How to measure sales uplift of individual media channels?

When running a successful marketing campaign, it's important to understand how each media channel contributes to your overall sales and success. This can help you optimize your marketing strategy, allocate your budget more effectively, and ultimately drive better results for your business.

Sales uplift or lift can be measured through a variety of methods, including controlled experiments, A/B testing, attribution methods, and Marketing Mix Modeling (MMM). The goal of measuring sales lift is to determine the true impact of a marketing campaign on sales, taking into account other factors that may influence sales, such as seasonality or external events.

As previously mentioned, here are a few effective approaches you can take to measure the sales uplift of individual media channels. One method is to use a controlled experiment, where you compare the sales of a group of customers who were exposed to a specific marketing channel (such as a TV ad or email campaign) to a group of customers who were not exposed to that channel. By comparing the sales data of these two groups, you can determine the incremental sales lift that the marketing channel provided. Similarly, you can perform A/B testing for two specific audience groups.

Alternatively, you can use MMM to measure your sales lift as part of your overall marketing measurement strategy. MMM analyzes your sales data to determine which media channels were most effective in driving sales. This approach allows you to see how each channel contributes to the customer journey and gives you a better understanding of the overall impact of your marketing efforts. In the following sections we will dive more into these methods.

What is “sales lift” in Marketing Mix Modeling?

Essentially, sales lift is a metric used to measure the incremental increase in sales that can be attributed to a specific marketing campaign or media channel. It is the increase in total sales you observe during a campaign in comparison to the level of sales you would have archived without running the campaign.

Simply put, sales uplift can be calculated like this: Total Sales - Base Sales = Sales Uplift

Base sales refer to the level of sales that your company would achieve without any additional marketing investment. It represents the existing demand for a company's products or services that is driven by factors such as brand awareness, reputation, product quality, and customer loyalty. Therefore, if you would stop all your marketing investments immediately, you would still, hopefully, have a certain amount of base sales.

When a company invests in marketing, the goal is to increase sales beyond the base level by attracting new customers, increasing purchase frequency, or increasing the average order value. The increase in sales achieved through marketing efforts is referred to as incremental sales, which is the difference between total sales and base sales. This incremental sales, is essentially the sales uplift, because it represents the additional “lift” on top of the base sales.

Let’s visualize it:

In the chart above you can see an example of sales timeline during a 12 months period. The grey section represents the base sales and the colorful sections represent the sales lift across the different media channels (like Youtube, Tiktok, Pinterest etc). Hence, everything above the base sales represents your sales uplift. You can try our interactive demo (including the chart above) here.

Why is "sales lift" measurement important?

Understanding sales lift is important because it provides a benchmark for measuring the effectiveness of marketing campaigns. By comparing incremental sales to base sales, businesses can determine the true impact of their marketing efforts and optimize their strategies for maximum ROI. Additionally, understanding sales lift can help businesses set realistic sales goals and allocate their marketing budgets more effectively.

Shortly, the sales lift metric can help you:

1. Measure the effectiveness of your marketing campaigns and media. By measuring the sales lift generated by your marketing campaigns and media, you can determine the true impact of your marketing efforts and identify the channels or tactics that generate the highest ROI.
2. Optimize your marketing strategy. You can allocate your resources more effectively and optimize your marketing strategy for maximum impact once you understand which channels generate the highest sales lift.
3. Set realistic sales goals. Understanding your sales lift can help you set achievable sales goals based on the incremental sales you expect to generate from your marketing campaigns.
4. Make data-driven decisions. You can make informed decisions about where to allocate your marketing budget and how to optimize your marketing efforts to generate the highest possible return on investment.
5. Provides a benchmark for evaluating future marketing campaigns. By understanding the sales lift generated by past marketing campaigns, you can use this information as a benchmark for evaluating the effectiveness of future campaigns and identifying areas for improvement.
6. Justify your marketing investment. Sales lift measurement can help you demonstrate the value of your marketing investment to stakeholders and justify future marketing budgets based on the incremental sales generated by your campaigns.

How is sales lift measured?

As mentioned previously, the sales uplift can be measured in multiple ways. Let’s look at the most common ones.

Controlled experiments

In a controlled experiment, a group of customers is randomly assigned to either the treatment group (exposed to the marketing message) or the control group (not exposed to the marketing message). Sales data is then collected from both groups, and the difference in sales is calculated to determine the sales lift generated by the marketing campaign. However, depending on your market size, the results may not be generalizable to the larger population. Moreover, it may be difficult to isolate the impact of the marketing message from other factors that could affect sales (e.g., seasonality, external events) and it could be expensive to continuously conduct experiments in order to get the necessary data.

A/B testing

A/B testing involves testing two variations of a marketing message or tactic and measuring the difference in sales between the two variations. This allows you to determine which variation generates the highest sales lift. Similarly to the controlled experiments, A/B testing might not be a sustainable long term method of continuously measuring the sales uplift.

Multi-Touch Attribution Modeling (MTA)

MTA follows the customer journey through touch-points (likes, clicks, impressions etc) and attributes successful conversions to the different media channels. You could identify the channels or tactics that generate the highest sales lift, and potentially, you can optimize your marketing strategy for maximum ROI. Some downsides of MTA are:

• It may not capture the full impact of marketing activities on sales (e.g., brand awareness).
• Possibly, it might not account for interactions between different marketing channels. For instance, attribution completely ignores the importance of concurrent TV campaigns in favor of the last touchpoint, such as a SEA click.
• It may be difficult to accurately attribute sales to specific marketing activities.

Marketing-Mix Modeling (MMM)

MMM is a statistical method used to measure the impact of various marketing channels on sales. It involves analyzing historical sales and marketing data to determine the contribution of each channel to sales. MMM measures sales lift by using historical sales data, preferably, with a high granularity Meaning, MMM looks at daily historical data in order to understand the performance of your investments.

Some of the downsides of MMM are:

• It could be costly to develop on your own due to the high technical expertise needed and also the lengthy time frame. If you’re interested in MMM, a SaaS provider or an open-source framework might be more effective. Here we compared some of the most common ones with SaaS. However, MMM provides many other additional insights aside from sales lift. Therefore, it might be more cost effective if you look at all the other use-cases for MMM.
• You need historical data in order to be able to carry out MMM, and, depending on the media channels you use, you might need to have around 2 years of historical data to properly account for yearly seasonalities. Normally, this is something your marketing agency or the digital ads platform should provide you with.

How to measure sales lift on individual media channels with MMM?

Essentially, measuring the sales lift on individual media channels implies looking at each media channel you are using (for example TV or YouTube) and seeing the sales lift for that specific channel. Finally, you can see the total sales lift if you sum up all the channels you are using.

We want to focus on MMM when discussing the sales lift measurement on individual media channels because this is what we know best, and, the previously mentioned solutions are highly customizable and will vary from company to company.

The following steps are involved in measuring sales lift using MMM:

• Collect data on historical sales and marketing activities.
• Segment the data by channel (e.g., TV, online, print) and analyze the impact of each channel and the sales data by desired sales dimensions (e.g. sales channel, customer segment, product category, etc).
• This is where the magic happens. MMM uses statistical methods to establish an estimate of the baseline sales and the estimates of the media uplifts.
• Analyze and compare the relative performances of the various media channels, utilize the results in determining the profitability of past investments and media budget optimization.

MMM measures sales lift by using historical sales data with a high granularity. It looks at daily historical data in order to understand the performance of your investments on each media channel. This approach allows you to see how each channel contributes to the customer journey and gives you a better understanding of the overall impact of your marketing efforts on each channel. By analyzing this data, you can determine which channels were most effective in driving sales and allocate your resources accordingly.

MMM looks at historical data to determine the impact of marketing activities on sales. By analyzing trends in sales and marketing data over time, MMM can identify the drivers of sales and forecast the impact of future marketing activities. To put it simply, by analyzing historical data MMM can understand the variations in data during a time frame down to a daily level. For example, you could understand the difference in performance of different media channels on a daily level or even compare the same day in two years.

We believe we clarified that MMM relies on a wide range of data sources to measure the impact of marketing activities on sales. This can include data on advertising spend, website traffic, customer demographics, and sales performance, among others. However, MMM can also include brand KPIs, impressions etc. Essentially almost all sales and marketing related data you posses. Moreover, MMM can also include offline data such as TV and Radio to avoid over-attributing sales lift to last-click media, such as SEA.

MMM can attribute sales to specific marketing channels (such as ad spend, website traffic, and sales) and determining the extent to which each variable affects sales. This allows marketers to see which channels are most effective at driving sales and which ones are underperforming. Armed with this information, marketers can optimize their marketing mix to maximize sales lift by reallocating budget from underperforming channels to more effective channels, marketers can drive better results and improve ROI. We talked more about marketing budget optimization in our blog post about Diminishing Return Curves.

What are the benefits of measuring sales lift on individual media channels with MMM?

1. Accurate measurement: MMM is a highly accurate way to measure the impact of each media channel on sales. It uses statistical models to isolate the effects of each channel, taking into account factors such as seasonality, price changes, and external events.
2. Granularity: MMM provides a high level of granularity in terms of measuring sales lift. It can analyze daily or weekly data to provide insights into how each media channel is performing.
3. Optimization: By using MMM to measure sales lift on individual media channels, marketers can optimize their marketing strategy to focus on the channels that are most effective at driving sales. This can improve the overall ROI of the marketing campaign.
4. Cost-effective: Compared to other methods of measuring sales lift, such as A/B testing or randomized controlled trials, MMM is generally more cost-effective. It allows for analysis of historical data rather than requiring the creation of new experiments. Moreover, it offers additional functions aside from sales lift measurement.

Curious to learn more? Book a demo.