What is Incrementality? True Marketing Impact Guide

Incrementality, in a marketing sense, refers to the additional business results that can be directly attributed to your marketing. Consider it this way: the sales, the conversions, the revenue that would not have happened had you not marketed. It's not traditional attribution, which is about correlation, it's incrementality that's about causation. The big question that incrementality addresses is "What extra value did my marketing create?"

This more scientifically grounded approach has evolved into the gold standard for marketing measurement. As per recent industry research, 71% of advertisers now consider incrementality their top key performance indicator (KPI) for obtaining a holistic understanding of real marketing impact in a privacy-first universe. The reason? It is one of the few ways you can demonstrate the impact that your marketing dollars are actually making.

Understanding incrementality vs. attribution in marketing

Where they differ is in what they measure: incrementality and attribution. Attribution is about giving credit to marketing touchpoints that the customer encountered on their journey, and telling you which channels they interacted with before they converted. Essentially it is a correlation-based measurement that cannot account for people who would have converted anyway and were not really affected by marketing.

Incrementality, in contrast, assesses causality through controlled experiments. As Avinash Kaushik, who spent 16 years at Google and now works at Croud, explains: "Attribution and Incrementality are not the same thing. Chalk and cheese. Incrementality reveals which Conversions would not have happened without various marketing tactics."

Here's an example from actual client studies at Measured: A customer visits a brand's website with purchase intent. In the days to come, the brand serves a retargeting ad, and the attribution credit for the sale goes to the retargeting ad. But in an incrementality test, you might discover that this was actually a customer who would have purchased anyway, the ad didn't incrementally cause the purchase.

This is important because, as Kaushik provocatively puts it, "you could fire everyone in marketing at every company and profits go up. The truth is most of us do marketing that is not incremental."

The practical implications are significant. Attribution tells where your conversions came from, incrementality tells which conversions your marketing actually caused. This is made even more critical as privacy rules get rid of tracking as we know it. Google's official research notes that incrementality testing "is now the industry's gold standard for understanding advertising's true impact in a privacy-first way."

How incrementality testing plays out in practice

Incrementality testing functions like a randomized controlled trial in the medical field. The process splits your audience, or your markets, in two: A test group to which you expose the marketing campaign and a control group that isn't. By contrasting the results between these groups, you can strip out the true causal impact of your marketing.

There are three main ways to perform incrementality testing:

Geographic testing

This method splits markets into treatment and control areas, which is ideal for privacy-driven measurement and for larger campaigns. This approach paid off for local services provider ServicePro who found 40% of their Google Ads leads were genuinely incremental, delivering $100,000 in extra revenue from $25,000 monthly spend, a 400% ROI that would have been hidden by attribution alone.

User-based testing

This randomly places users into a testing group and a control group. Meta's test methodology has brought promising results, advertisers using their Incremental Attribution feature have seen an average increase of over 20% in incremental conversions, from a test group of 45 advertisers across 11 verticals.

Synthetic control methods

These are the forefront of incrementality testing, leveraging machine learning to build artificial control groups from historical data. This methodology makes it possible to measure on an ongoing basis without interrupting any campaign. As more platforms democratize these advanced techniques, Google has lowered testing minimums to only $5,000 with Bayesian testing compared to tens of thousands previously required.

Kate Minogue, Head of App and Gaming at Meta's Marketing Science team, stresses this scientific validity:

i

"What we are trying to avert is attributing all of your impact to an ad that just so happened to be at the very end of a touchpoint but really didn't make a difference. That's where experiment design and incrementality come into play, it's using scientific process to properly validate and evaluate what would've happened had the ad not been there."

— Kate Minogue, Head of App and Gaming at Meta's Marketing Science team

Real examples showing incrementality in action

Real world use cases show the value of incrementality.

Shinola, the Detroit-based luxury lifestyle brand, was among the many companies that encountered that quandary when Apple privacy changes sent Facebook's reported conversions into a nosedive. Through geo-matched market testing at the zip-code layer via Measured's testing platform, they found out that their Facebook campaigns were actually driving a 14.3% incremental lift, but Facebook's own platform was under-reporting performance by 413%. This knowledge empowered them to keep the faith in their Facebook investments despite low-looking platform metrics.

Greenworks, the outdoor tools brand, used insights on incrementality amid tough weather delays and economic headwinds. For their Prime Day campaign in 2022 they teamed up with CommerceIQ and used inventory position, margin and incrementality data to intelligently automate their advertising bids. The numbers were jaw-dropping: 225% year-over-year growth in ad-attributed sales, a 42% spike in total ROAS and 69% more new-to-brand shoppers. A single campaign did $3M+ in sales on only $20k in spend.

The most sobering example is probably from Uber, who learned from a well-cited incrementality test that their Meta ads were not driving new riders like attribution indicated. Leveraging the large-scale holdout testing, they found minimal incremental impact and diverted the $35 million to other uses. This case study is legendary in marketing circles because it serves as a cautionary tale about the disconnect between attributed and incremental value.

Monday.com, the B2B SaaS platform, uses incrementality as a calibration for their attribution model. They found that for some cases, there are as much as 100% discrepancies between platform attribution and actual incremental impact. And by running always-on Conversion Lift Studies, they were able to see 20% reduction in blended cost per acquisition, showing how incrementality can complement your existing measurement systems, not replace them.

The business value of measuring incrementality

The bottom line of incrementality is much more complicated than just optimization. In a deep-dive study by Forrester's comprehensive research, brands that run comprehensive incrementality testing can improve their marketing ROI by 30%. This enhancement is accomplished through three primary value drivers that are directly attributed to the bottom line.

1. Budget optimization provides the most immediate benefit. By knowing which channels really generate incremental results, advertisers are able to shift spends from underachieving areas to those high-impact campaigns. The data supports this: a Lifesight market analysis estimates the loss of $100 billion in media spending in 2023, resulting in no return from poor measurement practices. Companies that use incrementality testing see consistent 15-25% uplift in overall marketing efficiency through better allocation.

2. Reduced waste is equally important. About 20–40% of attributed conversions are estimated to be organic conversions that would have occurred without exposure to an ad. In an anonymous B2B SaaS company their display advertising appeared to perform well, only to discover the real ROI was not a cumulative positive return but actually only a 1.2% increase in subscriptions, with a ROAS of 62% less than they were expecting. They then reallocated 80% of that budget to content marketing and high intent retargeting.

3. Strategic confidence teaches marketing how to go from being a cost center to a growth-driver. To quote Jason Goldberg, Chief Commerce Strategy Officer at Publicis:

i

"Incrementality is something we can show brands that's 'real math,' that says your investment in a retail media network will make you money, not just theoretical money that you'll never see, but money that actually shows up on your GAAP P&L."

— Jason Goldberg, Chief Commerce Strategy Officer at Publicis

This finance-friendly marketing method makes marketing budgets bulletproof to financial scrutiny.

The move to incrementality also solves the core problem of marketing measurement in 2025. And with 71% of advertisers now identifying incrementality as their most valued KPI, and major platforms, such as Google and Meta, incorporating incrementality into the core of their offerings, the business case is unequivocal. Businesses that succeed with incrementality secure an edge in the market through smarter working of budgets and a greater sense of assurance in taking decisions.

Implementing incrementality in your marketing strategy

Incrementality testing, when done effectively, is a carefully planned and methodical process. Here's a practical guide to doing it, based on insights from industry experts and successful use cases.

Begin with clear goals and hypothesis development

Define what you want to know and how you would use the answer BEFORE you start any test. As Joseph Enever, Senior Research Director from Gartner's marketing research team, contends: "When executed well, marketing leaders can use holdout testing to closely measure the incremental effect of marketing investments in channels and tactics." This includes setting KPIs, whatever that might mean to you, be it revenue, conversions, or customer lifetime value, to track against during testing.

Choose the appropriate methodology for your situation

Geographic tests are a happy medium for most brands in terms of statistical rigor and practical application. Split your markets between test and control groups in a way that makes them comparable, based upon historical metrics, demographics, and market size. Make sure your test groups are sizable enough to have statistical significance which typically requires no less than 30 days for digital campaigns and can take as long as 90 days for complicated B2B sales cycles.

Leverage platform tools to simplify complexity

Native testing is supported by Google's Conversion Lift, Meta's incrementality studies, and Amazon Marketing Cloud. These handle the statistical heavy lifting and will enhance your campaigns. For brands that need more nuanced analysis, third-party platforms, among them Measured, Haus or INCRMNTAL, offer more complex features, such as synthetic control group methodology and cross-channel measurement.

Build organizational buy-in before you get started

Incrementality findings can often be counterintuitive to the marketing performance we assume. Head off all stakeholders, especially those in control of channel results, for results that show lower performance when not evaluated through the attribution-based lens. Frame incrementality as an optimization target, not a performance benchmark.

i

"In today's privacy-first marketing landscape, incrementality testing has become the definitive method for proving true marketing impact. We've seen clients achieve remarkable ROI improvements of 25-40% simply by reallocating budgets based on incremental insights rather than attribution assumptions."

— Tessar Napitupulu, CEO of Arfadia and Digital Marketing Expert

At Arfadia, we've found that clients get the best results by framing incrementality as something that enables them to make the biggest impact, rather than cut spend.

Common challenges and how to overcome them

Although incrementality testing provides valuable insights, it has several practical challenges for marketers. Awareness of these challenges and their resolutions are critical to the success of deployment and obtaining reliable results.

Statistical complexity

This often intimidates marketing teams. The great thing is that advanced stats have been democratized by modern platforms with easy-to-use interfaces. If your team doesn't have any data scientists, choose platform-native solutions that do the calculations for you. For example, Google's advanced Bayesian approach yields robust outcomes with smaller sample sizes and budgets compared to traditional frequentist methods.

Organizational resistance

This emerges when incrementality findings subvert prevailing beliefs on channel performance. Marketers can be reluctant about discoveries that can prove that their channel is underperforming. Solve this issue by bringing all the parties together in creating the test design and concentrating on total business impact rather than individual channel success. "It's really the gold standard to actually show the actual impact, or causal impact, that ads have on driving sales," says Ali Miller, VP Ads Product at Instacart.

Budget constraints

These have traditionally confined incrementality testing to enterprise brands, but that's changing fast. As Google lowers testing minimums to $5,000 and new tools emerge enabling synthetic control methods without holdout groups, incrementality insights are available to mid-market brands. You may want to test your highest-spend channels first, or split the costs of a test with other brands.

Test contamination

This happens when results are affected by factors outside of the study. Big sales events, competitive activity or seasonal peaks can skew incrementality measurements. You can minimize this by controlling when you test, staying clear of big promotional peaks, and by controlling for known factors statistically. Modern systems also use machine learning to automatically compensate for various extrinsic factors.

The secret to navigating these hurdles is to begin with the basics, and to grow in complexity and proficiency gradually. "Easy incrementality testing will be democratized for all brands at a low cost... these tools will be much more available in 2025," says Michael Kaminsky, Co-CEO at Recast. This democratization means that brands no longer require million-dollar budgets or a PhD in statistics to measure true marketing impact.

Wait, what makes this even more thrilling? The pace of technological advancement is so brisk that what once took huge teams and budgets, is within reach of almost any business. And then it gets really interesting...

Frequently asked questions about incrementality

How is incrementality distinct from last-click attribution?
Last-click attribution credits 100% of the conversion to the touchpoint that is the last before the conversion is made, it shows correlation, but not causation. Incrementality quantifies what would not have happened without marketing in a controlled experiment, and thus uncovers true causal impact. Just as last-click might reward a brand search ad with a conversion, incrementality testing might show that customer would have bought anyway, and so that ad spending is non-incremental.

What is the minimum budget required for incrementality testing?
Budget requirements have dropped dramatically. Google Ads enables $5,000 incrementality testing with Bayesian methodologies now. Geo-testing usually costs $50,000-100,000 based on market exposure. Cross-channel-based enterprise solutions typically start in the low six figures annually, but offer complete measurement of all marketing activity.

How long should incrementality tests run?
Of course, time in test is dictated by your sales cycle and the amount of data you need. 30-60 days is the typical minimum window required for most digital campaigns to capture delayed conversions and achieve statistical reliability. For B2B companies with longer sales cycles, it can take them 3-6 months. Don't get into the habit of cutting tests short because it 'looks good', this undermines statistical significance.

Is incrementality testing useful for small companies?
Absolutely. Incrementality is increasingly available for any size business, largely due to platform advancements. Small businesses can use Google's Conversion Lift at a mere $5,000 minimum spend or geo-testing in some very limited markets. The magic is starting with your biggest-impact channels and layering on measurement sophistication over time.

Does incrementality replace marketing attribution?
No, incrementality is complementary to attribution, not instead of it. Leverage attribution for day-to-day insights as well as an understanding of customer journeys, while incrementality provides a periodic validation of true causal impact. At Arfadia, we feel incrementality results can be used to help calibrate and improve your attribution models to measure better over time.

What's the ideal channel for incrementality testing?
Typically paid search and social provide the easiest starting points with clear control mechanisms. Display, video, and connected TV are working nicely for geo-based testing. Simple A/B holdout tests are possible with email marketing. Even notoriously trickier channels such as SEO can be addressed with synthetic control methods, but it would need more advanced approaches.

What is the impact of privacy legislation on incrementality testing?
Privacy laws practically lend themselves to incrementality testing, no tracking of individual users required. Geo-based and cohort-based testing protocols are all GDPR, CCPA, and platform privacy changes friendly. With cookies vanishing and tracking limited, incrementality is more valuable than ever as it is one of the last effective ways to measure marketing impact.

Related marketing measurement concepts

Understanding incrementality requires knowing some other measurement concepts related to it, which are part of the modern marketing analytics toolbox. This integrated approach ensures that all these methods are combined for a complete view on performance.

Related Terms

Marketing Attribution - Framework for assigning credit to marketing touchpoints that lead to conversions
Conversion Rate Optimization (CRO) - Systematic process of increasing percentage of visitors who convert
Customer Acquisition Cost (CAC) - Total cost of acquiring new customer including marketing and sales expenses
Return on Investment (ROI) - Key performance metric measuring campaign profitability and marketing effectiveness

Best practices from marketing measurement experts

The industry's top performers in incrementality testing echo similar sentiments when it comes to how to get the most value out of this measurement strategy. Their learnings are based on managing billions in ad spend, so there's a blueprint to follow.

Develop a testing calendar, instead of ad-hoc experiments. The best-in-class brands plan incrementality tests throughout the year, testing channels and campaigns in a systematic rotation. This method results in continuous learning and avoids testing fatigue, while maintaining statistical rigor. Schedule tests during "typical" business times to avert seasonal bias.

Begin with your hardest questions, not your lowest-hanging channels. Although it may be tempting to experiment where implementation seems the easiest, prioritize testing channels that are consuming the most budget or where attribution feels suspicious. Avinash Kaushik says: "If your internal team and your agency are not reporting channel silo incrementality to you, they are planning to fire you secretly. Because it's so cheap and easy a question to answer."

Build incrementality into existing workflows, not as an independent task. Leverage incrementality results to inform attribution models, validate the results from MMM, and inform quarterly budget reviews. The most successful organizations have baked incrementality in as an ongoing component of their measurement rhythm vs one-off special projects.

Document everything meticulously. Record testing hypothesis, design decisions, external factors during test periods, and post-test actions taken. This becomes very useful in future tests and helps in building institutional memory. Develop a consistent test design and results reporting template to be used for all teams and reporting periods.

Invest in organization-wide training. Some concepts about incrementality may seem counterintuitive, especially when incremental effects do not match up with the data coming from attribution. Training, case study analysis, and methodological transparency all build confidence in results. Concentrate specifically on finance and executive-level decision makers behind the approval of budget decisions through incrementality insights. And don't overlook the significance of ensuring everyone is on the same page.

Looking ahead: The future of incrementality in marketing

The path of incrementality measurement leads to widespread adoption as the standard for measuring marketing effectiveness. Several trends are driving this revolution, transforming the digital marketing discipline.

Artificial intelligence is revolutionizing incrementality testing through automated experiment design, live optimization, and the creation of synthetic controls. Today, platforms deployed for trials are starting to include machine-learning capabilities to find the most effective testing parameters, estimate the needed number of samples, and adjust for confounding factors automatically. The automation lowers the knowledge bar for advanced testing while boosting the accuracy and speed to insights.

Privacy laws drive incrementality adoption as historical tracking methods disappear. With the phaseout of third-party cookies and restricted platform tracking, incrementality is one of the few remaining avenues for measuring accurately. This leaves incrementality-savvy marketers well positioned for success in this post-cookie world as their competition stumbles with measurement.

Democratization continues as costs drop and interfaces become easier. Michael Kaminsky of Recast's predictions for 2025 foresees the incrementality tools "becoming much more available... through open-sourced packages or vendors who package them into easy-to-use, affordable software." This makes incrementality available not just for big brands with big budgets.

Integration becomes seamless as measurement platforms combine incrementality with attribution, along with MMM and other approaches. Marketers will not need to decide between these approaches, they will use an integrated dashboard, displaying attributed performance grounded in rigorous incremental impact. This offers a complete perspective that helps you to make decisions more confidently and optimize cycles more quickly.

Real-time incrementality appears to be the next frontier. Testing today takes weeks or months to generate results, but the evolution of statistical methods and computing power will make it possible to measure incrementality continuously. Just think of changing campaigns based on incremental performance updating every day rather than waiting months for a test to conclude, that's the future we're heading toward.

The digital marketer's takeaway here is simple: incrementality is not "best practice," it's necessary practice in order to succeed in the future. The folks who master incrementality today position themselves as irreplaceable strategic partners who can demonstrate the business impact of marketing. We are living in an age of tightening budgets and growing accountability, and the capacity to prove true incremental value becomes the ultimate competitive edge.

At Arfadia, we see incrementality measurement being as standard in marketing as the process of creative development or the act of buying media. The debate isn't about whether to use incrementality testing, rather it is about how fast you're able to build this capability, to not be behind the curve. Begin small, keep learning and allow those incremental insights to propel you to maximum marketing impact.