3 ms·
For user-generated content, a blank slate is usually discouraging to contributors. Additionally, you need to set expectations for what sort of content you wan
by twobyfour 9y ago
For user-generated content, a blank slate is usually discouraging to contributors. Additionally, you need to set expectations for what sort of content you want users to generate.
The thing you should do that doesn't scale is to seed your content area with content generated in-house that mimics the type of content you want to see. Then invite a couple dozen users, engage with them heavily, and continue to generate your own content.
Your goals are 1) to make the community seem like it's active and something users will return to frequently to find new content; 2) to make users feel like contributing is a rewarding act that gets them the sort of interaction and encouragement they seek; and 3) to establish norms for your community.
Then invite another cohort and another. Each new cohort can be an order of magnitude larger than the last, and once each new cohort is engaged, you can scale back your own contributions and engagement significantly.
- jaski 9y agoThanks a ton! A reddit-esque UGC seeding plan was already in place before I posted, but the slow growth wasn't appealing to me since I think we'll get 10k app downloads the day we open the gates. But thanks for making it clear that it's not just the number of downloads that matters... engagement is key. Since the UGC part is totally new, goals 2 and 3 may be difficult to reach with 10k users on day 1. So yes, we'll plan to grow slowly and focus on delighting users.