Performance teams are constantly arguing about what part of the revenue appeared thanks to advertising, and what part of the revenue would have arisen by itself. This is what describes incrementality in marketing. To explain in simple words, it shows what additional result advertising gives on top of natural demand and organic traffic.
Below we will analyze what incremental sales growth is, how the experiment with the control group works, what metrics to use and what conclusions the business can draw from such tests.
Below we will analyze what incremental sales growth is, how the experiment with the control group works, what metrics to use and what conclusions the business can draw from such tests.
Incrementality: definition in simple words
From a mathematical point of view, an increment is an increment of value. In marketing, the term "incrementality" refers to the share of the result that appeared only because of advertising activity.
In simple words, incrementality answers the question:
How many additional sales, conversions or revenue the company received thanks to the campaign, and not by a coincidence of circumstances.
Let’s imagine that without advertising, customers bought 100 products a day, and after the launch of the campaign it became 130. Then:
So it becomes clearer what incremental sales growth is and how to tie it to a real business result, and not only to clicks on banners.
In simple words, incrementality answers the question:
How many additional sales, conversions or revenue the company received thanks to the campaign, and not by a coincidence of circumstances.
Let’s imagine that without advertising, customers bought 100 products a day, and after the launch of the campaign it became 130. Then:
- 100 — basic demand
- 30 — incremental sales
- Increment itself is and is incremental growth
So it becomes clearer what incremental sales growth is and how to tie it to a real business result, and not only to clicks on banners.
Incremental effect and incremental approach
In English-language sources, the word incremental is often used. They describe an approach in which not the entire volume of sales is estimated, but only the growth associated with a specific activity.
Within the framework of this approach, they talk about an incremental effect. This is the difference between the indicators of two comparable groups:
If the test group brought a revenue of 2.3 million, and the control group — 2 million, then the incremental effect is 300 thousand. The same logic is applied to the number of orders, subscriptions, user registration or other purposes.
Incremental indicators are convenient because they allow you to honestly compare channels. One channel can give a lot of traffic, but almost does not affect the increment, the other is modest in volume, but sells perfectly to additional customers.
Within the framework of this approach, they talk about an incremental effect. This is the difference between the indicators of two comparable groups:
- The test audience saw the advertisement
- The basic control group did not touch
If the test group brought a revenue of 2.3 million, and the control group — 2 million, then the incremental effect is 300 thousand. The same logic is applied to the number of orders, subscriptions, user registration or other purposes.
Incremental indicators are convenient because they allow you to honestly compare channels. One channel can give a lot of traffic, but almost does not affect the increment, the other is modest in volume, but sells perfectly to additional customers.
How to put an experiment: groups, data and time
In order for incrementality to be not a theory, but a working tool, it is necessary to set the experiment correctly. The typical scheme looks like this:
The larger the sample and the more accurate the division, the more reliable the conclusions. For a large retailer, such a test can easily require a budget of several hundred thousand rubles, but it is this scale that provides qualitative data and allows you to accurately measure the increment.
It is important to ensure that users from the control segment do not fall under the same advertising by mistake. Otherwise, the incremental effect will be underestimated, and the conclusions will be inaccurate.
- The target audience segment is selected
- The segment is randomly divided into two parts: test and control group
- The test part falls under the influence of advertising campaigns: sees banners, videos or newsletters
- The control part lives without these touches
- After a specified period of time, compare the results of the groups by pre-selected metrics
The larger the sample and the more accurate the division, the more reliable the conclusions. For a large retailer, such a test can easily require a budget of several hundred thousand rubles, but it is this scale that provides qualitative data and allows you to accurately measure the increment.
It is important to ensure that users from the control segment do not fall under the same advertising by mistake. Otherwise, the incremental effect will be underestimated, and the conclusions will be inaccurate.
What metrics to use and how to calculate the increment
In the classic performance approach, they look at the clicks and the price of traffic. When calculating incrementality, the focus shifts to business metrics:
The sequence of incrementality calculation is simple:
For example, the campaign cost 500,000, and incremental sales brought 900,000 revenue. The approximate payback factor is 1.8. This format can be interesting for business, especially if it gives stable growth during repeated launches.
- Number of orders and sales in pieces
- Revenue and margin
- Key conversions — registration, subscription, shopping cart design
- Share of active users and growth of repeat purchases
The sequence of incrementality calculation is simple:
- We calculate the base: how much money and orders did the control group bring
- We calculate similar indicators of the test
- We find the difference — these are incremental results
- We divide them into advertising costs to understand how effective the experiment turned out to be
For example, the campaign cost 500,000, and incremental sales brought 900,000 revenue. The approximate payback factor is 1.8. This format can be interesting for business, especially if it gives stable growth during repeated launches.
Examples of applying incrementality in marketing
Below - several practical scenarios, where the increment helps to make decisions on the budget.
1. Checking the new media wave
The company launches a new image video. Some regions see the campaign, some don’t. A month later, they compare how sales, average check and repeat visits have changed. If regions with advertising show a noticeable additional effect, then the wave was successful, and the strategy of marketing activity can be scaled.
2. Testing CRM-mailing
The service divides the database into two groups: the first one sends letters with a selection of goods, the second serves as a base. After a couple of weeks, they count how many subscribers decided to buy something, what growth of orders each part showed and how the revenue changed. According to the incremental results, it is clear whether it is worth continuing the mailing and which segment it is better to direct it.
3. Evaluation of a new traffic source
The business connects an additional source of traffic and wants to understand whether it brings new customers or only intercepts people who would buy anyway. For part of the audience, the new source is turned off and the difference in revenue is seen. If the incremental indicators are minimal, it is easier for marketers and business owners to abandon this format and redistribute the budget.
In all these examples, incrementality helps to see where advertising really sells, and where it only creates the appearance of activity.
1. Checking the new media wave
The company launches a new image video. Some regions see the campaign, some don’t. A month later, they compare how sales, average check and repeat visits have changed. If regions with advertising show a noticeable additional effect, then the wave was successful, and the strategy of marketing activity can be scaled.
2. Testing CRM-mailing
The service divides the database into two groups: the first one sends letters with a selection of goods, the second serves as a base. After a couple of weeks, they count how many subscribers decided to buy something, what growth of orders each part showed and how the revenue changed. According to the incremental results, it is clear whether it is worth continuing the mailing and which segment it is better to direct it.
3. Evaluation of a new traffic source
The business connects an additional source of traffic and wants to understand whether it brings new customers or only intercepts people who would buy anyway. For part of the audience, the new source is turned off and the difference in revenue is seen. If the incremental indicators are minimal, it is easier for marketers and business owners to abandon this format and redistribute the budget.
In all these examples, incrementality helps to see where advertising really sells, and where it only creates the appearance of activity.
How to interpret results and build a strategy
After the experiment, it is important not only to count the numbers, but also to read them correctly. Here are some practical tips.
For long-term planning, not one or two one-time tests are important, but the system. Regular measurements show which incremental indicators are stable and how they react to the change of creatives, offers and frequency.
- Look at the relative indicators. One and the same incremental sales can be good for a niche product and weak for mass
- Match with other channels. If one format gives a stronger incremental effect at similar costs, it makes sense to strengthen it in the general media plan
- Consider the context. Sometimes the increase is associated not only with advertising, but also with external events: weather, competitors' promotions, assortment changes
For long-term planning, not one or two one-time tests are important, but the system. Regular measurements show which incremental indicators are stable and how they react to the change of creatives, offers and frequency.
Method limitations and typical errors
Incrementality is a powerful, but not magic tool. He has limitations that are worth remembering.
Nevertheless, it is this approach that shows what activities really move the business forward, and helps to measure the benefits of marketing in clear money.
- Volume is needed. With too small a sample, even one large purchase or an order sold in bulk distorts the picture
- Discipline is needed. If in the middle of the experiment creatives change, new stores open or the price changes dramatically, it will be difficult to interpret the results correctly
- Account for intersections is needed. When the same person sees advertising in several places at once, the incremental effect of one channel is partially "replaced" by the influence of another
Nevertheless, it is this approach that shows what activities really move the business forward, and helps to measure the benefits of marketing in clear money.
Why should marketers and business consider incrementality
For owners and managers, incrementality is a way to make decisions based on data, not the feeling "this campaign was cool".
The approach helps:
For advertising specialists, it is also a language in which it is easy to explain your work. When the report shows how much money the company has received above the base, you can simply show how this or that activity affects the business and why it needs to be continued or scaled.
The approach helps:
- See which campaigns give a real additional effect, and which ones only waste resources
- To protect the budget for the sake of really working activities
- Build a long-term growth strategy without unnecessary risks
For advertising specialists, it is also a language in which it is easy to explain your work. When the report shows how much money the company has received above the base, you can simply show how this or that activity affects the business and why it needs to be continued or scaled.
A short resume in simple words
Let’s sum up and collect the main thing in one block:
This view of marketing makes the work transparent: the marketing team is evaluated by the additional result it brings, not by the number of launched actions. That’s why incrementality is gradually becoming a standard practice for those who want to manage growth, not just watch the numbers in reports.
- Incrementality is the increment of key metrics under the influence of advertising
- The incremental approach is based on the comparison of the test and control groups
- Incremental indicators help to evaluate campaigns by real contribution to revenue and customers, and not only by clicks
- Regular experiments give the company confidence that the money for advertising and other marketing activities is spent in the most effective way
This view of marketing makes the work transparent: the marketing team is evaluated by the additional result it brings, not by the number of launched actions. That’s why incrementality is gradually becoming a standard practice for those who want to manage growth, not just watch the numbers in reports.
