Make a living giving it away, fifteen years on

Update, July 2026: I have had to withdraw from WordCamp US 2026. My passport will not be renewed in time for me to cross the border for the August dates, and I could not find a way around it. I was sorry to send that note to the organisers, who could not have been better to deal with. The talk itself stands, and I would still like to give this version somewhere before the year is out. What follows is what it was built to say.

The first time I gave this talk, in November 2011, the premise was still something you had to defend. Free software was obviously good for the world. Whether it was good for your own bank account was an open question, and I spent a decent chunk of that session making the case before I could get to the mechanics. I gave it in Toronto that month and again in Detroit a week later, and a rebuilt version of it went to Buffalo in 2023.

Phoenix was to be the fourth outing, and the first one at a flagship event.

The session that was accepted, Make a Living Giving It Away, was a breakout on the schedule for Tuesday, August 18, at WordCamp US 2026, which runs August 16 to 19.

I did not want to simply dust off the 2011 deck, because the question has changed shape underneath it. Nobody in that room was going to need convincing that a person can build a career adjacent to work they gave away. The people I talk to have already done it. They maintain a plugin with a few thousand installs, or they wrote the tutorial that a whole category of developer learned from without ever knowing whose name was on it. What they are doing in 2026 is arithmetic. They are looking at the hours going into the free side of the ledger and asking, out loud in a way they were not asking five years ago, whether the return is still there.

So the sceptical questions still come, just from a different direction. In 2011 people wanted to know whether this could possibly work at all. In 2026 they want to know whether it still does, and for whom, and that is a better question than the one I started with. It deserves a session built for it rather than a rerun. I wrote about the acceptance, and about what a strange week it arrived in, in Behind the Scenes: a WordCamp US yes, and four machines that lied to me politely.

The short version:

• The talk is about what you give away and what you charge for, and working out which is which before the giving is already done.
• The common failure is not giving too much away. It is giving away the one piece that was holding the paid work up, usually by accident, one reasonable decision at a time.
• The test I use is simple: the free thing should make the paid thing easier to find and more obviously worth paying for.
• It is for people who already contribute, maintain a plugin, or otherwise give work away, and who want that arrangement to stay financially survivable.
• Thirty years of doing this in public, and getting paid for most of it, is what sits behind the argument.

What the session actually covers

I want to be careful about what I am promising here, because there is a version of this talk I am not qualified to give. I do not know what large language models finally do to the economics of open source labour. My opinions on that are held loosely, and I am wary of anybody currently selling certainty about it. That is not the session.

What I do have is thirty years of doing this in public, and most of that getting paid for it. I have been building for the web since 1996 and working in WordPress since 2007, and for most of that time some meaningful portion of my output has been free to anyone who wanted it. Plugins, documentation, and a great many whole afternoons. That is a long enough run to have made most of the available mistakes at least once, and to have noticed which ones cost real money.

The paid side of that same work, when a team wants the documentation or the plugin built properly rather than handed over for free, is what my writing work covers.

The talk is about what you give away and what you charge for, and mostly about how you work out which is which before the giving is already done. The order matters. Almost everyone I know who got into trouble with this got into it retroactively, discovering after the fact that the free thing had become the whole product.

Think about how a restaurant handles bread. The basket arrives before you have ordered anything, nobody itemises it, and no owner in the world thinks of it as a loss. It gets you seated and comfortable, and it makes the thing you are about to pay for taste like part of a larger experience. The bread is doing work you are not being billed for. Every operator who does this well knows precisely where the bread stops and the meal starts, and they know it before service, not halfway through it.

The failure I run into most often is not people giving too much away. It is people giving away the piece that was holding the paid work up, usually by accident, usually one reasonable decision at a time. A free plugin picks up a feature that used to be the reason somebody hired you. A generous answer in a forum turns into an unpaid implementation. The wall moves on its own, because nothing in the day-to-day forces you to look at where it is standing.

The test I use is boring and it works: the free thing should make the paid thing easier to find and more obviously worth paying for. When the free thing starts substituting for the paid thing instead of pointing at it, you have crossed a line, and the sooner you notice the cheaper the correction. That is most of what the talk argues, with the specifics of how I have priced and occasionally reversed my own decisions over the years.

The test I use is boring and it works: the free thing should make the paid thing easier to find and more obviously worth paying for.

Who it is for

This one is aimed at people who contribute to open source, maintain a plugin or a theme, or otherwise give work away as a normal part of how they operate, and who would like that arrangement to keep being financially survivable. It assumes you already know why you give things away. It is not a beginner session, and I will not be spending time on the case for open source, because the room will have made that case with its own labour long before it walked in.

The Buffalo 2023 outing is on the site if you want to see where the material stood then.

If your own event needs a version of this, or something else entirely, my speaking page has the details. The arguments I get in the hallway are the reason this talk keeps changing, and I still want the next version to be different again. If your room is where that happens, you know where to find me.