MOBILE RETARGETING
FOR MAMSY
Mamsy — is the largest online store and club closed sales for parents, pregnant women and children of all ages. More than 8 million women from all over Russia are already using Mamsy app.
CPE — a revolutionary approach in mobile retargeting
Mamsy is one of the first clients that has launched a mobile retargeting campaign with Getloyal using CPE model. CPE is a pricing model in which we choose a post-install event and only pay for users who engage in that specific event (purchase). Working on CPE model means an effective budget expenditure since only the re-engaged users are paid. CPE can significantly increase the quality of the users being acquired – and therefore ROI. Moreover, this model is efficient for scaling the campaigns without binding to the click price.
Mamsy is one of the first clients that has launched a mobile retargeting campaign with Getloyal using CPE model. CPE is a pricing model in which we choose a post-install event and only pay for users who engage in that specific event (purchase). Working on CPE model means an effective budget expenditure since only the re-engaged users are paid. CPE can significantly increase the quality of the users being acquired – and therefore ROI. Moreover, this model is efficient for scaling the campaigns without binding to the click price.
CPE model and mutually beneficial interaction with the client — is the basis for scaling the retargeting campaign.
THE CHALLENGE
User re-engagement at all stages of the funnel;
ROAS > 300%, CPO < 150 рублей
User re-engagement at all stages of the funnel;
ROAS > 300%, CPO < 150 рублей
THE SOLUTION
1. User segmentation and development of creative approaches
We have developed and tested several creative approaches according to each audience segment. The banners differ in sizes and types, for example, we use dynamic banners as well as static ones. The examples of user segments are below:
We have developed and tested several creative approaches according to each audience segment. The banners differ in sizes and types, for example, we use dynamic banners as well as static ones. The examples of user segments are below:
1
Made only one order
We showed complementary goods or current discounts in the application
2
Placed items to the cart but didn't order
Motivate users to complete the order by showing items from the cart or by offering similar extra goods
3
Having viewed an item but didn't make a single order
Motivate users to complete the order by offering popular items, discounts and special first-order offers
4
Opened the app but didn't register
We showed goods with discounts and suggested to register
5
Showed interest in a brand
We demonstrated goods by this particular brand
6
Showed interest in a specific goods category
We demonstrated goods by this particular category
The usage of discounts in creative approaches always performs successfully. After we implemented the tests, we identified the following working approaches with the highest CTR rates and subsequently scaled them. Examples are below:
2. Deeplinks and communication with mobile tracker
Initially proper deeplinks and retargeting configuration is a key to future campaign success.
We always help the client to solve the problems with mobile tracking. For example, in case of Mamsy we noticed that direct deep linking conversions are not recorded by Adjust and, accordingly, are not counted by our tracker. Due to to fast communication, the problem was quickly localized. After that, all conversions were taken into account and reached 80% from click to reattribution.
Initially proper deeplinks and retargeting configuration is a key to future campaign success.
We always help the client to solve the problems with mobile tracking. For example, in case of Mamsy we noticed that direct deep linking conversions are not recorded by Adjust and, accordingly, are not counted by our tracker. Due to to fast communication, the problem was quickly localized. After that, all conversions were taken into account and reached 80% from click to reattribution.
3. Planning, placement and optimization of the retargeting campaign
Campaigns were divided: into a whitelist, which included only relevant applications, and into all other applications. First of all, we rely on CTR and conversions into open rates. These indicators allow you to optimize the campaign for rates and placement platforms at the first stage. We always dedicate special attention to the frequency capping (the frequency a specific app visitor is shown a particular advertisement). Our tests have shown that the most optimal value is 3 impressions per user per day.
The campaign that was launched with whitelist showed higher CTR; the conversion to reattribution and purchase showed also good results. However, there was less traffic than we expected. That's why we decided to scale without white-lists. The main goal was to motivate users for a spontaneous purchase.
Campaigns were divided: into a whitelist, which included only relevant applications, and into all other applications. First of all, we rely on CTR and conversions into open rates. These indicators allow you to optimize the campaign for rates and placement platforms at the first stage. We always dedicate special attention to the frequency capping (the frequency a specific app visitor is shown a particular advertisement). Our tests have shown that the most optimal value is 3 impressions per user per day.
The campaign that was launched with whitelist showed higher CTR; the conversion to reattribution and purchase showed also good results. However, there was less traffic than we expected. That's why we decided to scale without white-lists. The main goal was to motivate users for a spontaneous purchase.
AD INVENTORY
Integration with the client tracking system
Deeplinks
Feed
Getloyal Audience Builder
Getloyal DSP, Yandex
RESULT
CPE model allows to quickly scale the campaign: by the second month we returned 2.5 times more users to the Mamsy application than during the first month; by the third month - x4 times more. And by the current day, we are trying to increase our turnover. CTR banners in our DSP is an average of 3-6%.
Based on the three months campaign results, we performed client's KPI — ROAS> 300%, CPO <150 rub. At the same time, we manage to increase volumes due to the fact that CPO of the previously paying users is lower, so we are able to convert the new ones and stay within the KPI framework. CPE model focuses on user engagement rather than clicks or impressions. So basically, client only pay when users engage with their ad.
CPE model allows to quickly scale the campaign: by the second month we returned 2.5 times more users to the Mamsy application than during the first month; by the third month - x4 times more. And by the current day, we are trying to increase our turnover. CTR banners in our DSP is an average of 3-6%.
Based on the three months campaign results, we performed client's KPI — ROAS> 300%, CPO <150 rub. At the same time, we manage to increase volumes due to the fact that CPO of the previously paying users is lower, so we are able to convert the new ones and stay within the KPI framework. CPE model focuses on user engagement rather than clicks or impressions. So basically, client only pay when users engage with their ad.
Kirill Pyzhov
CMO Mamsy
We use two retargeting systems: Getloyal and Criteo in Mamsy mobile application. At first Getloyal slightly lagged behind its competitor in sales volumes and CPO rates, but due to good up to date performance and results, Getloyal is taking the lead by extracting more and more budget from its competitor..
