The Developer Advocacy Handbook

If you find this helpful, you can buy me a coffee.

Prepare for outreach

Lanyard and bag

As a developer advocate a lot of your job is going out there and tell the world about the things your company does (or technologies, techniques and methodologies it uses). Your success in this is to a large part relying on how the world sees you - are you a tech guru or somebody who tries to sell their company or some product? Your integrity is your main power - you have to make sure that it stays intact. This means first and foremost that you need to prepare properly before going out there.

Get your facts right

You will be asked to talk about a certain new product. Make sure that you are up-to-date on the matter before you go and speak about it. Do not promise things that are not under your control. Talk to the product team and ask them in meticulous detail what the product is about, what works, what doesn't and so on. Be as skeptical as possible as this is what the people you are about to talk to will be.

Know the audience and their needs

Your communication should be targeted to the audience. People came to listen to you or read your article with a personal agenda - if you fulfill that agenda you win. Know what people expect and need and you can deliver. Otherwise you need to hope for the best - and that is never a good plan.

Going to conferences costs money. Going to free events where you speak costs time. Make it worth while for the people who do either and try to get something into your presentation that they can go back to their company with to wow their bosses. That way they will be able to go to more conferences and your other events.

Example: A long time ago had to give a talk about Yahoo BOSS (now defunct) to a search engine optimisation crowd. They loved that Yahoo's search index is open for remixing but also were aware that in the country I gave the talk Yahoo only had 5% of the search market. I worked around this by building Keywordfinder, a tool using the Yahoo Search Database to create keywords. Giving them a cool tool to get keywords related to certain showed them how versatile BOSS is (well, was…).

Have expert backup

You cannot be the expert in everything. In the best case, when giving a presentation try to have an expert at hand to answer tricky questions for you. If there is no expert available at the time note down the question and follow it up after consultation. Do not promise to come back to someone and then forget to do so - that’ll make you look like you needed a fast way out! There are far too many speakers out there who play the “I'll get back to you” game.

Under no circumstances try to wing it and promise things you are not sure the product team will be able to deliver. You are here to promote what can be used, not put pressure on your colleagues by promising the world the moon on a stick.

Choose the right medium

Your communication should be in the right format for the intended audience. This can be many things: slides,videos, audio, live coding exercises, online step-by-step examples or streams.

Tip: My rule of thumb is - the more technical the audience, the fewer you should use PowerPoint or Keynote. Show how you can code with the product, not how shiny it is or what its workflow is.

Plan for failure

Things will go wrong and you need to be prepared. In the case of a presentation do this:

Example: Once I gave a keynote at a conference about Artificial Intelligence and my overall message was that whilst machines are faster and more efficient than we are they are also not infallible and augment the mistakes we made. This was a conference with a complex stage setup that needed me to send them my slides in advance and be played from their computer. Promptly the presentation machine crashed and my slides somehow got out of order. After the first five there was darkness and nothing worked any more. So I took on the stage and used the failure of the stage setup as a hook to tell the rest of the story. Halfway through my story the slides came back up starting at the wrong one which gave me a chance to talk about what machine learning can do wrong with data that isn’t in the correct format. The tech issues could have been the end of the keynote, but by rolling with the punches I got the audience on my side and brought the message home.

Next: Get speaking opportunities