Impressions vs Engagement: You’re Measuring the Wrong Thing

Impressions tell you how many times your content appeared in front of someone. Engagement tells you how many times someone chose to respond to it. They measure different things, they predict different outcomes, and treating them as interchangeable is one of the most common ways marketing teams quietly waste budget while producing reports that look fine.

Neither metric is inherently superior. The question is which one is relevant to what you are actually trying to achieve, and whether you are using either one honestly.

Why This Debate Keeps Resurfacing

The impressions versus engagement debate has been running for as long as digital marketing has had dashboards. And it keeps resurfacing because neither side is entirely right, and most teams are not honest about what they are actually trying to measure.

I have sat in enough quarterly business reviews to know how this usually plays out. When reach is up, the team talks about brand building. When engagement is up, the team talks about community and connection. When neither is up, someone finds a metric that is. The reporting bends to the narrative rather than the other way around.

That is not a measurement problem. It is a strategic clarity problem. And it tends to get worse, not better, as teams grow and reporting layers multiply.

If you want a cleaner framework for thinking about this, it helps to start with what each metric is actually built to capture, and where each one breaks down.

What Impressions Actually Measure

An impression is a served exposure. It does not mean the person looked at it, processed it, or remembered it. It means the platform counted a delivery event. On most platforms, that threshold is low enough to be generous.

That does not make impressions useless. At scale, consistent exposure to a brand does build something. Mental availability, the concept that a brand comes to mind when a purchase occasion arises, is real and commercially meaningful. You cannot build it without reach. The problem is that impressions as reported by platforms are a very rough proxy for the kind of quality exposure that actually moves the needle.

When I was managing large media budgets across multiple sectors, I watched teams report impression numbers that looked extraordinary. Hundreds of millions of served exposures. And then the brand tracking data would come back flat. The impressions were real. The impact was not proportionate. The gap between served impressions and meaningful impressions was significant, and most teams were not accounting for it.

Impressions are most defensible when you are in a genuine awareness phase, entering a new market, launching a new product, or reaching an audience that has no existing relationship with your brand. In those contexts, reach matters because you cannot build anything without it. BCG’s work on commercial transformation in go-to-market strategy is consistent on this point: growth requires reaching new audiences, not just serving existing ones more efficiently.

Outside of that context, impressions as a primary success metric is a choice that deserves more scrutiny than it usually gets.

What Engagement Actually Measures

Engagement is a behavioural signal. A like, a comment, a share, a save, a click. It means someone did something in response to your content rather than scrolling past it. That is a more active signal than an impression, and it is tempting to treat it as a proxy for interest or intent.

The problem is that engagement is heavily shaped by content format, platform algorithm, and emotional register, not by commercial relevance. Content that provokes a strong reaction gets shared. Content that is genuinely useful to a specific buyer rarely goes viral. The two things are not the same, and optimising for one does not automatically produce the other.

I have seen this pattern repeatedly. A brand posts something slightly provocative or culturally timely and it gets ten times the engagement of their normal content. The team celebrates. The CMO gets excited. And then someone asks whether any of those engagements came from people who were ever going to buy the product, and the room goes quiet.

Engagement is a meaningful signal when it comes from the right people on the right content. It is noise when it is driven by content that was designed to generate reactions rather than build commercial relationships. Platforms like Hotjar exist precisely because on-platform engagement data does not tell you what happens to behaviour once someone leaves the feed. You need a separate signal layer to understand that.

The other issue with engagement as a metric is that it is platform-defined. The platform decides what counts as an engagement, how it is weighted, and what appears in your dashboard. You are measuring what the platform wants you to measure, which is not always what is commercially useful to you.

The Funnel Position Problem

A lot of the impressions versus engagement debate is actually a funnel position debate in disguise. Impressions are broadly an upper-funnel metric. Engagement sits somewhere in the middle. Neither one is a conversion metric. And yet both are frequently used to justify spend that is supposed to be driving commercial outcomes.

Earlier in my career I overvalued lower-funnel performance metrics. I thought that if something was measurably converting, it was working. What I came to understand, slowly and through some expensive lessons, is that a lot of what performance marketing gets credited for was going to happen anyway. You are capturing intent that already existed, not creating it. And if you are not also creating new intent further up the funnel, you are slowly harvesting a field you stopped planting.

The analogy I keep coming back to is a clothes shop. Someone who tries something on is significantly more likely to buy than someone who just browses. But the person who tries something on got into the shop first. If you only measure the fitting room conversion rate, you will optimise the fitting room experience while the footfall quietly declines. Impressions and engagement, done well, are what gets people into the shop.

The Forrester intelligent growth model has long argued for connecting upper-funnel activity to downstream commercial outcomes, rather than treating awareness and conversion as separate programmes with separate success metrics. That framing is more useful than the impressions versus engagement binary.

If you are thinking about where impressions and engagement fit within a broader commercial strategy, the Go-To-Market and Growth Strategy hub covers the full picture, from market entry to channel selection to measurement frameworks that connect to revenue rather than just activity.

When Impressions Should Win the Argument

There are specific situations where impressions are the right primary metric and arguing for engagement instead is a mistake.

If you are entering a new market or launching a product that nobody knows exists, your first job is to build awareness at scale. Engagement on zero awareness is a contradiction. You cannot have a meaningful conversation with an audience that does not know you exist. In this context, optimising for reach, even imperfect reach, is the correct call.

If you are running a brand campaign designed to shift perception over time, impressions are the appropriate unit of measurement. Brand building is a long, slow, accumulative process. It does not convert in the quarter it runs. Measuring it by engagement rate is the wrong instrument for the job.

If you are in a category where purchase decisions are low-frequency and high-consideration, such as insurance, financial services, or B2B software, your job is to be mentally available when the purchase moment arrives. That requires consistent exposure over time. Impressions, at the right frequency and in the right context, are doing real work even when engagement is low.

The mistake is treating impressions as a vanity metric by default. They are only vanity if they are disconnected from a strategic purpose. If the purpose is clear and the targeting is right, a high impression count is a legitimate business outcome.

When Engagement Should Win the Argument

Engagement becomes the more useful metric when you are operating in a context where behavioural signals are genuinely predictive of commercial intent.

In B2B marketing, where audiences are small and every interaction with the right person matters, engagement is a more useful signal than raw reach. If a CFO at a target account watches your entire video, saves your whitepaper, and comments on a post, that is a meaningful signal regardless of what your total impression count looks like. You would rather have that than a million impressions served to people who will never buy from you.

In content marketing and community building, engagement is the mechanism through which the relationship develops. A newsletter open rate, a reply, a share to a colleague, these are signals of a relationship forming. Impressions do not capture that. The creator-led go-to-market work that Later has documented makes this point clearly: the value of creator content is often in the depth of engagement with a specific audience, not the breadth of reach.

In performance-adjacent contexts, where engagement is a leading indicator of conversion, it is worth measuring carefully. Click-through rate, save rate, and direct message volume on social platforms can all be early signals of purchase intent, particularly in e-commerce and direct-to-consumer categories. what matters is validating that the engagement actually precedes conversion in your specific funnel, not assuming it does because the logic seems plausible.

The Honest Problem With Both Metrics

Here is what most marketing teams do not say out loud. Both impressions and engagement are metrics that platforms report, and platforms have a direct financial interest in those numbers looking good. The incentives are not neutral. An impression count that inflates slightly in the platform’s favour is good for the platform’s ad revenue. An engagement metric that includes low-intent interactions makes your campaign look more active than it is.

I spent years managing significant media budgets across multiple platforms, and the discrepancies between platform-reported data and what we could observe through independent measurement were consistent enough to be a standing agenda item in client reviews. Not because the platforms were lying, exactly, but because their measurement methodology was built to serve their commercial model, not yours.

The practical implication is that you should never rely solely on platform-reported metrics for either impressions or engagement. Build your own signal layer. Use UTM parameters, first-party data, CRM attribution, and behavioural analytics tools to construct a view of what is actually happening downstream. Platform data is an input, not a conclusion.

Semrush’s analysis of market penetration strategy makes a related point: the metrics you choose to optimise for shape the strategy you end up executing. If your primary metric is platform-reported engagement, you will end up creating content optimised for platform algorithms. That may or may not align with what actually grows your market share.

Building a Measurement Framework That Actually Works

The practical answer to the impressions versus engagement question is to stop treating it as a binary and start building a measurement framework that connects both metrics to commercial outcomes.

Start with the business outcome you are trying to achieve. Not the marketing outcome. The business outcome. Revenue, market share, customer acquisition cost, lifetime value. Work backwards from there to identify which marketing behaviours are most predictive of that outcome in your specific context.

Then identify which metrics, impressions, engagement, or something else entirely, are most closely correlated with those behaviours. This is not a theoretical exercise. It requires looking at your own data, not industry benchmarks, and finding the connections that exist in your funnel rather than the ones that are supposed to exist in a generic model.

When I was helping turn around an agency that had been losing money for two years, one of the first things I did was audit the metrics the team was reporting to clients. Almost every client deck led with impressions and engagement. Almost none of them had a clear line connecting those metrics to the commercial outcomes the client actually cared about. The reporting looked thorough. The strategic connection was missing. That gap was costing us client relationships, and it was costing clients money.

BCG’s research on go-to-market strategy in B2B markets is consistent on this: the companies that grow sustainably are the ones that connect marketing activity to commercial value clearly and honestly, rather than reporting activity as a proxy for impact.

A framework worth building has three layers. First, reach metrics, of which impressions are one, that tell you whether you are getting in front of the right people at the right scale. Second, engagement metrics that tell you whether those people are responding in ways that suggest interest or intent. Third, conversion and commercial metrics that tell you whether any of that is translating into business outcomes. All three layers matter. None of them is sufficient on its own.

The Semrush breakdown of growth strategy examples illustrates this well: the approaches that produced durable growth were the ones that connected acquisition activity to retention and revenue, not just to top-of-funnel volume.

There is more on how to build measurement frameworks that connect to commercial strategy across the full Go-To-Market and Growth Strategy hub, including how to think about channel selection, audience segmentation, and the metrics that actually matter at each stage of a go-to-market plan.

The Question Worth Asking in Every Review

When I judged the Effie Awards, the entries that stood out were not the ones with the biggest impression numbers or the highest engagement rates. They were the ones where the team could explain clearly why the metric they chose to optimise for was the right one for the business problem they were solving, and then show that it had actually worked.

That is a higher bar than most marketing teams hold themselves to. It requires being honest about what you are measuring and why, rather than reporting the numbers that look best in a given period.

The question worth asking in every performance review is not whether impressions or engagement went up. It is whether the activity you ran moved the business closer to the outcome it needed. If it did, the metric you used to measure it was probably the right one. If it did not, no amount of impressive impression counts or strong engagement rates changes that.

Impressions and engagement are both legitimate tools. Neither one is the enemy. The problem is not the metrics themselves. It is the habit of treating them as ends rather than means, and the willingness to let good-looking numbers substitute for honest commercial thinking.

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