4 ms·
Great Article, thank you for sharing. It's interesting to follow a Product Launch and see what works / what doesn't work. I like how you've integrated a specifi
by shawnreilly 13y ago
Great Article, thank you for sharing. It's interesting to follow a Product Launch and see what works / what doesn't work. I like how you've integrated a specific UX into the Product. Right now the Signup is probably one of the most important aspects to focus on (as you onboard Users). Choosing to skip E-mails (and E-mail verification) for a faster signup lowers the entry point barrier and makes the UX more enjoyable. This is something I'm currently working on with my App (the initial Signup Process). I've performed some customer validation and the results were in line with my assumptions. The preferred experience seems to be minimal time investment, and minimal effort. Following this train of thought, here are few basic options:
1. User manually inputs an E-mail Address and Password (commonly w/ E-mail Verification)
2. User manually inputs a Username/Password (less to type, no Verification needed)
3. User taps twice and uses Social Media Integration (example, Facebook Login)
Of course each of these has Pro's and Cons. I will follow Popcorn and see how it works with the second option. I'm leaning towards the third option (minimal time investment), but I'm concerned about perceptions of the associated Social Network(s) and how that might impact the User's desire to create an account. For example, if a certain Network is "not cool" with the younger generation, and you present the "not cool" Network as the Logon method, would that make the App also "not cool"?
- drum 13y agoYou need to really think about 3. Asking someone to sign in using facebook or twitter is much more than the physical action of tapping twice. you're asking for something emotional, their personal data. It would help others and myself if you posted what your app does. i would only use those as a form of login signup if your app absolutely depends on it - Tinder is a great example of this.
- shawnreilly 13y agoI completely agree. Using option 3 has its own set of con's, and this is one of them. Many Users have been misused/tricked in the past and don't like the idea of associating a new (aka unknown) App with their Social Media Accounts. For all they know the App will steal their data and spam their friends. This is in line with some of the feedback I get when validating with potential customers. You may have hit a sweet spot there with option 2. Keep us updated! An interesting Metric would be comparing the Bounce Rate between these options above (ala A/B Testing).