Repositioning PostHog

As PostHog enters its next chapter, we have spent the last 6 weeks going all in repositioning as the company that makes your product self-driving. This is how we are specifically doing it, and how it's going. If you are trying to position or reposition your product, hopefully this will help you!

What does repositioning actually mean?

"Positioning" is when you set the context for how you want your product to be seen by customers. You're trying to get across how your product is the best in the world at something that your ideal customer profile (ICP) actually cares about.

"Repositioning" is when you change your positioning (woah). It's very normal to regularly review your positioning as the world around you changes.

Why did PostHog need repositioning?

We originally positioned PostHog as open source product analytics for developers back in 2020. Over time, we've expanded to position ourselves as a suite of tools that product engineers use to build better products. We've used terms like "Product OS" and "customer data infrastructure" that tried to capture this.

Despite this, most people still see PostHog as "analytics++" today even though we make money from a lot of different products. In part, this was because we did our repositioning incrementally, but mostly it's because repositioning is just hard, especially if you don't go all in. We updated website copy! We ran ads! We wrote blogs! But never all at once, and never in a way that felt like our entire team was pointing in the same direction and moving with the same urgency.

How PostHog is positioned now

Self-driving is now how we are positioning ourselves. We are going all in on this in an uncomfortable way - this means putting it front and center on our website, but also focusing on it in content, video, influencers, events etc.

Being all in means realizing this will probably confuse some people for a bit. It might hurt signups. We are willing to tolerate some short term pain, because we believe our long term success (and revenue!) depends on convincing people that you build with PostHog, not add it on afterwards. I feel like we wouldn't have a shot at doing this well if it weren't for the fact that our founders are totally bought in on this - it has to come top down.

You can see more in our public handbook, but here's the tl;dr:

• PostHog makes your product self-driving.
• Products are the surfaces a customer adopts to access self-driving. Today that's PostHog Web, PostHog Slack, PostHog MCP, PostHog CLI, and PostHog Desktop. PostHog Mobile is coming.
• Tools are the capabilities you access through the products: product analytics, session replay, feature flags, experiments, error tracking, surveys, web analytics, and so on. Tools can be as granular as annotations or comments even.
• Context is the data that feeds the self-driving loop: events, recordings, errors, and logs from PostHog, plus other business data (Slack, code, Notion, support tickets etc.)
• Context warehouse is the managed data warehouse plus the full context-ingestion pipeline (modelling, data pipelines, batch exports, and so on).

A large part of positioning comes down to some sets of specific words you are going to repeat everywhere until you are bored of them. They need to be specific and limited because our audience's attention is limited. Once your team is bored of hearing them from you, chances are that your customers have only just started to internalize them.

Where is PostHog playing now if not analytics?

We're now positioning ourselves as part of the market for self-driving software product platforms. We don't have a perfect competitor in terms of overlap, but some are reasonably close if you squint, especially if they are shipping "self-healing" or "self-improving" software.

The truest competitor for our ICP that we should care about is the old way of working. Triaging errors, watching replays back, writing the PR yourself, re-prompting an agent for every change, connecting your warehouse to your analytics to your coding tool - these are the habits that we are offering an alternative to.

Why are you inventing new concepts like self-driving and context warehouse?

When you're playing in a fuzzily defined or emerging market, you enter the realm of category creation. This is different to most positioning, where you are in an established market differentiating yourself against existing companies and products. Category creation is fun, high risk, and hard to pull off. Rather than building "Uber for dogs", you are trying to change people's mental model for how an established thing should work. This sometimes requires you to come up with new terms to describe what're you are doing, because it doesn't exist yet.

When a term has been widely adopted in the industry, we will use that term - we would not try and redefine "MCP" as something else for example.

However, "self-driving" and "context warehouse" describe things that don't really exist today, or where there isn't consensus yet. You may have heard the concept of "self-healing" or "self-improving" software for example, but these are very far from industry standards.

We have to work really hard to educate people about these new terms, to the point of feeling like we're repeating ourselves too much. We're confident we can succeed because we have some prior experience helping to popularize the concept of "product engineer". We didn't invent the term, but were one of the earliest people to hire product engineers, wrote about it a bunch (top of Google when you search the term), and now it feels pretty standard.

Are you open to changing your minds about this?

Maybe one day! But not right now - internally, we've had plenty of rounds of feedback and we've all had to disagree and commit to some extent to get this out there. Debating product names is the perfect way to distract your entire team for hours.

Part of repositioning is that you really have to commit to it for several months to see if it's landing. And of course, we'll change course if things look like they're going wrong. Early signs are promising from some of our best customers, and people are starting to talk about it on places like X. But it's only been a few weeks, and the majority of mentions still talk about us in the context of analytics -we'll get there.

As has always been the case, we will know whether we need to change course by talking to our users and making something they want.

This was mildly useful, but I'd appreciate a book recommendation

There are a million books about positioning. They are all compromised in some way. I speed ran April Dunford's process in a week from Obviously Awesome, and it's worked well enough for us.