[edanm]

I also founded, grew and sold a boutique software services co., and am now running a new one.

I agree with all of the above.

The only major note I'd add is: some of the above really depends on who your clients are.

For example, we work with mostly technical customers, taking on things that are not their core strength. But only customers who already have internal dev teams. The way we phrase is internally is "we want our deliverable to be a git repo, and only work with clients who know what to do with that".

This implies a bunch of stuff - for example, ongoing server maintenance etc. is seldom a problem, since our clients already have their own devops teams.

Other dev shops can target the other way - non technical clients, for whom the dev shop becomes their entire technical staff. Those contracts are often quite different, and just to continue with the example above, often will include some kind of ongoing devops work (which may be more of a profit generator than the actual project itself!).

[mipmap04]

I've started doing this as well too. I don't want to be someone's hosting provider or CSP. Too many headaches and building the thing is a lot more fun than watching it run and making sure you've in compliance with the 1000 governance documents the company has regarding production environments. There are a lot of good companies that already provide purely that service and, to your point, many companies have those teams already.

[oneelectron]

I also founded, grew, but have not yet sold a boutique software services co.

I see on your profile you sold to a client. We have had that option before but the multiple on our earnings was too low to actually consider it. Curious what your revenue/earnings multiple was on the acquisition. Also curious about the handcuff terms. Giving up autonomy over my daily life is hard to swallow without 8 figures.

[edanm]

I can't get into exact details. It was a fairly typical multiple, a bit higher than normal after some negotiation. That's smaller than a startup, obviously, as there was no IP, just a team. However, since we were bootstrapped, all of the money went to us, none to investors.

It ended up being fairly profitable to sell, though we were by no means dependent on this - we could have just as easily continued and possibly gotten more money in the long run by not selling (with a greater risk, of course).

[tixocloud]

How do you get started with a custom dev shop? What type of technical companies would you focus on?

[edanm]

Depends on the field. We tend to try to offer complimentary services - e.g. if you do web, typical clients might be companies that do embedded, and need a web UI. Or who do mobile usually but now need some web expertise, etc. Or who focus on building security products, and now need a web interface to them.

Basically, try to find companies who are technical, but are missing the skill your company can provide, and then provide that.