ASO Keyword Research - How to Choose Keywords that Get the Most Downloads
App Store Optimization is one of the fastest and easiest ways to acquire new users for your app by increasing its visibility in the App Store and on Google Play. A major component of ASO is keyword optimization, which when done correctly will help push your app to the top of the search results in those keywords. But to reap the rewards, you'll need to identify the right keywords first.
What is Keyword Optimization?
Apple's and Google's search algorithms are designed to connect users with relevant apps. To do that, the algorithms use app names, subtitles, and descriptions to "understand" what apps do, and then match them to user searches. While the algorithms continue to evolve one thing remains constant: they rely on the information you as the developer provide to do this matching.
Knowing this, you can help the algorithm by pointing out where you'd like Apple and Google to show your app. That's keyword optimization. But here's the catch, Apple and Google limit the number of characters you can use, so it's crucial that you pick keywords that get the most traffic. Keep reading to learn how to separate good keywords from those that are just a waste of time.
In this guide:
- Data Empowers Keyword Optimization
- How to Choose the Right Keywords
- Start by Collecting Keywords
- Focus on Keywords with High Traffic
- Find Keywords You Compete On
- Use the Top 5-10 Keywords
- Monitor and Iterate
- A Hands-On Example: Evaluating Keywords for a Weather App
Data Empowers Keyword Optimization
There are two ways you can tell how well you’d rank for a keyword and how much it’ll add to your bottom line: trial and error or using data and insights. Trial and error isn’t a terrible way to learn, but it’s very slow and means failing a lot more than succeeding. Using data and insights to identify the right keywords you can skip the failure and target what works right away with more immediate results.
In this guide we’ll go over how to save time while optimizing by using data and insights to filter your list of potential keywords. We'll also walk you through a step-by-step keyword research for a real app.
How to Choose the Right Keywords
Finding the right keywords for your ASO is a simple process that starts with identifying potential keywords and continues to ask questions about their performance:
Collect as many potential keywords as you can. Start by listing all keywords you can think of that you think your potential users are searching for. This list can include your app’s features and benefits, keywords competitors have in their names and subtitles, and keywords related to those.
Filter out the non-popular ones. Not all keywords get traffic. Look up the popularity score for each keyword and remove the low ones.
Find the ones you can compete for. The search algorithm determines which app to put at the top primarily by its downloads and ratings. Look those up for the top results in the keywords in your list and focus on ones where your performance meets or beats those.
Choose the top 5 and use them. After filtering out keywords you can't compete for, take the top 5-10 most popular keywords and integrate them into your app's metadata.
Monitor your results and double down on what works. Give the changes 3-4 weeks to live on the store, monitoring daily. See which ones got the best ranks the fastest and work to find more related keywords.
Accelerate Your Downloads
Discover new keywords, monitor ranks, and snoop on competitors.
Start by Collecting Keywords
To find the right keywords you’ll need to go through and evaluate more keywords than what you’ll end up using. At any given time, you should optimize for 5-10 keywords. How many you’ll need to evaluate depends on how much time you and put into your optimization efforts and the size of your niche. We recommend going through a minimum of 30 keywords and filtering them down.
Focus on Keywords with High Traffic
The first step at evaluating which keywords to optimize for is to identify the ones that are being searched and the ones that aren't. You may think a keyword is popular using intuition, but in our experience it's not always that straightforward and being the first result for a keyword no one is searching for is a total waste of time. You don't really want that.
Instead, you'd want to figure out which of the keywords you have on your list are the most popular and focus on those. You'll do that by looking up the keyword's popularity score.
A popularity of 100 means it's one of the most searched terms in the store. "instagram", which has a score of 100, is good examples for the most popular searches.
A score of 5 on the other hand, which is the lowest score possible, is given to keywords that get little to no searches at all. Many keywords you'll look at will have a popularity score of 5.
What's a Good Popularity Score?
You always want to aim for the keywords that are relevant and have the highest popularity score. But not every segment of the market is like Instagram's and it's very possible the list of keywords you assembled won't have high popularity keywords.
We studied ~11,000 keywords in the US App Store and found that most keywords have a popularity score between 10 - 40.
Tip: As long as your keywords aren't in the single digits they're good to use. If all of your keywords are however, you should go back to the beginning and come up with more keyword ideas.
Find Keywords You Compete On
Knowing the popularity of keywords, you can quickly determine which will help you get more downloads and which won't, but that's just the first half of the evaluation process. The second, and possibly more important part is finding keywords that you can actually get into the top results in.
Most users don't scroll too much after searching, so the higher you're ranked the higher the chance you'll get the user's attention. Those chances drop significantly if you're not ranked in the top 5 and become very small if you're not in the top 10.
Apple's and Google's algorithms use downloads to determine the order in which to display results, so the second part of the evaluation process will be to find keywords that have similar performance to your app. We'll do that by looking at the estimated downloads, category rank, ratings, and age of the top 10 results for each of our keywords.
Tip: The Competitiveness Score is a summary of all of those factors together. It's useful when you need to compare a large amount of keywords fast, but for this part of the process we'll look at the actual values.
Benchmarking a Keyword’s Performance
To determine the kind of performance you need to rank well in a keyword you should create a benchmark of its top results. To do that, look at the top 10 results, noting the app with lowest number of downloads and the average number of downloads across all 10. Do the same for ranks and ratings and that's your benchmark.
You want to optimize for the keywords where your app's downloads, ranks, and ratings meet (or beat) the minimum, and preferably get close to the average. You can try optimizing for keywords where your performance is lower, but you probably won't rank well so they won't net you any new downloads.
Use the Top 5-10 Keywords
Once you have your filtered list and know the numbers you need to compete, choose the 5-10 you can compete for that have the highest popularity and use them.
Using keywords means inserting them into your app’s metadata where it matters most for the search algorithm. For iOS and Mac apps that’s the app’s name, subtitle, and keyword list. For Android apps on Google Play that’s the app’s name, short description, and long description.
Monitor and Iterate
Once the new keywords are live, give the store 3-4 weeks to detect your changes and update its index. Make sure to watch your keyword ranks closely during that time to see if your app gets ranked in the new keywords, and if that rank is slowly going up. In addition to ranks, keep an eye on impressions to make sure they're not going down.
If you've done your evaluation correctly your keyword ranks, impressions, and downloads should go up.
To continue growing your downloads, repeat this process with the keywords you had on your list that you didn't end up using or with new keywords that you've evaluated and compare the results. You can then choose to keep some keywords but not all or continue on to a new keyword set altogether. Constantly iterating isn't something you have to do but will help you get closer to the best set of keywords. Just remember to give each iteration enough time.
A Hands-On Example: Evaluating Keywords for a Weather App
To help you put this guide into action, I'm going to share how I analyzed keywords for Weather Line, a beautiful weather app for iPhone and iPad that recently got a makeover.
I started my analysis by reading through the app's new description to uncover its features. I also took the time to read descriptions of other weather apps that are ranking well to get a feel for how they describe themselves. That led to me to the following list of exactly 41 keywords:
- fall forecast
- forecast weather
- hurricane tracker
- live radar
- live weather forecast
- live weather map
- local weather
- local weather alert
- local weather alerts
- local weather forecast
- local weather radar
- noaa maps
- noaa radar
- noaa tracker
- noaa weather
- noaa weather app
- rain alert
- rain forecast
- rain tracker
- storm alert
- storm alerts
- storm forecast
- storm map
- storm tracker
- track weather
- weather alert
- weather alerts
- weather app
- weather forecast
- weather map
- weather maps
- weather nyc
- weather radar
- weather this week
- weather today
- weather tomorrow
- weather tracker
- weather tracking
- weather warning
- winter forecast
Notice how I didn't include any competitor names? That's on purpose! When a user searches for an app by its name they're looking for that app and will download it as soon as they see it. Ranking below that app, which is what's most likely to happen, would do nothing for your downloads.
I start my evaluation by adding Weather Line to my Appfigures account. Then, adding the above keyword list to the Keyword Performance report so I can quickly see and sort the keywords by their popularity score.
Having the keywords on the screen it's obvious that while competitive, some of my initial attempts at guessing what users would be typing into the search box weren't all that great. This is normal. Let's drop all the keywords with single-digit popularity score.
Next up is benchmarking the performance needed to rank for each keyword. Weather is a pretty challenging category because there are quite a few options and some well-known brand names. So we'll need to rank high to have the most potential for a download. Because of that I'm going to base my benchmark on the top 3 results and not 10 like I normally would.
Let's look at "weather radar", which has a popularity of 59.
To squeeze into the top 3 for "weather radar" we’ll need to use the keyword in the app's name and have at least 90K downloads per month in the US. We'd also need a high number of ratings because the competition starts at around 500K.
This may be a keyword I end up suggesting, but first I continued doing this for all the popular keywords on my list. After doing that I decided to focus on the following keywords:
- Live weather forecast
- Local weather
- Local weather alerts
- Local weather forecast
- Weather alerts
- Weather forecast
- Weather map
- Weather maps
- Weather radar
- Weather tracker
Let's use them!
After a few creative iterations I came up with the following suggestion for the app's optimized name:
Weather Line - Live Forecast
and the following suggestion for the subtitle:
Local Alerts, Maps, and Radar
This name and subtitle combination targets all of the keywords I prioritized from our original while still being very readable to humans.
The last place to use keywords is the keyword list in App Store Connect. Unlike the name and subtitle, the keyword list has a higher character limit and is invisible to users, so it doesn't have to be readable. I suggested the following list:
The keyword list in App Store Connect is a good place to add keywords that combine well with the name and subtitle but may not be as important. If you have the space. This gives you more chances to rank in even more keywords such as "storm map", "rain tracker", and "hurricane alerts".
The algorithm automatically combines the name and subtitle into a single list of keywords but considers the keyword list as a separate source for keywords. Repeating them _between_ the two reinforces the keyword's importance.
Some marketers may not agree, but this is based on my experience and the data I've seen.
And that's how I research and analyze keywords. It's a pretty simple process once you go through it a few times.
App Store Optimization, and more specifically keyword optimization, is a powerful way to get more organic downloads. But only when done correctly. Now that you've read the guide and the step-by-step walk-through you know what it means to do it correctly, so go ahead and give it a try with your app and make sure to track your keyword ranks to see what works and what doesn’t.
Appfigures helps you track and optimize your apps. Hourly rank tracking, powerful analytics, app intelligence, alerts, and more. Try it out →