Incremental Marketing: Stop Crediting Conversions That Were Going to Happen Anyway

Incremental marketing is the discipline of measuring and driving growth that would not have occurred without a specific marketing action. Not total conversions. Not attributed revenue. Growth that is genuinely new, caused by your activity, and would not have happened otherwise.

It sounds obvious. It is, in practice, one of the hardest things to get a marketing organisation to take seriously, because honest incrementality tends to make the numbers look considerably worse before it makes the business perform considerably better.

Why Most Marketing Measurement Is Telling You a Comfortable Lie

Earlier in my career, I was a true believer in performance marketing. We had dashboards, attribution models, CPA targets, and ROAS benchmarks. The numbers looked good. The business grew. I assumed those two things were connected in the way the dashboards implied.

It took years of managing P&Ls, running agencies, and watching what actually happened when budgets were cut or channels were switched off, to understand that a significant portion of what performance marketing claimed credit for was going to happen anyway. The customer had already searched. They had already decided. We put an ad in front of them at the moment of intent, they clicked, they converted, and our system called it a win. But we did not cause that purchase. We just showed up at the right moment and took the credit.

This is not a niche technical problem. It is a structural flaw in how most marketing teams evaluate their own effectiveness, and it has real consequences for where money gets allocated and whether brands actually grow.

What Incrementality Actually Means in Practice

The cleanest definition of incrementality is the counterfactual question: what would have happened if we had not run this campaign, placed this ad, or sent this email? The difference between that hypothetical baseline and your actual result is your incremental contribution.

In practice, measuring this requires either holding out groups (audiences deliberately not exposed to a campaign, so you can compare their behaviour against those who were), geo-based experiments, or media mix modelling that attempts to isolate the causal contribution of each channel. None of these methods is perfect. All of them are more honest than last-click attribution or platform-reported ROAS.

When I was running performance campaigns across multiple verticals, we started running holdout tests on branded search. The hypothesis was that people searching for our client’s brand name were going to buy regardless of whether we ran a paid ad above the organic result. The test confirmed it. Incremental conversions from branded search were a fraction of what the platform reported. We reallocated that budget into upper-funnel activity. Revenue held. Efficiency improved. The dashboard looked worse. The business performed better.

That experience shaped how I think about measurement permanently. The goal is honest approximation, not false precision. If your numbers look suspiciously good, they probably are.

For a broader view of how incrementality fits within growth strategy, the Go-To-Market & Growth Strategy hub covers the commercial frameworks that sit around these decisions, from market penetration to demand creation.

The Lower-Funnel Trap

Performance marketing, particularly paid search and retargeting, is structurally biased toward low incrementality. That is not a criticism of the channels. It is just how they work.

Paid search captures existing intent. Someone has already decided they want something. They type it into a search engine. You appear. They click. The platform records a conversion. But you did not create that intent. You intercepted it. The question is whether they would have found you anyway through organic results, direct navigation, or a competitor, and whether the cost of interception was worth the margin on the sale.

Retargeting has the same problem, often more acutely. You are showing ads to people who have already visited your site, which means they already know you exist and have already shown interest. Some of them will convert regardless of whether you spend money following them around the internet. The incremental contribution of the retargeting spend is often far lower than the attributed conversion volume implies.

This does not mean you should abandon these channels. It means you should be honest about what they are doing. They are efficiency tools for demand that already exists. They are not growth engines. Confusing the two is how brands end up with optimised performance metrics and flat revenue.

Understanding the difference between capturing demand and creating it is central to thinking about market penetration strategy. Most brands optimise for the former when the latter is where the growth is.

Where Genuine Incremental Growth Comes From

If lower-funnel activity largely captures demand that already exists, then genuine incremental growth has to come from somewhere else. It comes from reaching people who were not already going to buy from you.

There is a useful way to think about this. Consider a clothes shop. A customer who walks in and tries something on is dramatically more likely to buy than one who walks past the window. The act of trying it on does not just measure existing intent, it creates new intent. The product, the experience, the moment of physical engagement, shifts someone from passive to active. Marketing at its best does the same thing. It introduces people to something they were not already looking for and makes them want it.

This is what genuinely incremental marketing looks like. It is not showing an ad to someone who has already searched for your product. It is reaching someone who has not yet thought about your category and making them think about it. It is brand building, awareness campaigns, content that earns attention rather than buying it at the point of decision.

When I was growing an agency from around 20 people to over 100, the growth that actually changed the trajectory of the business came from being visible in places where potential clients were not already looking for us. Speaking at events. Publishing thinking that attracted inbound interest. Being present in conversations before the brief was written. None of that showed up cleanly in an attribution model. All of it was incrementally valuable.

Brands that want to understand how to structure this kind of activity can look at how Forrester’s intelligent growth model frames the relationship between demand creation and demand capture. The balance matters more than most organisations acknowledge.

How Incrementality Thinking Changes Budget Decisions

The most immediate practical impact of taking incrementality seriously is that it changes where money goes. Channels that look efficient on an attributed basis often look considerably less efficient when you apply incrementality thinking. Channels that look expensive or hard to measure often have higher genuine contribution than the dashboards suggest.

This creates a genuine organisational challenge. Finance teams, boards, and even marketing leaders who have grown up in a performance culture find it uncomfortable to fund activity that cannot produce a clean cost-per-acquisition number. The pressure is always to put more money into what the platform says is working. Incrementality thinking pushes back against that pressure by asking whether the platform’s definition of “working” is actually connected to business outcomes.

I have sat in budget reviews where a channel with a reported ROAS of 8x was being championed for increased investment, and the incrementality data suggested the true contribution was closer to 2x. The difference was not fraud or incompetence. It was attribution methodology capturing organic and direct traffic that would have converted anyway and assigning it to the paid channel. Presenting that data to a CFO who has been told the channel is performing at 8x is not a comfortable conversation. But it is the right one.

Incrementality-adjusted budgeting tends to shift investment in three directions: more toward upper-funnel brand activity, more toward channels with demonstrable holdout-tested lift, and less toward branded keywords and retargeting pools that are largely capturing intent rather than creating it.

The Measurement Methods Worth Understanding

There are several approaches to measuring incrementality, each with different requirements and trade-offs.

Holdout testing is the most direct method. You split your audience into an exposed group and a control group, run your campaign for the exposed group only, and compare conversion rates. The gap is your incremental lift. This works well for digital channels where you can control exposure at the user level. It requires statistical rigour and sufficient volume to produce meaningful results, but it is the closest thing to a controlled experiment that most marketers can run.

Geo-based experiments work on a similar principle but use geography as the unit of analysis. You run a campaign in some regions and not others, then compare outcomes. This is particularly useful for channels like TV, out-of-home, or radio where individual-level holdouts are not feasible. It is also how most serious media mix modelling is validated.

Media mix modelling uses historical data to build statistical models of how each channel contributes to overall outcomes, controlling for external factors like seasonality, economic conditions, and competitor activity. It is not a real-time tool, it requires months of data and significant analytical resource, but it gives you a view of incremental contribution across your entire portfolio that attribution models cannot provide.

None of these methods gives you certainty. All of them give you better information than last-click attribution. The goal is not perfect measurement. It is honest approximation, making decisions based on the most defensible evidence available rather than the most flattering dashboard.

Tools that support growth experimentation, including some of the approaches outlined in resources on growth testing methodologies, can help teams build the infrastructure for this kind of ongoing measurement without requiring a dedicated data science team from day one.

Why Incrementality Is a Strategic Discipline, Not Just a Measurement Technique

The reason I think incrementality deserves attention beyond the measurement community is that it changes how you think about marketing strategy, not just how you report on it.

A team that genuinely internalises incrementality thinking starts asking different questions. Instead of “which channel has the best ROAS?”, they ask “which channel is creating demand that would not otherwise exist?” Instead of “how do we improve our conversion rate?”, they ask “are we reaching people who are not already going to buy?” Instead of “how do we attribute this sale?”, they ask “did our marketing cause this sale or did it just show up at the moment of a decision that was already made?”

These are harder questions. They produce less comfortable answers. They also produce better commercial outcomes over time, because they push investment toward activity that actually grows the market available to you rather than activity that efficiently harvests the market that already exists.

I have judged the Effie Awards, which exist specifically to recognise marketing effectiveness rather than creative execution. The campaigns that stand out are almost always the ones where there is a credible causal story between the marketing activity and the business result. Not correlation. Not attribution. A genuine argument that the marketing caused something to happen that would not have happened otherwise. That is incrementality, whether the teams used that word or not.

Building growth loops that compound over time, rather than just capturing existing demand more efficiently, is one of the principles behind growth loop frameworks that focus on creating self-reinforcing cycles of new audience acquisition. Incrementality thinking is what makes those loops genuine rather than circular.

The Organisational Resistance You Will Face

Introducing incrementality thinking into an organisation that has built its reporting around attributed metrics is not a technical challenge. It is a political one.

Performance teams have built careers on the numbers that attribution models produce. Agencies have built retainers on demonstrating ROAS. Platforms have built entire business models on attribution systems that, not coincidentally, tend to credit their own channels generously. Telling any of these stakeholders that the numbers they have been reporting are inflated is not a message that lands easily.

The approach that tends to work is not to attack existing measurement but to run incrementality tests alongside it. Let the holdout data speak for itself. When a channel that reports 6x ROAS shows 1.8x incremental lift in a controlled test, the conversation becomes much easier to have. The data is doing the work rather than a strategist asserting that the attribution model is wrong.

Scaling this kind of thinking across an organisation requires both the analytical infrastructure and the cultural permission to question comfortable numbers. The BCG framework for scaling agile practices is relevant here not because incrementality is an agile concept, but because the challenge of embedding a new way of thinking across teams with established habits and incentives is structurally similar.

The broader point is this: if your marketing organisation is rewarded for attributed metrics, it will optimise for attributed metrics. If you want it to optimise for genuine growth, the measurement and incentive systems need to change together.

Incrementality and the Question of Brand

One of the reasons brand investment is chronically underfunded in performance-led organisations is that it does not produce clean attribution numbers. You cannot easily draw a line from a TV spot or a brand awareness campaign to a specific sale in the way you can from a paid search click.

But brand investment is, by its nature, incremental. It reaches people who are not yet in-market. It builds familiarity and preference before the purchase decision exists. It means that when someone does enter the market for your category, you are already a known quantity rather than an unknown one. The incremental contribution of that prior exposure is real, even if it is hard to measure in a standard attribution model.

There is a compounding dynamic here that pure performance thinking misses. The more you invest in reaching people before they are in-market, the larger the pool of people who are already familiar with you when they enter the market. That familiarity reduces the cost of conversion and increases the probability that you are considered at all. Performance marketing then works on a better base. The two are not in competition. They are sequential, and the sequence matters.

I have worked with companies that cut brand investment entirely during difficult periods and doubled down on performance. The short-term numbers held up. Twelve to eighteen months later, the consideration metrics had eroded, new customer acquisition costs had risen, and the performance channels were working harder for less. The incrementality of the brand investment only became visible when it was absent.

Understanding how pricing and go-to-market decisions interact with brand perception is another dimension of this, and the BCG analysis of pricing in go-to-market strategy touches on how brand equity affects commercial outcomes in ways that do not show up in short-term attribution data.

If you are working through how incrementality fits into your broader growth architecture, the Go-To-Market & Growth Strategy hub covers the connected decisions around market selection, demand creation, and commercial prioritisation that determine whether incremental thinking translates into incremental results.

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