You probably have heard about App Store Optimization, some already use its methods to increase the inflow of units from the search App Store and Google Play, as well as improve conversion rates on the page of the application. In this article I would like to talk about one part of App Store Optimization, namely, the proper semantic kernel, which will allow you to find many targeted search queries for the app which will increase your traffic from the search App Store and Google Play in several times. Despite the relatively simple methods of finding the target of search queries, many mobile developers didnt do it still.
But in vain. Semantic kernel is a set of words and phrases that describe the functionality of the app, and the users who may decide the application. More simple language is a set of words and phrases that your target users can search for your mobile app.
Why it is needed. Semantic kernel allows you to build the most accurate picture of the needs of the target audience. Knowing the needs, you will be able to understand how you can look for your users in the app store, to understand the exact search queries.
Another interesting point. About 70% of search queries — long queries that include more than one word. In other words, your users can find, for example, not only for “work”, but for “work”, “paying job” and so on. How to make the most complete semantic kernel.
For building a semantic core, you need to understand how you can look for your users. This will require common sense and a few tools that will help your common sense to transform into high-quality semantic core. Common sense. If your app has not yet appeared on the market, you just going to let him out, then you should start with common sense, namely to sit down with your team and think about search queries that meet the functionality of your application.
For example, if your app is looking for tickets, your requests will be. “tickets”, “airline tickets search”, “plane tickets”, “cheap tickets” and so on. Analysis reviews.
If your app is already on the market, in addition to common sense, you can strengthen your list of requests, based on the feedback to your app, if of course they are. For example, from the review of the Duolingo app for iOS requests can be received “training English”, “training English”, “training English”. User surveys. If your application are set mechanisms of direct communication with your users, you can conduct a poll among them, asking them questions.
“What requests would you looking for our app in the search?” or “What problems can you solve by using our app?”. I note that the reviews and user surveys are the best sources of targeted search queries. Competitor analysis. Yes, where do without competitor analysis.
You need to understand which search queries your competitors have done the optimization. For this. Selecting Keyword Spy for app Badoo, we get a list of keywords.
The more requests that you collect in the first stage, the higher quality will be the semantic core in the end. At the end of the first stage you are given a list of dozens of keywords and phrases, but it cannot be called a quality of semantic core. With high probability this does not take into account the core of any search queries that your target users can use.
But where to get more words and phrases?. Synonyms and examples of usage of words. The Internet is just a huge database of synonyms and usage examples of words that are constantly updated by users.
For English you can use, for example, WordNet. The database is updated under the guidance of Princeton Univercity. Has a simple API to use.
For Russian language you can use Yet Another RussNet, which is also updated by users and has a simple API. A similar linguistic base have in other languages. It is also useful to look at examples of usage of words from dictionaries.
Yandex.Wordstat and Google Trends. These services are useful because they provide information about popular and similar search queries in the web. And you can display statistics only for mobile devices.
It is recommended that each meaningful keyword to run through such systems. How to use received data. After all these actions, you will have a list of several dozen, maybe even hundreds of search words and phrases.
It is likely that all of these requests you will not be able to consider when optimizing your app. But all search keywords and phrases to take into account not worth it for two reasons. How to understand whether there is traffic for a search word or not, where to get statistics for a word or phrase for the App Store and Google Play.
You can use the services of assistants, for example, SensorTower.com. They provide some data for keywords, like traffic, the current position of your app in search results, the number of applications that are already on this request. Traffic data is not accurate, is enough for more or less the right decisions when optimizing the obtained semantic core.
There are two algorithms how to clean your semantic core from a sub-optimal search queries. More detailed and clear explanation will soon appear as a separate item. Not worked out the semantic core.
The Preggie app, a social network for pregnant moms. Please note that the search queries are ordered in descending order in terms of “Traffic”. This means that the query “pregnancy test” compare the traffic with the query “pregnancy”. Nevertheless, only 2 of the app include it in your search engine optimization, column “iPhone Apps”.
Similar situation with the query “pregnancy calendar”, “signs of pregnancy”, “gestational age”. The above-mentioned applications, we have made search engine optimization that in the next re-issue will be taken into account. Worked out the semantic core.
The application Aviasales is well suited to work on search engine optimization. Elaboration of the semantic core can help you attract targeted users to your app absolutely free. Users themselves will find you. Moreover, such users will be the “qualitative” because these users have specific needs that your app can meet.