"Regardless of the ad format, type or mechanics you wanna try if the content is really interesting for the users and, of course, if the USP is something the users are willing to pay attention to or really is solving something on their end, then, definitely, a possibility to succeed in UA actions will be higher".
What we have talked about:
- Different formats of creatives (banner, inertial, video, playable, native ad and fake ads)
- How to test the creatives in a more efficient way. What geo/ target audience/platform to choose. Android examples, iOS14+ is a different story
- TikTok as one of the most performing channel: spark ads, winning tactics and challenges
- The most significant differences between SKAN 3 and SKAN 4: 2 new additional postbacks, new “Lock Window” concept, crowd anonymity, web-to-app attribution. A current state of transition process.
- Trends to follow in 2023
- Interstitial ads are too invasive, though they might be good for brand-awareness.
- Playable ads: it is a good format to provide a sneak peek into the app, however, it might be not applicable to all the cases (hyper-casual - doable, it shows good performance rate/ for hardcore or RPG - harder due to the limitation the ad usually has).
- Videos: in Meta the best ad format as it allows giving a more extensive explanation in 15 seconds (ideal length). It is recommended to blend animated videos with UGC or influencer content.
- Banners: the engagement rate is higher with native ads than a regular banner
- Content is a key ingredient for creatives rather than format. If we understand what USP is, what pain points we are solving with our target audience, then the likelihood of success will be higher.
- Testing is a must since Google, Meta and other networks are moving towards a more automated environment (AAA campaign in Meta, ACi campaign in Google). Targeting is becoming less relevant.
- It’s better not to launch testing in Google. Pick the network for testing where you already have your campaigns running or allocate more budget. You can run the tests in Meta and then apply the learning into SnapChat, for example.
Admiral Media testing strategy.
- The cheapest one (upload the creative to test in a new campaign, in one single ad set, let the algorithms decide on a winner) Cons: you won’t have enough data to compare the competing creatives as algorithms usually push 1-2 creatives, others won’t have enough spend rate
- More insightful. 1 campaign, but separate ad set for each creative. You will have the same spend rate and can decide on a winning creative based on the data
- Hybrid model. If you have many creatives to test, you can combine them into groups, i.e. 1 ad set per topic. Algorithm will push some creatives in each ad group, but no matter that you will have enough data to choose a winning creative in each topic
- TikTok: it’s among the 2nd or the 3rd best-performing channels. If you crack Meta and Google, TikTok is definitely the channel to try
- Spark ads - the basic difference is that you leverage organic posts from brands’ profiles or creatives’ profiles. It has a better performance compared to regular ads, therefore, that leads to skipping regular ads and running only spark ads continuously. Challenge: it’s harder now to find micro influencers at a reasonable price that could deliver a quality content
- SKAN 4.0. It’s confusing as we still don’t know how the ad networks will adopt the new features, however, the sooner you dig into SKAN 4.0, the more advantages you will have once its data starts rolling in
To keep an eye on:
- Ad networks & MMPs updates on SKAN 4.0 implementation
- The news around AI (bearing in mind the arguments around banning ChatGPT in Germany, Italy)
- NFT as it is likely to show a positive upward trend this year