Marketing Measurement Methods That Reflect Business Reality

Marketing measurement methods are the frameworks and tools used to connect marketing activity to business outcomes, from brand tracking and attribution modelling to incrementality testing and revenue contribution analysis. Most businesses use some combination of these, but the combination rarely tells a coherent story.

The problem is not a shortage of measurement options. It is that most measurement setups are designed to make marketing look productive rather than to determine whether it actually is. Fix that orientation, and the rest of measurement tends to follow.

I spent years managing large performance marketing budgets across dozens of clients, and the measurement question was always the most politically charged one in the room. Not because it was technically difficult, though it often was, but because the answers threatened budgets, relationships, and comfortable narratives. That tension is worth naming before we get into the methods themselves.

Why Most Marketing Measurement Is Built to Confirm, Not to Challenge

There is a version of marketing measurement that functions as a reporting exercise. Numbers come in, dashboards light up, and the story is written to match whatever the business wanted to hear. I have sat in enough quarterly reviews to know this is not cynicism. It is a description of how most measurement actually operates.

The structural problem is that the people responsible for measuring marketing are often the same people responsible for delivering marketing results. That is not a personal failing. It is an incentive problem. When your job depends on the numbers looking good, you will, consciously or not, make choices that favour confirmation over challenge.

Early in my career I was as guilty of this as anyone. I overvalued lower-funnel performance metrics because they were clean, attributable, and easy to defend in a presentation. Paid search delivered a cost-per-acquisition. Retargeting showed a return on ad spend. The numbers were right there. What I did not ask, and should have asked much earlier, was how much of that activity was capturing demand that already existed rather than creating new demand. The honest answer, once we started running proper incrementality tests, was: more than we wanted to admit.

This shapes everything that follows. The methods below are only useful if you approach them with a genuine appetite for what they reveal, including the inconvenient parts.

If you are working through a broader go-to-market strategy and trying to understand how measurement fits into the wider commercial picture, the Go-To-Market and Growth Strategy hub covers the full landscape, from audience strategy to channel selection to performance frameworks.

What Is the Difference Between Attribution and Measurement?

Attribution and measurement are not the same thing, though they are routinely conflated. Attribution is one method within the broader measurement toolkit. It attempts to assign credit for a conversion or outcome to specific touchpoints in the customer experience. Measurement is the wider discipline of understanding marketing’s contribution to business performance.

Conflating the two leads businesses to treat their attribution model as the definitive answer to the measurement question. It is not. Attribution tells you which touchpoints were present before a conversion. It does not tell you which ones caused it.

That distinction, between correlation and causation in the customer experience, is where most measurement frameworks fall apart. A customer who clicked a retargeting ad and then converted was going to convert anyway in a significant proportion of cases. The ad was present. It may not have been necessary. Attribution models, particularly last-click, cannot distinguish between the two.

The Core Methods: What Each One Actually Measures

Last-Click and First-Click Attribution

These are the oldest and simplest attribution models. Last-click assigns 100% of conversion credit to the final touchpoint before purchase. First-click assigns it to the first. Both are easy to implement and both are wrong in the same fundamental way: they treat a single touchpoint as the complete explanation for a conversion that usually involved multiple interactions over time.

Last-click in particular systematically benefits paid search and retargeting because those channels sit at the bottom of the funnel, closest to the point of conversion. Channels that build awareness and consideration, brand campaigns, content, organic social, get no credit because they did their work earlier. This distorts budget allocation in ways that compound over time. You defund the channels that generate demand and over-invest in the channels that capture it.

I managed one client where paid search was consuming over 60% of the digital budget on the basis of last-click return on ad spend figures. When we ran a geo-based incrementality test, the incremental return was less than half what the attribution model suggested. The channel was still worth running. It was not worth running at that scale.

Multi-Touch Attribution

Multi-touch attribution (MTA) attempts to distribute credit across multiple touchpoints in the customer experience. Linear models split it equally. Time-decay models weight recent touchpoints more heavily. Data-driven models use statistical analysis to assign credit based on observed patterns.

MTA is more sophisticated than single-touch models, but it shares the same foundational limitation: it is correlational. It tells you which touchpoints appeared in converting journeys. It cannot tell you whether removing any one of them would have changed the outcome. It also depends entirely on the quality of your tracking, which is increasingly compromised by cookie deprecation, iOS privacy changes, and the growth of cookieless environments.

Data-driven MTA from platforms like Google is also worth treating with appropriate scepticism. The model is trained on data from within the platform’s own ecosystem. It has a structural interest in attributing value to touchpoints that run on that platform. That does not make it useless. It makes it a perspective, not a verdict.

Marketing Mix Modelling

Marketing mix modelling (MMM) takes a fundamentally different approach. Rather than tracking individual customer journeys, it uses aggregate data over time to model the statistical relationship between marketing spend and business outcomes, typically revenue or volume. It can account for external factors like seasonality, economic conditions, and competitor activity.

MMM is better suited to answering strategic questions: which channels are driving incremental growth, what is the optimal budget allocation, how does marketing interact with pricing and distribution? It is less useful for day-to-day optimisation because it operates on longer time horizons and requires substantial historical data to produce reliable outputs.

The resurgence of interest in MMM over the last few years is partly a response to the deterioration of user-level tracking. As individual-level attribution becomes harder to do accurately, aggregate modelling becomes more attractive. That is a reasonable response to a real problem, but MMM has its own limitations. It requires clean, consistent data over a meaningful time period. It struggles to capture the effect of new channels with limited history. And the outputs are only as good as the model specification, which requires genuine statistical expertise to get right.

Incrementality Testing

Incrementality testing is, in my view, the most honest method available. The principle is straightforward: you create a test group that is exposed to a marketing activity and a control group that is not, then measure the difference in outcomes. The difference is the incremental effect of the activity.

This can be done through geo-based experiments, where you run activity in some markets and not others. It can be done through holdout groups in digital campaigns. It can be done through matched market tests. The methodology varies, but the underlying logic is the same: you are measuring what actually changed because of the marketing, not just what correlated with it.

The reason incrementality testing is underused is not that it is technically inaccessible. It is that the results are often uncomfortable. When we ran holdout tests on retargeting campaigns for several clients, the incremental return was consistently lower than the attributed return, sometimes dramatically so. Those results were correct. They were also unwelcome, because they implied budget reallocation and uncomfortable conversations with channel owners.

If you want to understand whether your marketing is working, incrementality testing is the method that gets closest to an answer. The broader challenge of making go-to-market activity measurable and defensible is something many teams are grappling with right now, and incrementality is one of the few tools that gives you a genuinely causal answer.

Brand Tracking and Awareness Measurement

Brand tracking measures changes in consumer awareness, consideration, preference, and perception over time. It is typically conducted through surveys, either continuously or at intervals, and it captures the parts of marketing’s contribution that attribution models cannot see.

The challenge with brand tracking is connecting it to commercial outcomes. Awareness is not revenue. Consideration is not conversion. The link between brand metrics and business performance is real, but it operates over longer timeframes and through mechanisms that are harder to isolate. This makes brand tracking easy to dismiss in a performance-first culture, which is exactly the wrong response.

When I was judging the Effie Awards, the entries that consistently demonstrated the strongest commercial cases were the ones that showed both the brand and the business outcome, not one or the other. The campaigns that could only show brand uplift were interesting. The ones that could show brand uplift leading to measurable sales growth were the ones that won. The measurement challenge is building the bridge between the two.

Revenue Attribution and Commercial Contribution Analysis

Beyond channel-level attribution, there is a broader question of how marketing as a function contributes to revenue and commercial performance. This involves connecting marketing activity to the metrics that the business actually runs on: customer acquisition cost, customer lifetime value, revenue per cohort, margin contribution.

This type of analysis is harder to do because it requires integration between marketing data and commercial or finance data that often sit in separate systems with separate owners. But it is the level at which measurement becomes genuinely strategic. When you can show the relationship between marketing investment and customer lifetime value by acquisition channel, you have a basis for making budget decisions that the finance team can engage with on their own terms.

Growing an agency from 20 to over 100 people taught me that the marketing conversations that got traction with the CFO were never about impressions or click-through rates. They were about cost per acquired customer relative to the value of that customer over time. That framing is available to any marketing team. Most do not use it because it requires access to data and a willingness to be held to a commercial standard.

How Do You Choose the Right Measurement Method?

The honest answer is that you do not choose one. You triangulate across several, because each method has blind spots that others can partially compensate for.

A reasonable starting point for most businesses is: use MMM for strategic budget allocation decisions, use MTA for tactical optimisation within channels, use incrementality testing to validate the assumptions that both models make, and use brand tracking to capture the longer-term effects that neither attribution nor modelling can see clearly. That combination is not perfect. Nothing is. But it is more honest than relying on any single source.

The question of which methods to prioritise also depends on your business model, your data infrastructure, and the decisions you are actually trying to make. A direct-to-consumer brand with strong transaction data and a short purchase cycle has different measurement needs than a B2B software business with a six-month sales process. The structural differences in how B2B markets operate affect not just strategy but how you define and measure marketing’s contribution.

There are also practical constraints. Incrementality testing requires sufficient volume to produce statistically meaningful results. MMM requires at least two years of consistent historical data to be reliable. Brand tracking requires budget and a commitment to longitudinal measurement. Most businesses cannot do all of this simultaneously, and that is fine. The goal is to move progressively toward a more honest picture, not to achieve measurement perfection.

What Are the Most Common Measurement Mistakes?

The most common mistake is treating platform-reported metrics as ground truth. Every platform reports performance in the way that makes that platform look most valuable. Google reports conversions using its own attribution window. Meta reports reach using its own methodology. These numbers are not fabrications, but they are perspectives shaped by the platform’s commercial interests. Taking them at face value without independent validation is a category error.

The second most common mistake is measuring activity rather than outcomes. Impressions, clicks, sessions, and engagement rates are easy to measure and easy to improve. They are not the same as business outcomes. A campaign that drives a 40% increase in website traffic but no measurable change in revenue has not performed well. It has generated activity. There is a difference.

The third is failing to account for the counterfactual. What would have happened without this marketing activity? This is the question that attribution models almost never ask and that incrementality testing is specifically designed to answer. Without some version of the counterfactual question, you cannot know whether your marketing is driving growth or simply accompanying it.

Think about it this way: someone who walks into a clothes shop and tries something on is far more likely to buy than someone who just browses. But the act of trying something on does not fully explain the purchase. Some of those people were going to buy regardless. Marketing has the same problem. Presence in the experience does not equal causation of the outcome. Good measurement is the discipline of separating the two.

Tools that support growth analysis can help operationalise some of this, particularly for teams that are building measurement capability from a relatively early stage. But the tools are secondary to the thinking. Getting the measurement orientation right matters more than the specific software you use to implement it.

How Should Measurement Connect to Strategy?

Measurement that does not connect to strategy is reporting. It describes what happened. It does not inform what to do next.

The connection between measurement and strategy happens when you use measurement outputs to make specific decisions: which channels to invest in, which audiences to prioritise, which creative approaches are generating genuine commercial return. That requires measurement to be set up with those decisions in mind from the start, not retrofitted after the campaign has run.

One of the most valuable shifts I made in how I approached measurement was starting with the decision, not the metric. What decision are we trying to make? What data would change that decision? How do we get that data in a form that is reliable enough to act on? That sequence, decision first, metric second, data third, produces much more useful measurement frameworks than the alternative, which is collecting everything available and hoping a story emerges.

Strategic planning frameworks that connect activity to commercial outcomes share this orientation. The measurement question is not separate from the strategy question. It is embedded in it from the beginning.

Measurement also needs to operate across different time horizons. Short-term measurement tells you whether a campaign is working. Medium-term measurement tells you whether your channel mix is right. Long-term measurement tells you whether your marketing is building the kind of brand equity and customer relationships that compound over time. Most businesses are reasonably good at the first, adequate at the second, and poor at the third. The third is where the most value sits.

If you are building out a full growth strategy and want to understand how measurement sits within the broader commercial architecture, the Go-To-Market and Growth Strategy hub covers how these pieces connect, from market entry decisions through to performance measurement and optimisation over time.

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