Now! The answer is: now. Let’s dig into why and how to.

Olle Pridiuksson

Aug 31, 2019·7 min read

Again and again, each time I am consulting a #middleware #startup they’re asking the same question: “when is it the right time to show the devs what we’re working on.” My answer each time is: NOW! And then I hear: “but we’re not ready yet” — YES YOU ARE!

TL;DR:

The software is never ready, you know it. There’s always a backlog, always a way to improve it, always an ugly code to rewrite. You’re not solving anything by being heads down. Yet you’re missing all the opportunities for early feedback and the power of #community when you’re eventually ready to start selling your software or services to fellow programmers.

Your software is never perfect anyway

You estimate that in 8 months you’ll have the docs, the demos, no ugly code and gonna ship your software for other devs to use. We both know your estimation is on the optimistic side of things, but what I’d love to focus on now is the shipping part.

How do you ship your stuff to devs? I mean, do you can hire a PR firm to push out a press release? Do you buy some ads on Facebook? Do you just ship your plugin to the Unity Asset Store (or whatever is a relevant plugin/SDK store for you)?

See where I am going? If you take these 8 months to assemble a fan base, a community of developers who you’re solving the problem for with your new tool, then you’ll have a loyal prewarmed community to ship to. Many of them will be the first to try, to buy, to join the alpha and beta programs, to spread the word about you.

Yes, your current code base is early, the product is in its infancy and the duct tape keeps everything together. But what if you’re humble and open about where you are now while very loud and ambitious about where you want to be in 8–12 months? Would that sound appealing to fellow programmers?

What do you talk about while you’re 8 months away from shipping

From the day 0 till forever you talk about what exactly problem you’re solving, why it matters, how cool it gonna be when it is out. You can drill down into some of the specifics of your software, what is revolutionary about it, explain elegant solutions in your code, your toolchain, your processes. When nothing else is on top of your head, you can just push your sprint demos and even the sprint planning backlog online…

See, the middleware/SDK/plugin user and the purchase decision-maker is a programmer, so the earlier you start showing your work to fellow devs, the faster you figure out what is your main value and the killer feature in their eyes.

Often it differs from your investor slides deck and your backlog priorities. You’d rather figure it out early instead of when you ship.

How do you talk to your community while you don’t have one

Starting a community is always a leap of fate. Talking from a twitter account with 3 followers is frustrating. So shall you think about a place where your potential future users/customers may hang out? A subreddit, a forum, Stack Overflow — try googling for the keywords that matter to what you’re doing and watch where the buzz is happening.

Then think about the kind of content you can create there. It should be of OK value for the community but super cheap to produce for you. Maybe a sort of weekly thoughts on a tech problem you’re solving? Maybe some help and advice here and there?

You’re not selling anything. You don’t have much to sale yet. You’re farming your social score in those communities, your credibility. You still have 8 months ahead.

This is a long shot to assemble a community at some point, so you can talk more about what you’re working on.

Then blogging comes to mind. Good technical long reads overall are cool yet consume so much time to do well that they happen once or twice a year. Maybe. While twitter can survive any kind of content. #covfefe…

All your daily thoughts, week logs, sprint demos, pictures, screenshots — everything can find a place in your company’s twitter feed. Multiple tweets a day are fine. Also, be sure to tweet (even as screenshots) your replies and overall activity at all that other communities.

Glueing two thoughts together we have an action point: start a company twitter account and link to it from all the professional communities you’re participating at.

Again, don’t be shy with content quantity and quality for twitter. Use Tweetdeck if you want some structure.

Your GitHub is your landing page, probably

Programmers like watching code, playing with code, seeing how code works and don’t like flashy marketing pieces with catchy slogans. Guess, you can relate?

Shall we name your GitHub as the main landing page then? And your company website is there just for the “Buy” button plus marketing and legal. Also the support forum. When you have the “Buy” button, you’ll need to have a support forum for people to complain and vent and for you to promise them bugfixes. You don’t want that to happen on twitter.

Now take a deep breath, remember about your code not being good anytime anyway, and think about opening your GitHub repo to your community 8 to 6 months before you ship.

Right. All your source code, if you’re with the open-source model. And all your issues, docs, API ref, the source for the demo projects, the license file and everything else that could help you get the trust of other developers, your future users.

Yes, it is hard and scary to open up when there are duct tape and hacks everywhere. But you do want your main repo to get 100 stars, right? And all that influencer marketing thing with people following your work on GitHub and their friends noticing that and possibly also following you there.

Breathe out. That GitHub part was a long shot. You know it is probably a good idea, just a bit tricky to make right. I’ll think aloud about where to start next time. I hereby inspire you to follow me here =)

Last bit about your website is that as soon as it is out, be sure the “Subscribe to the newsletter” button appears first. Even earlier than your marketing slides and the “Buy” button.

The collected emails will be your launch community. You’ll remember about it each time you have big news to roll out — the first beta, the big demo, the new venture round you had…

Remember GDPR. Respect privacy. Don’t spam.

You’re also a YouTuber suddenly

When you were looking for online communities and when you’re googling for pretty much anything, there are always links to YouTube on top. Now, what if other devs looking for relevant keywords would find your tool as a solution to their problems?

Probably a good way to think about a new YouTube channel for a middleware startup is: marketing your #software to programmers looking for a solution to a problem that your tool solves.

The challenge here is to figure out the cheapest possible way to produce good enough quality content for possibly unsophisticated users to get answers to their specific questions.

I am framing YouTube like this since until you build the audience your channel will only be getting one-off views, mostly from your social channels and a bit from the search. So high production value content probably makes little sense at the start.

I’ll blog more about YouTube if there are enough claps.

All these activities sound like a full-time job

I’ve covered a lot today, even while touching only the basics of what can be called developer relations. A full-fledged developer relations program owns even more by taking as much as possible off your engineering department’s plate. Here’s a relevant long read about what is #devrel.

If you’re still here and need a one-line takeaway, let it be this:

if your primary customers are programmers, you’ll have to roll out a developer relations program anyway. So consider it as early as only possible.

3 months ago I was en route to solving the world’s developer relations and developer experience problems by building a framework, a set of developer relations/developer experience practices and implementing that all as a contractor. Here’s my long read about it.

I expected my [nonexisting] sales and marketing pipelines to be a bottleneck. But nope. The demand for #developerrelations #consultancy was there, all around.

Everyone nowadays is building the next big SaaS thing, looking to expand to new developer verticals or has finally figured out that the world’s programmers own the budgets now and have the power to decide on what services and software to use in their company. And these programmers are immune to most marketing and sales funnels yet are very open to developer advocates.

So where am I in that process of solving the world’s problems? I am starting 3 local devrel meetup groups (in #Sweden, #Lithuania and TBA, maybe #Minsk?) to connect those interested to hire an evangelist/advocate/tech writer to those willing to jump on the travelling extrovert programmer role.

I’ll put an update on how this goes. So probably follow me here or on twitter, if curious about the #devrel #dx topic.