OPINION Decisions, decisions: the importance of incrementally

InternetRetailing
Image: Fotolia money profit incremental growth

The basic thesis of all marketing activity is that it must have an impact on sales, or else we are wasting our time on vanity metrics. Incremental sales are sales that would not have occurred without a marketing action. Retail Media is a form of marketing action and also the fastest growing advertising channel. Effective incrementality measurement of retail media is investments will be essential to its continued growth.

Incrementality obstacles

Let’s start with the obstacles to incrementality measurement. These include:

  1. lack of understanding of why measurement of incremental sales is important (!)
  2. lack of data or resources to measure incrementality
  3. lack of agreement of how incrementality should be measured
  4. lack of communication on how incremental sales are lower than ‘attributed’ sales
  5. lack of the right tools? Econometrics is expensive and all depends on your choice of baseline demand.
  6. knowing whether we are getting new or repeat buyers – which can vary based by cookie window!

Incrementality approaches

What are the approaches to measure incrementality? 

Experts all have their own opinions on which methods are best when measuring incrementality, but they all involve a combination of testing and experimentation. Paul Dahill of Koddi suggests five approaches:

  1. Geo Holdouts – Run campaigns in some regions while keeping others as control, then compare sales lift.
  2. A/B Testing – Split audiences randomly into test (exposed) and control (unexposed) groups, then measure the difference in outcomes.
  3. Pre/Post Analysis – Compare sales before and after a campaign, adjusting for seasonality and trends.
  4. Matched Market Testing – Select similar markets, run media in one but not the other, and analyze the lift.
  5. Media Mix Modelling (MMM) – Statistical modelling to quantify the impact of media spend while accounting for other marketing factors.

For digital media, Dahill points out that A/B Testing is the most rigorous for measuring true incrementality, directly measuring causality, but is harder to execute at scale.

Sample incrementality solutions

Kroger Precision Marketing (KPM) is bringing incremental sales reporting to self-service programmatic campaigns placed through The Trade Desk across display, streaming, audio advertising.

The new incremental measurement capability measures total and incremental uplift of sales (including performance by key audience segments and modality), household penetration, units, and visits.

KPM’s incremental measurement offers post-campaign insights into the sales lift generated by offsite advertising efforts. They offer “in-flight attributable” sales measurement to optimise campaigns in near real-time and at the conclusion of the campaign. 

To achieve this, the Kroger approach uses a control audience that mirrors the exposed audience, ensuring both groups share identical purchase behaviours except for the media exposure. By isolating and measuring the difference, Kroger Precision Media believe that advertisers can confidently real campaign effectiveness.

The new incremental measurement capability measures total and incremental uplift of sales (including performance by key audience segments and modality), household penetration, units, and visits.

Commerce media adtech vendor, Koddi introduced managed-service incrementality testing and reporting and is now introducing self-serve with real-time experiment reporting and statistical significance insights.  Koddi are also expanding measurement capabilities to integrate directly with advertisers’ existing media mix modelling.

Coles 360 Media in Australia, the retail media arm of one of Australia’s largest retail chains launched a new comprehensive measurement solution, Coles 360Impact, designed to provide a 360-degree view of campaign performance powered by Circana.

The new capability usess two incremental techniques, namely customer and market lift. To start with Coles 360 answers sales incrementality, across all the [Coles] channels and optimisation within certain channels as well?.

The partnership with Circana allows control groups to be set up for addressable media (online shopping and loyalty members) and one-to-many channels like in-store digital screens, radio and supermarket aisle activations.

Incrementality can be Subjective

One of the questions not delivered with all the conversations about incrementality is around the depth – because, like many things in marketing and advertising, a lot is subjective, as pointed out in a recent conference.

SOURCE: Tinuiti Retail Media Summit 2024

Making decisions based on incrementality – a simple guide

Even if you understand incrementality, what are your going to use it for: the answer is better marketing decisions. That’s the topline outcome. But there are distinctions within this topic.

The key question isn’t whether a single ad led to a sale—it’s whether your overall investment in a campaign, ad product, or strategy is driving additional sales that wouldn’t have happened otherwise.

To make informed decisions, start by identifying the level at which you need insights. 

Professor Koen Pauwels argues that there are three typical levels at which you can and do make decisions on incrementality:

1. Campaign-level decisions: expand or cut?

If you’re deciding whether to increase or decrease spend on a specific campaign, digital Multi-Touch Attribution (MTA) is often the best tool. MTA tracks individual consumer touchpoints and can use randomised control trials (RCTs) to measure whether a campaign is driving incremental sales.

  • Example: You test whether boosting a retail media campaign spend by 10% results in a proportional lift in sales.

2. Ad product-level decisions: where to allocate budget?

If you’re deciding how to split budget across ad types, such as Sponsored Search vs. Sponsored Display, Marketing Mix Modeling (MMM) is typically more useful. MMM looks at overall trends, including ad types that don’t have direct customer tracking, and factors in external influences like seasonality, economic trends, or weather.

  • Example: You might find that Sponsored Display takes longer to drive impact compared to Sponsored Search but provides greater long-term sales lift.

3. Product, Brand, or Category-Level Decisions: What to Advertise?

If you’re deciding which products, brands, or categories to prioritize, incrementality analysis helps determine whether spending should focus on:

  • A low-performing SKU that needs visibility vs. a high-performing SKU with organic traction.
  • Building brand awareness vs. driving direct product sales.
  • Growing an entire category (e.g., increasing demand for pizza in general) vs. winning market share within that category.
  • Example: If advertising a new product doesn’t significantly increase overall sales but just shifts revenue from an existing product, it might not be a smart investment.

Key takeaway

Incrementality is not about individual ads: incrementality is key to understanding whether your marketing investments are truly creating additional sales at the campaign, ad product, or category level. However, Paul Dahill of Koddi says “should the industry use them on every campaign? If we track everything against incrementality, do we risk optimising only for the short-term? Could this approach overlook long-term brand impact, halo effects, and the value of upper-funnel marketing?”

One thing is true: the conversation about incrementality will continue forever. The simple question to prove the point is this: have arguments about metrics, measurement and incrementality have gone away for Google or Meta even though they believe they have robust answers? (Answer: No!)

Read More

Register for Newsletter

Created with Sketch.

Receive 3 newsletters per week

Created with Sketch.

Gain access to all Top500 research

Created with Sketch.

Personalise your experience on IR.net