We have been making a lot of assumption in designing our online community because online communities is not a novel idea. We wanted to ask people questions about how they interact with their existing communities (e.g., neighborhood, hobbies, and friends) and how they find things to do or eat.
We posted across several existing communities (e.g., Reddit, IH, Slack) to get volunteers and ended up getting 12 volunteers, including several friends.
When doing interviews, we asked open-ended questions (e.g., How do you connect with your neighborhood?) and used their answers to ask them what they like or dislike about each solution they mentioned. Extracting the good and bad about existing solutions allowed us to discover common frustrations people had. We took these frustrations and developed a new solution to add to our MVP based on real data.
Interviewing people was a great way to connect because you build rapport with them by listening to their "problems," and we believe this kick starts our community around the solution we are building.