Today we have a very special guest post from Marcus Carlsson. Marcus wrote a post in the Facebook group “App Entrepreneurs and Marketers Group” on which he told how he went from 1,000 downloads to 2,500 downloads a day with a single keywords change using TheTool, and agreed to share his story with our users. Moreover, he doubled revenue to around $200 / day. Let’s see how he did it in his own words 🙂
How to Double App Downloads with ASO & TheTool
Whenever I used to change my keywords I didn’t have any special method. I merely looked at the keywords individually which would rank good and used them. I started to believe that I should use mid- and long-tail keywords and came up with a great way of adding those.
The app of this case study is a two-year-old iOS camera-app in the photo-section. It has been featured by Apple as Today’s app all over the world, however, during the time of this keyword change and after that, it has not been featured again. So that will not influence the data.
How I came up with the keywords
Before the change, I had around 200 single keywords that I usually use. But with this change, I wanted to see how I could improve the downloads by looking more at mid-tail keywords instead of just single keywords.
So, I started by going through all the 200 keywords and looked on which one I could combine as logical mid-tail keywords containing two keywords. Example:
Let’s say that I had these three keywords before:
I combined them into these mid-tail keywords.
- Dark image
- Dark filter
- Filter image
I went through all of the keywords I had and ended up with a total of 1500 mid-tail keywords that contained two keywords each.
After that, I used TheTool to get all data for each mid-tail keyword.
To make it easier to make the calculations (I will explain them later), I exported the data as a CSV-file and opened that in Excel.
My main idea was to see what type of mid-tail keywords that I already ranked for position 1-10 for. My assumption was that the words with similar data hopefully would rank for 1-10 even though they didn’t do it right now.
So, I looked at the ones that the app ranked for 1-10 already, and noticed what the keywords averaged on traffic, difficulty, and number of apps. For instance, all the ones that ranked 1-10 would have these averaged values:
- Traffic: 23
- Difficulty 40
- Number of Apps: 500
To be sure that I found the mid-tail keywords that fell into one or more averaged value, I used a point system.
If a mid-tail keyword had ranked 1-10 already, that keyword would get 2 points (I used 2 points here since ranking 1-10 is important).
If a mid-tail keyword’s traffic would be close to the average I calculated above, it would get 1 point.
I did the same for difficulty and number of apps.
For instance, if a new mid-tail keyword had these numbers:
- Traffic: 10
- Difficulty: 45
- Number of Apps: 200
- Ranking: 40
That keyword would get 1 point only for difficulty since it’s almost the same as the average value of difficulty.
- Traffic: 20
- Difficulty: 10
- Number of Apps: 450
- Current ranking: 8
This keyword would get 1 point for traffic, 1 point for the number of apps and lastly 2 points for already scoring 1-10. That is a total of 4 points.
So, now I calculated the points for all of my 1500 long-tail keywords, and the maximum points a mid-tail keyword could get were 5 points. I looked at all the mid-tail keywords that got 4-5 points.
That is, they had to have lots of data that were similar to the ones that I already rank for 1-10. I got around 250 mid-tail keywords.
Now, I continued and looked at the individual words in those mid-tail keywords that I just got. I made up a new spreadsheet where the number of times each individual word was included in a mid-tail keyword. For instance, if I would have these words (out of the 250 very good mid-tail keywords):
- b&w photo
- b&w image
- enlarge image
That would make this list:
I then figured that the words that combined the most mid-tail keywords are the most important keywords, and I put that in the title of my app. Naturally, the title had to have some sort of grammar correctness. I then kept on putting the words in the title, subtitle, and keywords. So, in my title, I have the words that I can combine the most and, hopefully, I have other combination that I didn’t come up with but will rank for those too since the important words are in the title.
Three days after the release with the new keywords, the app doubled the download and it has been steady for over a month now:
Thoughts and Conclusion
I have never had any mathematical or more thoughts whenever I choose a keyword. I used to look at the individual keyword and hope that they would rank for 1-10 and then I would use them. But these days when the App Store is so crowded there aren’t that many single keywords that one rank for position 1-10. So, using mid- and long-tail keywords are very important.
By doing the way as above I have put in some sort of logic in which keywords to choose. You can tweak the range where a keyword should get a point and by doing so get more or fewer keywords to choose from. Just to be sure that I don’t lose a vital keyword that I already rank position 1-10 for, but maybe don’t get 4-5 points, I go through the 1500 mid-tail keywords and look at the ones that I wouldn’t get in the new update and analyze if I should add those keywords anyway.
I used TheTool to track the keywords I chose and to get all my data. It also had the advantage that I was able to get all the data as CSV-files.
“TheTool helped me very much when finding the best keywords. It’s very easy to use and I believe the data is very accurate. The ability to export data as CSV-files makes it so much easier when I need to really evaluate keywords.”