Contents
Guide to the developer audience (based on insights from devs on Hacker News)
Developers who hang out on Hacker News are a notoriously tough crowd to market to.
But why? And what would good marketing look like to them?
Running a developer audience research on this group would be really difficult. But "by accident" I got a ton of insights from them which I documented in this article.
So here is what happened:
- I wrote an article about advertising to devs
- someone posted it on Hacker News
- it got to first place and stayed there for ~8 hours
- it got 170+ comments from devs and dev marketers
This is the Hacker News discussion for those interested.
Ok, so what did I learn?
Quick caveat.
Just remember that this research is based on some comments from one post on HN which is not the entire dev population. So it doesn't apply to every developer persona.
I even wrote a separate post about marketing to devs on Hacker News if you are interested.
But, you can still draw some insights from it. Here are my top learnings:
- Devs do hate advertising. The level is extreme, unreasonable even. With strong emotions (hate, disgust, contempt) it may be hard for some folks to stay analytical and think straight about your ad. Running ads can have a bad long-term brand impact even if you see short-term conversions.
- Devs love swag. Unreasonably so. They can jump through sooo many hoops for a t-shirt.
- Devs appreciate good tutorials and blogs. Some don’t even consider developer content marketing. But you have to be helpful, not salesy. Great examples are Digital Ocean blog and Interrupt (Memfault) blog.
- Devs only really trust other devs. 10x more than non-devs and more 100x than vendors. Even devrels should beware of being thrown into the “saleseman bucket” which can happen if they push their product too much.
- Say what you don’t do to build trust. Devs want to understand what you do. No fluff, no rainbows, and “saved time” but what you actually do. But maybe even more importantly say what you don’t do. That builds trust and makes them believe in other things you say. (btw, I wrote another post about creating dev-focused value propositions if you are interested).
- Devs use a lot of ad blocks. Not only typically display ads, or youtube ads, but even sponsored mid-video promo content blockers. Not all devs obviously, but many (my guess is around 30%).
- Podcast ads work. Potentially because of no ad blocks yet. Expensive but people got results.
- Say what you do, let me try, wait for me to reach out. dev marketing 101. Devs want to DIY, let them. Show your price, make it easy to try with no credit card, and don’t bombard with outreach.
- Devs think marketing == promotion and ads. Avoid using the word “marketing” with devs. Try to do marketing things that don’t immediately fall into promotion/ads bucket.
Devs mentioned a few companies that do great dev marketing (and seeing devs say it is like, yeah):
- JetBrains (for pricing and packaging)
- fly.io (for ads and posts on HN)
- DigitalOcean (for blog, went deep into how their dev content marketing works here)
- Tailscale (for ads and posts on HN)
- Memfault (for blog)
- Framework laptop (for ads)
Now, let's go deeper into each of these.
Devs do hate advertising
Turns out that many devs do hate ads :)
And when I say “hate” it is not “meh another ad”, but more like “I will hunt this advertiser down” feeling.
Not all devs obviously, but that is the direction where this audience is shifted. The HN community is probably on a more extreme and vocal side but still.
Ads trigger extremely negative emotions in devs
Things like, hate, disgust, contempt, and anger come up a lot. “sociopathic behaviour”, “offense to human psyche”, “visceral response of disgust”.
Those are very strong words.
With this level of emotional reaction, can’t imagine anyone is thinking analytically and pragmatically about your offer or creative.
Paradoxically devs who praise themselves for thinking “logically and practically” are not thinking “logically and practically” about ads.
But there is a good reason for it.
They got burned so many times before that as soon as they classify something as an ad they stop thinking straight about it.
Understanding that state of mind when creating ads (or any marketing materials for that matter) is crucial.
Think about handling this “it is just a f-in ad written by a non-tech sociopath” objection before you work on anything else.
Aggressive conversion tactics have long-term brand effects
Optimizing conversion with pushy tactics may give some short-term results (CTR, sign ups) but it has negative long-term effects.
Devs will remember you being pushy and even if they did download your ebook or signed up for a free version of the tool they will think twice about doing something with you in the future.
And if they don’t like interactions with your brand it is unlikely that they would recommend it to a dev friend.
That is perhaps the biggest problem as word of mouth is absolutely crucial with the dev audience.
Can your ad actually be relevant?
Yeah, this is not easy. Targeting the right people is hard, especially with devs. And so showing a relevant ad is hard.
This is interesting, but it is exactly why running ads may be an important part of the larger strategy imho.
Showing new products that solve new problems or problems people didn’t realize they had.
One more that touches targeting.
Again, show relevant ads. If people signed up don’t show them “sign up” ad. But you can show an ad to a resource/docs about a use case to activate users.
Devs love swag. Unreasonably so.
Turns out many devs will do a lot for a t-shirt.
There is even a website with free dev tool swag promotions. WHAT?
Even though they could have just bought it with a fraction of the time they invested in getting it.
Why do they do it then?
IMHO it is the challenge and/or belonging to a tribe.
Some swag t-shirts make it easier for folks to show the world their cool personality.
So good swag t-shirts could be about:
- belonging to a company: Google, Facebook hoodies/t-shirts
- belonging to a craft: “We are DevOps” from GitLab
- being a part of a group that understands the idea “Everything will be 200 OK”
Swag campaigns can be an acquisition strategy
I mean if people want t-shirts so much and will try an API to get them…
… then just do it. Do it at scale.
One campaign style mentioned is this:
or this one:
And it doesn’t have to be t-shirts.
I loved this campaign that someone mentioned.
Battle axe. Yep.
Devs appreciate good tutorials and blogs
This should be the core of any dev marketing strategy IMHO.
Even devs consider this “technically marketing but helpful and useful”.
Devs google and search optimization matters
Only some devs are on socials or YouTube but pretty much all devs google things (or search in other browsers).
It means you have to make your content easy to discover in search results.
You can probably pay for that traffic too.
Imho paid search traffic, especially when someone is looking for a “tool for XYZ” is fine.
Folks are looking for products and are happy to see what is out there.
Don’t be salesy
But that article you write should not be just one long sales pitch though.
Happens less often when devs write articles.
But even devrels/dev advocates should beware of “hard selling”.
One thing is you lose your super-power of being considered by other devs as a dev (trust factor).
The second is that persuading devs is just not working most of the time.
People love the Digital Ocean blog
Digital Ocean blog was praised by many people in the thread.
They focus on tutorials that solve particular dev problems.
They have thousands of those “How to do X” articles getting millions of unique readers every month.
And the promotional part?
CTA with free $100 bucks to run stuff (from the article or not) on Digital Ocean.
I go deep into their developer content strategy in this post.
People love Interrupt blog
The interrupt blog form Memfault was also mentioned.
Took a quick look at it and it seems to bring a lot of deep insights to embedded devs.
I will probably write a separate article about learnings from their blog.
Devs only really trust other devs.
Devs don’t expect non-tech people to understand their problems, the least of all marketing people.
So every “marketing page” or content is looked at from this lens. Probably a bit of bs, with too many benefits, and nothing about the cons, with a lot of fluff in between.
Because of that, they’d rather see what other devs are saying about your tool or content.
Say what you don’t do to build trust.
Devs know all too well that everything comes at a cost.
They got burned many times before.
Saw a shiny new tool that will solve all their problems, tried it, and then it wasn’t what they said on the website.
(almost) every product that does something, screws something else.
And it is ok.
But devs want to make that decision for themselves. They want to understand the pros AND the cons of using your tool.
And most marketing doesn’t deliver on that. It is mostly benefits, hidden under some vague specs. It feels like “propaganda” not like a spec. Devs love specs though.
So when you say what your tool doesn’t do you can win a ton of trust karma.
And start building foundations of a relationship between a dev and your brand.
Devs use a lot of ad blockers
Probably not a big surprise but devs use a ton of ad blockers.
Most of the regular stuff including display ads, search ads, and YT pre-roll ads will be blocked on most machines used by devs.
But even more “organic” things will be blocked as well.
Yeah, there are tools to block sponsored parts of Youtube content as well. And devs use them.
That leaves us with in-feed ads on social platforms (Twitter, Linkedin, Reddit, Stackoverflow).
One note is that the reality is a bit different from what HN commenting devs would claim.
The acutally percentage of devs blocking ads is probably closer to 30%. Not 90% + which some comments may led you to believe.
Podcast ads work.
Seems that ads on podcasts work well for folks.
Perhaps because there are no ad blocks yet, or because podcast hosts have gained enough karma with listeners by providing value that they let it fly.
Ads on dev podcasts are really expensive though.
Say what you do, let me try free, and wait for me to reach out.
Devs are building and want to do it themselves (DYI).
Let them.
For me, a big chunk of developer-focused go-to-market can be summarized by this:
- say explicitly what you do and don’t do. Let me see docs, specs, case studies, and examples.
- be clear and transparent about the pricing
- don’t make me talk to sales
- let people try your tool, preferably on their use case
- support them as they are building
- wait for them to reach out with problems
Or as a fellow developer marketer put it. Treat them as cats :)
But most dev-focused go-to-market motions are not exactly how devs would want to buy.
They are more about how companies want to sell.
I know it is easier that way, but sorry, it doesn’t work like that.
The best dev-focused companies are really good at that:
Too much fluff in dev-focused communication
Say what you do, what you don’t do. Be explicit, and to the point. Cut the fluff.
This alone will get you 90% there.
Don’t push for sales/contact too early
Devs are DYI they don’t want to talk to anyone to see the demo or get the pricing.
They want to explore on their own with examples, docs, tutorials, and the real product.
Let them try it out with no commitment.
Be transparent and clear about your pricing
It is not that devs don’t want to pay for anything and build everything themselves.
Maybe some but most understand that their time is more valuable than a monthly subscription for your tool.
But they need to understand what that subscription is.
Is it 10x more than their budget?
That is what they will assume if the pricing is not on the website.
And they will go with a tool that does tell them (even roughly) what it will be.
Devs think marketing == promotion and ads.
People just don’t get what marketing is.
People think it is just pushy pop-ups, fluffy copy, and irrelevant ads.
I think this is our fault, marketers. We did this to ourselves.
Now when someone does good marketing (in positioning, audience research, content strategy) it is “maybe technically marketing”, or devrel, or product strategy.
The lesson from this is that even though we marketers know marketing is so much more than ads, people don’t.
Especially devs don’t want to have anything to do with “marketing” or “advertising”.
So when you want to get devs help with marketing, you should avoid words like “marketing”, “advertising”, “promotion” and use “product adoption”, “community growth”, “helpful tutorials” instead.
BTW we need to do better with marketing the marketing itself, as it seems we really suck at it, people :)
What is next?
OK, so what is the plan?
If you based your marketing just on the insights from this one HN thread you should do the following.
Start an education-first blog:
- make it education-focused
- be actually helpful to devs
- SEO-optimized. Most of the traffic comes from people googling stuff
- cut pushy promotional stuff
- don’t make “one long sales pitch” articles
- soft CTA at the end is fine
Create a buying experience that is DIY focused:
- explain what you do and what you don’t do on the landing page
- make pricing available and easy to understand. Nobody wants to talk to sales to get the quota.
- let people try your tool with almost no commitment
- make it easy for people to reach out when in trouble
- don’t “attack” devs with sales outreach early
Facilitate word of mouth
- as devs trust other devs you need to make it easy for them to recommend you
- remember that every interaction with the dev builds the brand that someone wants to recommend or doesn’t
- build a feel of a brand that is helpful to devs to build some positive brand affinity
- make it easy for devs to tell what you do and send other devs somewhere on your site to find out more (what you do and what you don’t do again)
- run swag campaigns so that devs can show that they associate themselves with your brand
Run swag campaigns:
- do t-shirts contests
- t-shirts should help devs tell something about themselves: belonging, sense of humor, the tech they believe in, the tech they don’t like.
- connect those contests to brand building and word-of-mouth
Run ads but be careful:
- Google search ads are fine: people are googling a ton
- podcast ads are expensive but work
- make sure your ads are relevant. Seriously.
- handle the “this is a fk-ing ad” obstacle before you do anything else
Hope this was helpful. What did you think?
