How we spend our marketing budget at PostHog (with actual $ figures)

Nov 28, 2023

Posted by

On this page

It really annoys me that no one is willing to talk about how much they actually spend on marketing in startup-land.

When I’m trying to figure out what a sensible amount is for PostHog to spend, all I can find are articles that talk very generally about marketing budgets in terms of things like % of revenue or general benchmarks. Which is nice, but the ranges given are wild and I want cold, hard numbers.

How much does a typical startup like PostHog spend on B2B marketing. And why doesn’t anyone talk about it?

“But Charles, you naive idiot” asked nobody, “surely competitors will use this information against you?”

Um, sure I guess? But especially if you’re into product-led growth (like we are), how you spend marketing $ seems pretty far down the list of competitive advantages. We already tell everyone what we’re going to build next, and that doesn’t seem to have hurt us so far.

By sharing some of this info, I’m hoping it’ll help others as a data point. I am not saying this is the best way to spend your marketing budget. I have no idea! That’s the point!! But I’m hoping maybe someone will see this and tell us how we could do better...

But first, some context...

Startups are different (woah, much insight!)

Here's some background about PostHog before you start copying this stuff and then sending me angry messages when this tanks your company:

  • B2B open source dev tool
  • 43 people, $27m raised
  • Team is 70% engineers, product-led growth
  • Post product-market fit

Let’s get into it.

Show me the money

Our typical monthly marketing spend looks like this:

  • People: $72,000 per month
    • 3 content marketers
    • 1 product marketer
    • 1 graphic designer
    • 1 frontend developer
    • 0.5 exec who mostly gets in the way
  • Not people: $73,000 per month
    • Tools & subscriptions: $2,000
    • Paid ads: $40,000
    • Agency to manage the paid ads: $5,000
    • Sponsorships:
      • Newsletters: $15,000
      • Influencers: $5,000
      • Open source projects: $2,000
    • Merch: $3,000
    • Events: $0
    • PR: $0
  • Total: $144,000 per month

Some notes

  • We do all of our content in-house. We have very occasionally worked with freelancers in the past, but we haven’t done this for a long time.

  • Technically our frontend developer is not part of the marketing team (Website & Docs is its own team at PostHog), but I appreciate at most companies your website will be considered a marketing expense.

  • The 50-50 split is coincidental.

If you want to learn more about how we implement these things in practice at PostHog, it’s all in our public handbook.

Going deeper on paid ads

I’ve written about running paid ads channels before, but this is how that $40k split looked over the last 3 months on average (as the channel mix varies month to month):

  • Conversion
    • Google Search - brand keywords: 10%
    • Google Search - product keywords: 50%
  • Awareness
    • Twitter: 10%
    • LinkedIn: 10%
    • Quora: 10%
    • Random experiments (e.g. G2, Bing): 0-10%

The barrier to implementing PostHog is reasonably high in the grand scheme of SaaS tools - you need a software product that has users, instrumenting PostHog isn’t totally trivial etc. This means we don’t use the channels listed under awareness to try and drive sign ups at all.

What we’re thinking of changing

In 2024, we’re going to keep our budget broadly flat in the ‘not people’ category, but will probably move things around a bit:

Reduce:

  • Paid ads spend across a few channels we’ve been using purely for awareness. These don’t seem to be driving high paying customers for us. Hypothesis is that investing more in posting organic content will be much more effective.

Increase:

  • Spend on developer influencers, as we’ve been running some small experiments here that are looking promising.

  • Start doing events. We’re not sure what this looks like yet.

  • Partner marketing. Probably more a sales thing, but this is something we’re potentially interested in.

On the hiring side, we don’t currently have plans to hire additional people onto the team – we’re much more reactive here depending on how things go!

What do you think?

Are the figures what you expected? Should we be spending more money somewhere else? Are we doing something very stupid?

Let me know!

Moar reading

Subscribe to our newsletter

Product for Engineers

Helping engineers and founders flex their product muscles

We'll share your email with Substack