6 Steps to Improve App Search Rankings
Working as developers, we are focused on the low level technical details of a product, being that a website or an app. This heads down approach often makes us not to pay attention to many high level details that are crucial to bring our product to success. The typical path for a developer, at least at the beginning, is to build an app, put it on the store and then see it fail miserably. If that has happened to your app, then this post is for you!
There are many different ways for an app to get attention from the public: ads, blogs, search engines, word of mouth, social networks, top rated apps and searches on the play or app store, etc.
More often the most used channel is searches on the Google Play Store and Apple App Store. That’s what we will focus on here.
Whenever someone looks for something on the app/play store they will make a search and get back a list of sorted results. The way results are sorted is very important, as users will only look at small subset of those results. In general, we can say that statistically the chance of catching a user’s attention decreases exponentially with an app’s position in a search. Hence our goal to get attention and accordingly downloads on the store requires us to try to rank in the highest possible position in popular searches.
Nevertheless, people don’t know how exactly the app/play store ranks apps in searches and can only make guesses about it and validate them through experiments and data analysis.
So in this post I’ve collected some of those guesses to come up with a list of things that most likely matter in determining an app’s rank in search.
Downloads make downloads!
It seems that downloads have a very clear impact on the apps ranking, especially when they are recent. This study (http://www.mobilesort.com/blog/ios-sale-numbers-by-app-store-rank.html) shows the correlation between downloads and rankings in searches:
The graph above shows the results of an experiment conducted by keeping track of the average ranking in the same searches for a particular app. It has to be read backwards (the lower the rank the better), and we can see how a higher number of daily downloads tends to mean a better ranking, in particular spiking up after a certain threshold.
2) Ratings and Reviews
Ratings and reviews are very important both in their quantity and content. This graph shows the impact of their quantity:
We can see that the top ranked applications always have a higher number of ratings for every rating value. It could be objected that they have a higher number of ratings because they rank high and hence get more downloaded by people and not the other way round. It seems however, given the same number of downloads, that if someone pushes to get more ratings from her users the rank does get a benefit from it.
This graph instead shows an average of different applications’ rankings around the time people believe apple started using the ratings as a measure for ranking:
The takeaway is to always ask your users to rate your app. There are several libraries that allow to integrate this functionality very easily, showing pop ups to users that link to submitting a review.
3) App Updates
Whenever an app gets updated and in particular when an app is launched, the app store is believed to give it a boost to promote apps people are actively working on. Several people have confirmed this belief with experiments.
So if your app doesn’t start on the right foot and hasn’t many active users to be sacrificed it might be worth it to delete it and submit it a second time in an attempt to exploit that boost.
Whenever we submit an app we are asked to fill in a keywords field. Those keywords are the searches in which we want to show up. Since they can be too many we have to pick them wisely.
if we add keywords for searches that are really popular, chances are that the competition will push our app down in the results rank and it won’t be reached. If instead we add keywords for searches that aren’t popular at all, our app simply won’t get searched. There are numerous tools that can help estimating the volume of searches for a certain keyword and its number of results and that can accordingly help spotting out the best keywords with low competition.
5) App Title
The app title both in the app and play store can be 30 characters long. Although for the title itself we want to choose something catchy, we can use the rest of the space to add a very short description of what the app does. The reason to that is that keywords in the title are proven to be the most effective way to rank high.
For example, one of Grio’s games, which is a fun memory game, is called “Whobert’s Path”. More effectively than leaving the title as such, we could change it to “Whobert’s Path – enhance your memory!” or “Whobert’s Path – memory game” or “Whobert’s Path – train your memory!”.
The app and play stores allow to publish an app in different countries with a different title, keywords and description. As it might sound obvious and also stats confirm, in countries with different languages people for the vast majority don’t make searches in english but in the local language. This implies that solely by translating our title “Whobert’s Path – train your memory!” to “Whobert’s Path – allena la memoria!” and publishing the translation into the Italian store we could potentially expand to a new market with little to no effort.
It has to be mentioned that if we are not translating the content of our app (which might be more of a time consuming activity) what said only applies to apps that are not much text based like visual games otherwise the app won’t get any interest from the public being gibberish and unusual.
Although finding a translator online could be fairly simple and standard, it doesn’t cover it all, as we also need to take into account that the translated keywords may not be trending or popular in the selected country.
A few tricks that serve this purpose are explained in this post: http://makeappmag.com/iphone-app-localization-keywords/.
The most interesting approach uses the app/play store search autocompletion. Let’s assume that we are looking for what might be trending keywords in Italy related to our memory game “Whobert’s Path”. We could translate the word memory in Italian and get “memoria”, then type it in the search in the store: http://drt.io/alwij and see what gets suggested in the typeahead.
Those suggested keywords are likely to be suggested as the result of being popular searches linked to our keyword on the local store and are very good candidates for us to be used. We then could list those suggestions and check the estimated volume of searches and number of competitors and in the case of good numbers we accomplished our task.
So there are many ways to improve your app’s search rankings. You should use all of them!