How to Set Up and Test UA Campaigns
How do you translate your business goals into a marketing goal, and what processes are needed to ensure you are setup for success? This is the key question you need to ask yourself before getting started.
Let’s examine some of the steps needed to maximize your success through test UA campaigns.
Set ROAS goals
As a performance marketer, you seek ROAS (Return On Ad Spend) and aim to improve LTV (Lifetime Value).
The goal is to work backwards from LTV to discover what levers to pull.
Understanding the user journey within your own environment gives you the ability to predict what users are likely to do when they trigger certain app events. This allows you to understand the value of your various paid media efforts.
Applying these values makes it easier and more predictable to forecast ROAS and optimize your campaign. Ask yourself:
- What kinds of user actions in your app can you map to different stages of the journey towards purchase?
- Are you trying to optimize campaigns for maximum installs or for maximum post-install in-app events?
- Will higher CPI costs deliver higher ROAS?
Deciding on what to measure is therefore a critical component of set up as it is the foundation of success. Aim for granularity, but don’t measure too much at first. Check out this guide to find out which events should be measured in your vertical. Also, this video offers a great app measurement framework to adopt.
You need to setup granular tagging in your attribution link that can pinpoint different elements and segment based on variables such as the creative, the campaign, the test, and the channel.
Set budget strategy
As we discussed in Chapter 3, understating how much we should invest into the growth of our products is a vital component of successful UA.
Test the creative
Many campaigns are test campaigns. Why? Because you need to know what resonates with your audience. To arrive at the answer you can test variations of images, headlines, copy, and any other creative element.
A campaign built on three images with three titles produces a test bed of nine creatives. Of course, you can easily add additional variations to create another layer.
Develop a framework that enables you to test your hypothesis, learn from your winners and losers, and respond accordingly. Stay focused and make sure you emphasize tests that have the potential to yield actionable insights.
- Start with the hypothesis. Determine why you are conducting the test and what you expect to learn.
- Define data-driven goals for success. Determine what your minimum sample size for statistical significance should be and make sure to compare results with a “clean” variant. Only name your test winner when you’ve used data from a trusted attribution provider. Last
ly, define the time frame given to the media source to reach the goals.
- Know in advance what steps to take after gathering results.
- Examine the resources required to setup your tests.
- Understand if your tests are scalable.
- Research testing tactics. Read, investigate, ask peers, join a community and take advantage of the many experienced, knowledgeable, and friendly marketers out there.
- Get key info from media partners that are new to you before experimenting with them.
- Are they self-serve or managed?
- Is their inventory direct or non-direct?
- What level of transparency is offered?
- How much of their inventory is “exclusive”?
- Which ad formats do they offer? Always evaluate based on an actual live preview – not just the listed specifications.
- Define legal terms and include an opt-out clause in the contract where it is clearly stated under which terms a test can end.
In general, you must adopt an always-be-testing mindset in whatever you do. This video offers a proven method to get your app marketing testing going.
Segment for sharper targeting
Your marketing reaches new levels—and the quality of your UA reaches new heights—when you build smart, dynamic, targeted audience segments.
Audience segmentation is dividing people into homogenous subgroups based on criteria such as:
Here are five ways you can improve UA performance with segmented audiences:
- Drive Quality Users —Constantly update your targeting and allocate more budget where you get the highest quality.
- Lookalike audiences—Target prospects that ad platforms identify as similar your best users.
- Uninstall recovery/exclusion—Target users who recently uninstalled your app to drive reinstalls, or exclude them from further targeting.
- Exclusion segmentation—Remove users you are targeting on Network A and need not target on Network B. Or create an audience segment of problematic users and exclude them from your re-engagement campaigns.
- Cross-app promotions—Target users that were once active, but aren’t currently with ads about your newest release.
The process of optimizing your UA campaigns begins with setting ROAS goals and requires careful analysis and testing of your creative executions. Apply the testing tips presented here to continuously refine your campaigns. You’ll also want to create targeted audience segments to further improve UA performance.