margin. The effective CPM you pay and the CPM that reaches the publisher are often materially different numbers. Understanding that gap matters if you’re trying to optimise media efficiency.

Audience targeting through programmatic is powerful but not magic. Third-party data segments are built on probabilistic modelling, not certainty. A segment labelled “in-market for financial services” is a best guess based on behavioural signals, not a confirmed list of people actively shopping for a financial product. The quality of your targeting is only as good as the data underpinning it, and that data is often less precise than it’s presented to be.

Private marketplace deals, where you negotiate directly with premium publishers for programmatic access to their inventory, sit between open auction buying and traditional direct buys. They give you more control over placement quality while retaining the targeting and automation benefits of programmatic. For brands where context and brand safety are important, they’re worth the additional complexity.

If you want to go deeper on the mechanics of paid media buying and how programmatic fits into the broader landscape, the paid advertising section covers the channel-level decisions in more detail.

CPM on Social Platforms: Different Logic, Different Expectations

Social platforms sell CPM inventory with a significant additional asset: first-party audience data. When you’re buying CPM on Meta, LinkedIn, or TikTok, you’re targeting against declared interests, behaviours, and demographic data that users have provided directly to the platform. That’s a different proposition from third-party programmatic data, and it generally produces better audience precision.

The trade-off is that you’re inside a walled garden. The platform controls what you can see, what you can target, and how your results are reported. Cross-platform attribution is difficult. The platform has an obvious commercial interest in showing your campaigns performing well. That’s not a conspiracy, it’s just the structure of the relationship, and it means you need to maintain healthy scepticism about platform-reported metrics.

LinkedIn CPMs are notably higher than most other platforms, often significantly so. That premium is justified for B2B advertisers reaching professional audiences by job title, seniority, or company. It’s harder to justify for consumer categories where the same audience can be reached more cheaply elsewhere. I’ve seen B2B clients waste budget on LinkedIn because it felt credible, not because the targeting was actually better than alternatives for their specific objective.

Mobile advertising has its own CPM dynamics. Mobile ad campaigns have shown distinct effectiveness patterns compared to desktop, and the creative requirements are different. Formats that work on desktop often don’t translate to mobile environments, and a lot of CPM spend is wasted because the creative hasn’t been adapted for the context in which it’s actually appearing.

The Creative Variable That Media Plans Ignore

Media planning and creative development are too often treated as separate disciplines, and CPM campaigns suffer for it more than most. The targeting can be perfect, the placement premium, the viewability high, and the campaign still fails because the creative doesn’t do anything in the moment of exposure.

In a CPM context, you’re often competing for attention in an environment where the audience is primarily there for something else. They’re reading an article, scrolling a feed, watching a video. Your ad is an interruption. The creative has to earn attention quickly, communicate something meaningful, and leave a residue. That’s a high bar for a 300×250 display unit or a six-second pre-roll.

The most common failure mode I’ve seen is creative that works well in isolation but disappears in context. It looks fine in a deck. It’s invisible on the page. The second most common failure is creative that’s been designed for one format and adapted (poorly) to others. A landscape video cut to square for Instagram, a desktop banner compressed for mobile, a rich media unit served in an environment that doesn’t support it.

When I was running a paid search campaign for a music festival at lastminute.com, the speed of the commercial return was striking, but that was a high-intent channel where the creative job was relatively simple: match the search query, present the offer clearly. CPM is a different creative challenge entirely. You’re not responding to expressed intent. You’re trying to create it, or at least prime it. That requires more from the creative, not less.

When CPM Marketing Is the Wrong Choice

There are situations where CPM is genuinely the wrong buying model, and being clear about them is more useful than defending the format in all circumstances.

If your objective is immediate direct response and you have limited budget, CPM is rarely the right starting point. You’re paying for reach before you’ve validated your offer, your creative, or your conversion path. Start with CPC or CPA buying to test what works, then scale with CPM once you have evidence that the broader awareness investment will pay back.

If your targeting is weak, CPM amplifies the problem. A poorly defined audience on a CPM model means you’re paying to serve irrelevant impressions at scale. The cost of that mistake scales linearly with your budget. At least with CPC, you’re only paying when someone engages, which provides a natural filter.

If you can’t measure anything downstream of the impression, CPM campaigns become very hard to defend commercially. You’re relying on faith that awareness creates value, which it does, but without any proxy metrics to track, you have no way to know whether your specific campaign is contributing or whether you’re just spending money. That’s not a position any serious marketing leader should be comfortable with.

The broader point is that CPM is a tool with specific applications. Pay-per-click marketing has its own logic and its own limitations. Neither model is universally superior. The question is always which model fits the objective, the audience, and the measurement framework you have available.

What Good CPM Campaign Management Looks Like

Running CPM campaigns well requires a different discipline than performance marketing. The feedback loops are slower, the signals are noisier, and the temptation to optimise toward metrics that feel good but don’t mean much is constant.

Start with a clear audience definition. Not a broad demographic, a specific description of the person you’re trying to reach, what they care about, where they spend time, and what you want them to think or feel after seeing your ad. That definition should drive every targeting and placement decision.

Set frequency caps. Serving the same impression to the same person thirty times in a week is not marketing. It’s annoyance, and it has a measurable negative effect on brand sentiment. Most platforms allow frequency capping. Use it.

Monitor placement quality actively. In programmatic campaigns especially, your ads can end up in environments that are technically within your targeting parameters but contextually wrong for your brand. Regular placement reports and exclusion list management are basic hygiene that gets skipped when campaigns are set and forgotten.

Build in creative rotation. A single creative running for three months will experience significant fatigue. Audiences who’ve seen your ad twenty times stop registering it. Rotating creative, even with modest variation, maintains the effectiveness of the impression over time.

And connect your CPM activity to the rest of your funnel. If you’re running awareness campaigns, you should be able to see their effect in your consideration metrics, your branded search volume, and your conversion rates over time. If you can’t see any signal at all, that’s worth investigating before you continue spending.

I’ve found that the marketers who get the most from CPM buying are the ones who treat it as part of a system rather than a standalone channel. The impression is the beginning of a conversation, not the end of it. What happens after the impression is as important as the impression itself, and that means the campaign strategy has to extend beyond the media plan.

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