Re. #1: Good question :) The day-to-day is pretty relaxed. Skylar and I meet every Monday morning to catch up on the latest stuff: what we’re working on, what client work is in the pipeline, and what needs done to get to where we want to be—whether that’s launching a new feature or something a bit more long-term. Outside of the Monday morning catch up, we don’t really have any other meetings unless we’re meeting with a client. Most of the back-and-forth between the team is through Slack, and we have a channel for each product and client to keep things organized. Most of the stuff we know we need to do is a GitHub issue in the corresponding project’s repo and we try to use GitHub’s Milestone feature to group the issues by priority. This setup isn't perfect but it works okay for a small team like ours.

One thing that we do with Dropmark that pushes us to keep things moving forward is a monthly email that we send out. This was partially inspired by something Nathan Kontny wrote about doing when he took over Highrise from 37Signals:

On day one, I established a train schedule – we'd make major announcements on a regular basis. If something isn't ready, it misses the train. But an announcement is going out; something better be on it.

We started doing a similar thing at the beginning of the year, and at the end of every month we send out an email — most months we’ve had a new feature or improvement that we’ve been able to announce.

Re: #2 It’s definitely tricky to balance self-initiated projects with client projects, especially in the beginning of your company/project. Consulting brings in money (and can pay really well!), and self-initiated projects usually just cost money (and getting paying customers is hard!). From the start we mostly just worked on our own products in downtime between clients, which resulted in a very sporadic schedule. It really helps to be passionate about the self-initiated project, and that drives you to find the time in whatever small ways you can.

We’ve also experimented with different ways of building our own products. With Symbolset, for example, Mike and I slowly poked around with ideas for nearly a year before we actually launched something.

Gazette on the other hand, was built in one crazy week. Myself, Sawyer, and Larry spent a weekend hacking together some code and polished it throughout the week. When we launched it wasn’t even fully functional yet (except for the sign up form), but we knew we had 7 days to complete and ship the first issue. We made the deadline, slightly stressful but super fun. Even though Gazette is no longer around, it was a great learning experience (pro tip: if you build a product, charge more than $1.99/mo if you want it to be sustainable).

Re: #3 I wish we could say we have a secret formula for this, but really just involves experimentation and determination. Not everything we’ve tried has been a success, but typically everything we do is built with a purpose in mind and one that scratches an itch. We feel like if it's something we’d use, others may feel the same way. Even if it's not, just having the thing exist for yourself (or one client) may be “real” enough to start. We actually had 2 working versions of Siteleaf for years prior to the v1 public launch. Nowhere near as polished as it is now, but all along the way we had clients using it. So it served a purpose (we didn’t have to use Wordpress!) and was the “real deal” to at least some people.

Re. #4: We really focused on building a CMS that developers would love to build for and that everyone would love to use.

So some things developers will love is:

• Full compatibility with Jekyll — it’s open source, you can run it offline, and it’s the same thing GitHub Pages uses to power over half a million sites.
• GitHub Sync, which keeps your source files locally, in GitHub, and in Siteleaf all in sync. When you make an edit in Siteleaf or push a local change to GitHub, everything syncs which gives you a time machine for your content.
• A full API that developers can use to manage their content. Hell, you could create your own CMS using just our API. (The Siteleaf Admin uses the same API endpoints). I’m using the API personally to export my Kindle and iBooks highlights online.

From the client or content manager side: I’m a bit biased but I think we’ve built and designed a pretty kickass admin interface for you to manage your site in.

Re. #5 for me:

I’m obsessed with usesthis.com — it’s super interesting to me to see what other software developers are using in their practice, but also what scientists, chefs, and musicans are using. The site even has an API.

I also have to mention Kevin Kelly’s new book, “The Inevitable”. You gotta read it.