The App Store is a black box. Years ago, doing App Store Optimization (ASO) was like having an unfair advantage —the search engine was too rough, there were fewer apps and few people actually optimized their apps.
This has changed as the App Store gets more and more complex and intelligent with each update. But how smart is it now?
When working on Keyword Optimization (KWO), it's common to face the singular-plural form dilemma. You have a limited space to work with when choosing keyphrases, so you'd like to use different keywords. Having "car, cars" in your Keywords field seems like a waste of space. But to delete one of them, you have to trust that the App Store will actually consider your app eligible to both singular and plural forms by having just one of them.
There isn't a definitive conclusion yet on how intelligent the search engine algorithm is when understanding that singular and plural forms mean the same thing, thus combining search results for both forms. We never trusted the App Store enough to ignore the decision between singular and plurals, or to go with just one of them. We've always searched singular and plurals and most of the time we got significantly different results.
We decided to further investigate this matter using the two apps for which we had access to keywords.
We all want to know if we should or shouldn't worry about singulars and plurals. To get there, we need to answer a few other questions first:
I chose two high-performing apps to serve as a basis for my research. I had access to its metadata on iTunes Connect, so I know the keywords are from the App Name and Keywords field only.
I decided on searching the App Store using iPhones. Although faster and easier to use, the desktop iTunes search has a different ranking algorithm. In other words, searches there don't simulate a user looking for apps directly from their device, because the positions are quite different. Why didn't I use an iPad? Because it has fewer apps than the iPhone and thus is a little bit easier to rank on.
So, looking at my table, I had two iPod 5's running iOS 8. While they're not actually iPhones, their App Store search works exactly the same. One of them had the US App Store, the other the French one.
On my computer, I had a spreadsheet with all keywords from the App Names and Keywords field, from the English and French localizations. I matched each keyword with their respective singular or plural forms, even when they didn't exist. Why? It would help me understand if the App Store just combines everything with “-s” and marks them as plural, or if it's more intelligent than that.
Looking for an app inside the results for a single keyword search could take some time—it’s very hard to appear in the top results for single words. So, I used keyphrases (more than one word) in all searches. Usually, it was the most relevant keyword from the App Name plus the keyword I was actually testing. So, there we were, searching keyphrase by keyphrase, looking for the app, and counting the position. Antiquated? Yes, but it feels better to get the results directly from the device's screen.
1. Does the app appear on search results for both singular and plural forms even when using just one of them?
We’d love to give you a “yes” or “no” answer, but unfortunately we have to give you an “it depends,” because nothing in the App Store is black and white.
Let’s look at the data:
Connection Between Singular and Plural Forms
App AUS Store - appeared in the opposite form search results for 92% of keywords
App BUS Store - appeared in the opposite form search results for 38% of keywords
The connection App A had in the US App Store now looks like more of an anomaly. Since I used some plurals that don’t exist (like the and thes), my first guess was checking if the App Store knows when the plural exists and when it doesn’t.
All the singular forms from this research exist as words, so I checked the plurals.
Comparing App A's English localization with App B's, there could be a positive correlation between made-up words and lack of connection. Maybe the App Store recognizes indeed when a word exists before connecting singular and plurals.
As for the French localization, even with a good number of real plurals, the connection didn’t pass 30%. The French algorithm is not as advanced as the English localization one.
2. Does singular trigger plural (and vice versa)?
This is more related to the word and language itself than the form, it seems. I couldn’t find anything that pointed to a preference for singular or plural words.
3. Does the search results position for one form match the other?
Most of the time: no. The English localization for App A had the highest number of matches: 37.5% of positions were the same when looking for singular and plural forms on the US App Store. The worst case was App B in France at 0%.
Does Search Results Position Match for Both Singular and Plural (when only one of them is present in the app’s keyword list)?
App A - US Store
App A - French Store
App B - US Store
App B - French Store
What can we take away from this?
Choose the form (singular or plural) you want to rank higher on. Even if you appear on both forms, the one you actually have in your keyword pool will receive a bonus thanks to being an exact match. If relevance between singular and plural is the same, go for the one with less competition. But always keep an eye out for unintended meaning with your plural or singular forms. Take "new" vs. "news" as an example. The top results are drastically different:
In languages other than English, if both keywords have similar stats, I recommend using both. For instance, App B was ranked 21 for a targeted keyphrase in singular and it was the second result for the same keyphrase with a plural noun. App B had both forms in the keyword pool.
To further prove this point, search for the same keyword or keyphrase in both forms, and observe how the search results positions change between apps. In most of the cases, it’s definitely not a perfect (nor good) match.
4. Is it better to use singular or plural forms?
For the English localization, if you’re using keywords that actually have both singular and plural forms and they share the same meaning, it’s pretty safe to go with singular forms (if you’re OK with losing some positions for most of the plurals).
Now, for languages other than English, don't count on the algorithm to combine the forms. It didn't match most of the simplest plurals (like words whose plural form was produced with just an -s or -x suffix). Just pretend that singulars and plurals don't exist, consider everything.
The ideal scenario, for maximum optimization, is always analyzing singular and plural forms as different words—regardless of language. The search results position difference is significant enough (20th to 15th, 16th to 10th, 10th to 6th) to be worth taking into consideration both forms, even when working with English Keyword Optimization.
This research could’ve contained hundreds of apps with 15 languages each. But we do get a great sense of the App Store’s singular-plural abilities with this small sample There’s ample room for improvement in languages other than English. For this language though, if the word has real singular and plural forms, the match is intelligent enough.
I still recommend ignoring the match. Even if with a bigger research I got numbers that pointed out most of the keywords were correctly combined, there’s the exact match bonus. This alone is enough to study the metrics for singular and plural.
Should you use both forms then? When using both, you take space away for different keywords. Compare the stats for the keyphrases you’d be losing with the ones you’d be ranking with both forms —again, looking for relevance first, than competition, than traffic. Think like you’re comparing different keywords, as usual.
If you want to learn more about app store optimization, take a look at the WordData blog.
What about your experiences? Have you noticed differences when changing between singular and plurals? Share with us in the comments below!
Thanks for signing up. Look for your first email shortly!
We’ll reach out shortly to schedule a time to talk.