Before marketers or app publishers can even think about their mobile engagement plan, they must facilitate a fantastic App Store Optimization (ASO) strategy. ASO is an expansive subject with a multitude of resources available online, but it can be downright overwhelming to keep up with the trends and research. ASO needs to be beyond keyword refinement and the usage of different app screenshots to split A/B test. That’s why we’re happy to welcome mobile marketing expert, Liza Knotko, from SplitMetrics, who’ll share her top 10 best practices in ASO split A/B testing.
Let the countdown begin! Here are 10 split A/B test tips that’ll skyrocket your app store optimization (ASO).
1. Eliminating Office Bias & Banter
Results of A/B experiments are based on mathematical and statistical principles, so they’re deprived of bias and human error. These characteristics make A/B testing a trustworthy and impartial source of ASO insights which are critical for a publisher’s success.
For instance, the Head of Mobile marketing at Wargaming admitted that A/B experiments helped his team to cut out unnecessary office debates. Sure, team members’ disagreement and open discussion are inevitable during any work processes and executions. However, it’s possible to avoid all the unnecessary commotion and conflicts with help of A/B testing. It mitigates ‘whose opinion is right’ dilemma by providing cold, hard facts. Thus, the process of making a decision becomes easier, faster, and more transparent.
2. Perfect Your App’s Store Page
Some publishers still don’t understand that even the development of the greatest app doesn’t guarantee success and profitability. You have to make app store visitors install your app before they discover its usability, UX and all the glamorous features. High conversion is inconceivable without a store product page with carefully crafted elements (screenshots, icon, description, etc.).
How can you perfect your app store page? A/B testing is the answer. There’re two options here: you can use either Facebook ads campaigns to identify the most converting individual layouts or use services which emulate real app stores providing a more relevant context and therefore gives higher quality results.
Indeed, product page optimization can give sensational results. For example, ZiMAD got a 32% conversion lift for their game ‘Bubble Birds 4’ thanks to the redesign of multiple app screenshots. They had a few hypothesis to be tested and after a series of A/B experiments. They found out that it’s better to:
- Shorter captions
- Recharge captions with highlights of the experience instead of cataloging features
- Replace plain screenshots with artistically redefined elements.
Just a couple of alterations in screenshots resulted in 32% conversion boost of installs.
3. Ad Channels Assessment
It makes sense to experiment with various traffic sources before initiating major ad campaigns. You can draw users to your experiments via news, Facebook, cross promo, AdMob, and basically any ad network of your choice. Mind that targeting should be as identical as possible. Analytics of the test should be reviewed carefully, you are to determine which marketing sources can boast better conversion and attract more loyal users.
A/B testing is an easy yet efficient way of identifying the most performant ad channels. Its results will do you a great favor with tailoring further marketing strategy allowing you to leave all unqualified ad spends behind.
4. Pinpointing Your Targeting Users
It’s equally surprising and sad that many publishers don’t even understand who their ideal customers are. When it comes to setting up targeting for an ad campaign, very few developers nail it. A/B testing app product pages on various age groups, sexes, locations, etc. can be really useful in such cases.
An app for photo books creation ran a series of A/B experiments for their screenshots. This activity didn’t only maximize their converting potential, but it perfected their target audience as well. They found out that the app should be mainly targeted at 25+ people who are interested in digital photography, photo sharing, photograph albums, VSCO, wedding and travel photography. At the same time, such audience interests as a canvas print, photo book, and canvas printers online didn’t work that well.
Such target audience insights are essential for effective ads campaigns. Moreover, they give you a hint where your ideal users can be found.
5. Competitive Research
Advanced A/B testing platforms are capable of emulating store search results pages. Such tests give you valuable data on your competitors which can be used to better your conversion strategy. The higher your position in search results, the better for sure. However, actual results are not always that straightforward.
The second position in search doesn’t guarantee the second best conversion rate result and the third doesn’t always beat the fourth one, etc. For example, in the following case, the fourth icon had better conversion than the second and the third options despite its lower position.
The elimination of such analytical elements of surprise is important if you want to see a clear picture of where your app stands. That’s why search and category A/B tests are a must.
6. Tracking User Behavior
Using specialized A/B testing platforms you’ll get profound data on user behavior as they provide scroll, hit maps, time on page, and other core metrics. It’s really important to understand how your app users interact with your app store page and uncover app page bottlenecks.
Such insights clear pain points of your product page. For instance, the majority of tests show that users hardly ever scroll screenshots, so it’s vital to make the first one as attractive and enticing as possible. Studying your user behavior can take your conversion rate to the brand new level.
Unfortunately, a lot of publishers understand localization as a literal translation of their product page in a country app store. It’s vital to understand that localization stands for adapting every app page element to another culture or country.
FiftyThree determined the formula of an ideal screenshot set for Chinese-speaking audience via thorough A/B testing of their app “Paper”. It turned out the more screenshots that differ from each other, the more engaging the page is for the Chinese user base.
Localized screenshots in Chinese with multi-colored backgrounds reached 33% better conversion than the original ones in English. It proves once again that marketer is to translate product pages beyond the text, but to also take into consideration cultural elements.
It’s always hard to decide which app features to emphasize in the promotion of an app. There’s one thing clear: these features are to resonate with your audience. Split A/B tests assist in identifying which messages instigate users’ desire to install your app.
For example, Rovio A/B tested heavily before the launch of ‘Angry Birds 2”. One of the experiments was aimed at identifying the most appealing character. As a result, a gray bird was put on the first screenshot. Thanks to all pre-launch A/B testing activity Rovio got 2.5M extra installs in just a week.
9. Pricing Policy
It usually hard to decide which pricing model will work for your app. Free apps with in-app purchases turn out to be more lucrative at times. However, once you decide you want your app to be paid, it’s extremely important to find out the price which won’t scare potential customers. A/B experiments of various prices can solve this problem. Sometimes offering a free trial is also a good start before you pivot on pricing.
10. Product Image & Value Proposition
Everybody knows that there’s no use investing time and money in doomed enterprises. Pre-launch product page A/B testing (yes, it’s possible too) are really useful on early app development stages. Thus, publishers learn potential customers reaction to various concepts, characters, features, etc. If the app’s basic image or value proposition doesn’t resonate with the desired audience, it’s a sure sign to change your focus.
Overall, A/B testing is a powerful method which can fulfill multiple functions within ASO activity providing you’re ready to observe basic rules. Split tests require precision and dedication to give trustworthy results. Nevertheless, thousands of developers learned that incredible exact results, time-saving, and conversion boost are worth the effort.
Liza Knotko is Сontent Marketing Manager at SplitMetrics, an app A/B testing tool trusted by Rovio, MSQRD, Prisma and Zeptolab. SplitMetrics helps mobile app developers hack app store optimization by running A/B experiments and analyzing results.