Our client is one of the largest retail and e-commerce apps in Latin America, and one of the top 10 platforms by the number of downloads and monthly active users. The app features more than 100 million products in different categories. The app’s brand is well-known in the region.

The client has been on the market since 2011, so he has deep expertise in marketing. He sets clear and non-trivial tasks and has extensive experience in working with different sources. But perhaps the most important thing is that the client is well-represented in the market, which has further expanded following the initiation of our collaboration.

Starting a journey

Our team had to acquire new clients for a popular app. We were not using sources that the in-house team was already working with. We also had to avoid sources with a high rate of fraud. Specifically, these were Google Ads, Meta, TikTok, Reddit, and sources with a bad reputation from a client’s blacklist.

Karina Vasyukevich, Head of User Acquisition at Zorka.Agency, describes how to find new users for an e-commerce application using OEM sources:

“Firstly, we had a difficult task. We had to find new users for a very well-known application that had been on the market for 12 years. Secondly, we worked in cooperation with the client’s in-house team and contractors, and could not use sources that were previously used. Thirdly, the client was against traffic with even the slightest bit of fraud. To put it in simple terms, we had a significant problem to solve.

Our team had to acquire new users to a popular e-commerce app, excluding Google Ads, Meta, TikTok, Reddit, etc., which had already been used, all the while foraging sources.”

The terms of cooperation and KPI of the campaign were as follows:

• Pay the contractor a CPA rate for the first purchase within the first 7 days of install;
• iOS/Android: $ 9;
• GEO: LATAM;
• There was no payment for all of the fraud traffic;
• If we find fraud in the source, the source was excluded;
• Tracker and revision by Adjust.

At the very beginning, we decided to exclude all sources that could bring in fraud traffic. Therefore, we focused on OEM sources. Later on, we test-connected in-app sources, but they showed worse results.

Why OEM?

Original Equipment Manufacturer (OEM) — these are formats of ad placements that are provided by smartphone manufacturers themselves. You can find more information about it here.

OEM source traffic does not have fraud. Many mistakenly attribute the quality of in-app sources to OEM traffic, including the presence of fraud. That is incorrect because OEM traffic is fraud-free due to its design and user experience.

Of the OEM group, we decided to work with Appnext. Appnext is a recommendation platform that sells 3 billion advertising messages daily. On the platform, users contact advertising messages up to 20 times a day. Top mobile device manufacturers such as Samsung, Xiaomi, and Oppo have already successfully integrated Appnext into their mobile devices.

Creatives

The special aspect of OEM sources is that in addition to advertising creatives, application icons themselves participate in the promotion, which makes the placement native on the smartphone.

In order to test within the platform itself, we decided to use an icon in the placements and several variants of creatives with different USPs. On the one hand, it helped us understand which type of creatives worked better, and on the other hand, we were able to compare the performance of scenarios with different USPs.

Following the strict rules of the brand book, as the main advantages to create banners, we used:

• free shipping, wide product line, cashback;
• a promo code for the first purchase.

During the campaigns, we noted that placements with creative icons were more effective.

Pic 1. Formats of OEM Advertising

The average CTR percentage of placements was 7.68%.

Results

We are grateful to the client for their trust, and to the partners at Appnext for their resulting work. During the work on the project, we tested 3 traffic sources, one of which was in-app.

Pic 2. Campaign results (data includes repeat purchases)

Campaign results were like this. OEM sources on Android brought 44 344 downloads with CPI 3.35 and 19 893 purchases with CR first order 14.66%, CR purchase 44.86%, CPA purchase 7.472$.

In-app sources on Android brought 2 602 downloads with CPI 1.23 and 573 purchases with CR first order 5.42%, CR purchase 22.02%, CPA purchase 5.624$. In-app sources on IOS brought 20 561 downloads with CPI 0.75 and 680 purchases with CR first order 3.31%, CR purchase 16.95%, CPA purchase 4.457$.

As you can see, in-app sources results are often times worse than the OEM sources. This is due to the fact that OEM advertising is integrated into the smartphone. Since the smartphone has become the most prominent screen in most people’s daily life, the credibility of this type of advertising is higher.

If you are interested in leveraging the power of OEM in your strategies, don’t hesitate to get in touch.