Have you ever thought which of the search queries are the most popular apart from top brands? How many unique high-volume searches are there at all?

For the app discovery, the app store search is crucial. According to a survey, 47% of app users found an app via the search engine on the App Store. And 53% of Android users have found the app the same way on Google Play. According to the same survey, 65% of downloads arise from searches on the AppStore.

But it’s very interesting what exactly users type into the search box on the stores for finding their apps. According to another survey, an average smartphone user has over 80 apps on his phone. And the answer isn’t simple. The app store searches are just a black box for everyone, it’s impossible to find it out, unlike in web search. So, just the two leading platforms are aware of it. However, the situation might change in the nearest future.

This information also can be disclosed to different third-party technology providers, for example, attribution or analytics services. Various search estimates of the ASO tools rely on Apple Search Ads popularity or Apple Search Priority very often. But they provide only a score, not actual volumes of search. In major cases, they are used along with other signals, as a rule, from web-based searches.

John Koetsier from the TUNE attribution provider studied the 500 top store keywords in March 2017. What did he find out? His main discovery was that 9 out of 10 top keywords are branded, which means that users usually search the apps for brands they know or have heard of already. The TUNE research showed that 86% of the top 100 keywords were exactly brands.

The non-branded keywords in the top were related to simple utility apps or games, for example:

  • Vpn
  • Games
  • Music
  • Weather
  • Calculator
  • Free games
  • Photo editor

So, does a significant long-tail of app search exist, actually? If yes, what does it look like?

Users’ search requests apart from brand names

According to Relly Brandman, a Google product manager, there are two types of searches, namely:

Users who know exactly what they are searching for

Users who do not know what they want and search by such keywords, as “gobbling”, “puzzle”, etc.

Google Play’s head of search and discovery Ankit Jain offered a more detailed insight into Google Play users’ search at Google I/O in far 2013:

He pointed out two types of search queries: categorical and navigational. To categorical queries belong broad search terms, such as train schedules, free games, etc. And navigational ones include exact search terms, for example, viber, zombie tsunami, etc. –i.e. branded searches.

According to Jain, 12% of daily active users search for the apps on a daily basis, 50%-on a weekly basis, and, according to Google search, there are 6 million unique phrases searched on a monthly basis.

But there still remain millions of such unique search phrases monthly. But how to find out which of them are branded and which are not? And what amount of them makes an essential volume?

To find that out, it’s necessary to start from Apple’s Trending Searches. Knopp shared the following data on the global number of unique trending searches on iOS in different countries for 2016 comparing to 2017.
Trending Searches
According to these unique data, users are searching for twice as many high volume queries in 2017, than in 2015. There were 2.455 unique search phrases trending in the USA in 2017. And up from 2.247 in the year 2016.

Autocomplete and what it tells about search behavior

Knopp also provided a list of autocomplete App Store searches for a more qualitative analysis. But first, what is autocomplete? It is a common feature for search engines suggesting a shortcut for users showing the most likely result when typing your query.

Once a user is typing just a single letter, let’s say, “c”, the AppStore shows a list of 10 suggested searches starting with the letter “c”. In case users keep typing, “co…” there appears a new list of suggestions starting with “co”, etc. Apple’s Priority Index score identifies the order of showing auto-suggestions.

Knopp provided the list of 9.000 queries suggested by the App Store’s autocomplete option on February, 19th. The autosuggestions are provided for a-z and 2-character combinations for these letters from «AA”, “AB”, etc. to “ZZ” together with the Priority score assigned by Apple.

And there was no surprise; brands also dominate those auto-suggested lists. If making a quick manual analysis, we will see that out of 260 suggested words (10 for each alphabet letter), just 13 of them (i.e. 5%) are non-branded terms, such as calculator, bible, email, games, mail, news, maps, Olympics 2018, photo editor, etc.

After removing duplicates, there are 8.300 items in the list for all of the 2-letter combinations. If you search for commonly repeated phrases apart from brands, the most frequent are:

  • App&apps -338
  • Mobile-195
  • Games & game-158
  • Inc-89
  • News-84
  • Free-84
  • LLC-73
  • Radio-62
  • Ltd-53
  • Tv-47
  • Weather-46
  • Remote-45

By the way, such brand-related terms, as ltd or inc. appear in the most common words, too. Besides, video and audio apps and utilities dominate, too.

The word “games” a part of a users’ query appears 127 times, in such genres, as zombie games, arcade games, etc. The data contain just about 25 navigational queries with the word “apps”, for example, nursing apps, dating apps, music apps, etc. The word “app” in singular appeared over 300 times referring mostly to brands’ names.

The Answer for App Developers

The brands still handle the short- and mid-tail of app search. The longer tails of keywords remain in secret which every app developer solves individually. But Google has assisted in that issue, too.

Google released its new organic search data in the “User Acquisition” section in the Google Play Console in January. It was rolled out to a beta-testers group first.

This data makes it possible to split organic traffic from Play Store into Search and Browse. It means that if a user finds the app listing via a search query vs. browsing the store. It also gives data around installs generated per a keyword for the first time. Google showed the top 1.000 search terms the app is ranking for, together with their conversion rate, including store listing visitors, installers, buyers, or retained installers.

The app marketers might find out the answer to the question “How are users searching for the apps?” in the coming months. It will be confirmed by real and relevant data.