Öffnet Profilfoto
Folgen
Felix Krause
@KrauseFx
Currently building contextsdk.com - Security Researcher - Creator of fastlane.tools (previously ) @KrauseFx@mastodon.social
Vienna, Austriakrausefx.com/aboutSeit Juni 2009 bei Twitter

Tweets von Felix Krause

Gotta love when this happens when switching from VS Code to Xcode. Has been useful for Twitter engineers recently I heard
Bild
4
87
Did anyone catch any changes around the use of in-app browsers in iOS 17?
Tweet zitieren
🔥 New Post: Announcing InAppBrowser - see what JavaScript commands get injected through an in-app browser 👀 TikTok, when opening any website in their app, injects tracking code that can monitor all keystrokes, including passwords, and all taps. krausefx.com/blog/announcin
Diesen Thread anzeigen
Bild
3
43
Can’t believe phone cameras have come so far 😍 This is my iPhone 13 Pro. And yes I cleaned the lens.
Bild
12
88
6 larger apps now with enough Context SDK performance data, all of them looking very promising! For each app, there are two buckets, showing the conversion rate: → Bucket "Bad": Context SDK predicted it's a bad time to show a prompt → Bucket "Good": Same, but for good moments
Bild
4
20
Diesen Thread anzeigen
Just published many additional insights, graphs and takeaways:
Tweet zitieren
Have you ever considered that your app may be used 28% of the time while the user is in their bed? Most of us don't really think about in what kinds of situations our apps are running. However there are huge differences when it comes to conversion, and optimising your app:
Diesen Thread anzeigen
Bild
1
5
Diesen Thread anzeigen
Larger apps already leverage this kind of technology to optimise their apps: - Audible shows a different UI with larger buttons when driving - Uber is smart about when to actively ask you to upgrade to a Uber Black based on location, time and other signals
Bild
Bild
2
10
Diesen Thread anzeigen
Besides the activity, you should consider other signals as well: - Users in low power mode convert 37% less - Users with their phone in critical thermal state convert 27% less - Users on a phone call convert 26% less
Bild
4
19
Diesen Thread anzeigen
The activity detection happens completely on-device within less than 2 seconds. Context SDK comes with pre-trained models trained on dozens of hours of activity data generated by people of various age groups and body types. Every person walks and sits differently.
Bild
Bild
Bild
1
1
Diesen Thread anzeigen
Grouped by each activity, you can see clear signals: This particular app has more positive outcomes (sales) when the user is in their bed or sitting or standing. While the phone lying on the table, the user walking or on a bike or car causing sales to go down.
Bild
1
4
Diesen Thread anzeigen
Have you ever considered that your app may be used 28% of the time while the user is in their bed? Most of us don't really think about in what kinds of situations our apps are running. However there are huge differences when it comes to conversion, and optimising your app:
Bild
7
115
Diesen Thread anzeigen
What’s your favorite example of an app using context (time of day, location, etc) to subtly improve UX? Some examples • Strava suggesting photos from a run • Spotify car mode • Health’s Medications prompting you after a time zone change
4
29
Antwort an
• Audible renders larger button when you're driving • Uber does many things around context, e.g. upselling UberBlack • YouTube deciding between Shorts vs Videos when launching Context matters, but until now only the large players had the resources to really make use of it
Bild
Bild
1
8
Setting the Spotify Podcasts playback speed. They really came up with the best options here, combined with the best UI 💯
Bild
14
86
The fact that the first 4 results when you search something on Google are ads is proof that Google is going to lose search dominance. There's no way they can get away with this for much longer. I'll take ChatGPT with a few hallucinations any day over sponsored content.
178
1.016
You can use Context SDK to - Get a boolean value on if now is a good time for showing a prompt - Get specific information on the user's current activity (which was trained on dozens of hours of activity data from people of different ages, heights and body types)
Bild
5
37
Diesen Thread anzeigen
Conclusion 4: The real fun begins once you combine multiple signals - you can see clear patterns in what makes in-app conversions successful. Use that info to optimize when to ask the user to upgrade or enable a feature.
Bild
2
34
Diesen Thread anzeigen
Conclusion 3: There is a big difference in conversion rates depending on the iPhone model, but more importantly, which iOS version the user has installed. People who update their OS are more willing to spend money.
Bild
3
54
Diesen Thread anzeigen