[jds375]

That data appears to be relative to the number of native apps in the app store, so that doesn't necessarily mean that moving towards non-native development is a misconception. In addition, many apps are not moving to non-native entirely... but instead module by module. I wonder how they account for that and how it impacts their statistics.

[arielm]

We (Appfigures) looked at the SDKs that apps use to determine what's native vs. non-native. This method will catch "partially" non-native apps as well as fully non-native apps as the former will need to have the non-native SDK even for a single module.

[mi5ha]

So basically these stats are representation of the usage of your SDK over your userbase right? And it seems that your SDK is not that popular with those using non native technologies?

Are your SDKs installed on at least 90% of all apps in PlayStore? If not 90% what is the percentage?

[arielm]

This data is for all new apps released and has nothing to do with our user base. In fact, we don't have an SDK.

The report uses data from our SDK identification engine, which works for all apps.

[mi5ha]

It would help if you released more details about your methodology. So you are not working on a sample of the millions of apps? You are working on all of them? Your "SDK identification engine" can detect in what technology each and single one of million apps in both stores is created with? Really? How?

Do you download each binary and examine its content? Even paid ones? Large ones? Obfuscated?

What is the error margin of your methodology? Or you are 100% precise? Sorry I am honestly interested in the details, and I didnt find error margin anywhere for this study or details about methodology.

[ASalazarMX]

You might want to correct a typo in the landing page: "Our indespensible reports..."

[arielm]

Good catch. Thanks!

[austenallred]

Is React Native “native” in this instance?

[arielm]

Nope