Mobile marketing analytics for precise targeting with coupons

by Julia Pott

While cross-market user behavior on websites is already well documented and analyzed, user journeys in mobile applications remain a black box. This is where the start-up Murmuras comes in, offering in-depth insights into the use of retail apps. The aim is to provide retail companies with valuable insights for their mobile marketing strategies.

Murmuras 2024 was awarded the prestigious German science prize for this pioneering work. Reason enough to find out from co-founder Qais Kasem how these findings are obtained, what retailers can learn from them and what trends in mobile commerce Qais is observing.

Qais, you at Murmuras promise your customers valuable insights for their mobile marketing. How do you do that?

Qais Kasem: We investigate how consumers use retail apps. We do this in collaboration with market research companies by paying people to allow us to measure their digital activities. This allows us to collect a large amount of data, which we then process. We use this information to generate insights that we offer to retailers and manufacturers.

Which aspects of app use are you interested in?

Kasem: For example, we record which coupons the participants see and use, whether and for how long they look at the brochures or which promotions they take part in. We also know what customers are buying – both online in the app and in stores using digital till receipt data. From 2025, paper receipts will also be added as a data source via scanning.

Can a retailer who wants to use this data filter within this group according to certain factors?

Kasem: Yes, that is possible. Let's say a retail company is interested in the behavior of the user group of women aged between 30 and 39 who have a child, then we can recruit specifically for these criteria if required.

We currently have 3,000 active users of retail apps in Germany. We could use this database to serve the following request, for example: A retailer would like to understand how coupons are used in the major food retail apps such as Lidl Plus, REWE or EDEKA. We have 500 users per app and therefore gather very good insights: In our experience, you can get very far with this high-quality data set.

How does the measurement work?

Kasem: The technology was developed over many years at the University of Bonn. We started out in science, where our focus was on behavioral analysis in connection with smartphone addiction research. In the meantime – within a year of launching the product – retail apps have become our main market. This shows how great the need for these insights is.

One challenge here is to only read out the data that interests us and that we are allowed to collect in this context. The second step is to generate insights from this, which is where artificial intelligence helps us.

What new insights into mobile marketing does a retail company gain from your analysis?

Kasem: Our unique selling point is that we provide insights into what user behavior looks like in other apps. Until now, retailers and providers have only received answers to such questions from their own apps: Which coupons work best in which apps? At what prices are products sold? How are coupons personalized? This is where we bring transparency to the market.

Another advantage of our solution is that we can track customer behavior across apps and gain insights from this: Do customers compare the coupons from the Rossmann app and the dm app before making a decision? Do they only look at the coupons when they are in the store or do they find out about the offers before their shopping trip? How well do competitions or free coupons work? What short-term, but also long-term effect do discount campaigns or seasonal promotions have on customer loyalty?

Why is this type of insight becoming increasingly important for retailers and brands? What trends are you observing?

Kasem: A huge change is taking place with retailer apps: Offers are becoming more and more personalized, meaning that one user receives completely different discounts on certain products than another user. This degree of individualization can only be achieved through targeting on the smartphone.

On the one hand, this happens because providers have more and more data, for example on shopping behaviour, and can therefore personalize their offers in a more targeted way. On the other hand, it has been proven that personalized coupons work better on average than non-personalized ones. This is logical: if I offer coupons for diapers to all users, they are not nearly as effective as if I use them to specifically target parents. If you do individual targeting well, you can use the app to attract people to your store pretty effectively.

What developments are you observing in the mobile commerce sector?

Kasem: This is difficult to predict in the long term, but at the moment we are seeing retailers pushing their own apps massively. New loyalty programs are being set up or existing ones optimized. There are many collaborations with cashback providers. Whether the trend will eventually move in a different direction – for example, one app for everything or an alternative to smartphones – remains to be seen.

How will AI change behavioral analysis in the future?

Kasem: We currently offer a traditional insights platform that provides our customers with statistics, graphics and provider comparisons. Our vision: an AI-supported Insights-as-a-Service platform that – instead of just providing data – gives concrete, automated recommendations for action. An example would be: “Provider XY has personalized its offers, which has had this effect. Adjust the following parameters to achieve similar success.” Retailers could even ask questions such as: “How do I increase sales of natural cosmetic lipsticks?” – and our AI provides targeted answers.

An AI can only ever be as good as the database. Thanks to our unique data from retailer apps and consumer behavior, this vision is becoming reality: intelligent insights that not only inform, but also directly enable action.