I agree with a lot of the points in this article, but I definitely believe it is possible to overcorrect and harm user experience in pursuit of these goals, and it’s something that I’m definitely noticing with PostHog.
> the biggest risk isn’t that they’ll become a bad company or fail outright, it’s that they become a fine company that could have been exceptional.
> You become afraid of: Losing your culture, market position, and momentum
From the outside, it seems like PostHog is trending towards the first quote while forcing the second one.
The issue I find is that it appears PostHog is focusing more on marketing/quirkiness than product experience.
This really came across with the new homepage experience. As a dev, I totally respect the project, it looks great and I can tell a lot of work went into it. IMO, it also shows the wrong prioritization.
As someone who has worked at a company with a similar business model (open source software with a dedicated hosted version), these are the issues I’ve really come across:
* Broken builds: There have been npm packages that have been released that cause apps to fail to build. In the case I noticed, it took your team more than 24 hours to realize this, remove it from npm, and release a new fixed version. Does your team have CI/CD testing to check builds and prevent this from happening?
* Outdated documentation: Your React and Nextjs docs still include links to outdated sub-docs, some including code for methods that no longer exist in the latest versions of PostHog js. If this is intentional, you don’t include a notice that the approach is different after release #X.
* GitHub issue backlog: Your PostHog js GitHub project has 185 open issues. (And I do understand it can be hard to maintain issue backlogs), but when I submitted a valid issue, it took weeks to finally get a response with a fairly simple fix in the next release. Transparency helps, a message should have been created as soon as the issue began being worked on by the team.
* Communication: In yesterday’s SHA1 HULUD attack, it took a coworker to send a link to the PostHog tweet saying to downgrade packages above a version number. A comment on this post (and I can confirm) mentioned that an email should have been sent out to PostHog customers immediately with this info, the odds of devs seeing a tweet are super low for a critical security issue.
* The ingestion downtime last week
Of these issues, several were occurring while you were updating your website to include a page on employee feet pics.
And sure, maybe you value the quirkiness that comes along with hiding things like this on your website to maintain company culture, but it’s not necessary, it’s harmful to product image and experience.
I think there’s a disconnect in your approach to company culture. Instead of a monolithic unit, company culture is something that should be split between internal and external culture.
I’ve worked at big tech that had no company culture. I’m also incredibly lucky to currently work at a company with amazing internal culture. We’ve gone on multiple incredible retreats, and I will gladly hangout with every single one of my coworkers outside of work. The work we do is meaningful, and we have a lot of fun developing it, even using it in our spare time to identify new features that we would personally benefit from, even if they’re not the most serious. But: At no point did we ever prioritize uploading employee feet pics to our website over fixing issues with our product.
Is PostHog truly not afraid of “losing company culture”, while prioritizing unprofessional Easter eggs over fixing actual product issues? Even if you make the argument that the teams are siloed, and those working on the new website UI don’t interact with product fixes, I think this means you need more product (and infra) engineers!
TLDR: Take feet pics off your website and fix your product. In your own words, you are losing your exceptionality by focusing on external image. Your product prioritization hurts your image with those implementing your services, and really makes our team wonder if your bundling-style benefit truly outweighs the reliability that other established products have, even if it means using multiple at once.
so good, thanks for this post
There is so much value in this post, thank you
I agree with a lot of the points in this article, but I definitely believe it is possible to overcorrect and harm user experience in pursuit of these goals, and it’s something that I’m definitely noticing with PostHog.
> the biggest risk isn’t that they’ll become a bad company or fail outright, it’s that they become a fine company that could have been exceptional.
> You become afraid of: Losing your culture, market position, and momentum
From the outside, it seems like PostHog is trending towards the first quote while forcing the second one.
The issue I find is that it appears PostHog is focusing more on marketing/quirkiness than product experience.
This really came across with the new homepage experience. As a dev, I totally respect the project, it looks great and I can tell a lot of work went into it. IMO, it also shows the wrong prioritization.
As someone who has worked at a company with a similar business model (open source software with a dedicated hosted version), these are the issues I’ve really come across:
* Broken builds: There have been npm packages that have been released that cause apps to fail to build. In the case I noticed, it took your team more than 24 hours to realize this, remove it from npm, and release a new fixed version. Does your team have CI/CD testing to check builds and prevent this from happening?
* Outdated documentation: Your React and Nextjs docs still include links to outdated sub-docs, some including code for methods that no longer exist in the latest versions of PostHog js. If this is intentional, you don’t include a notice that the approach is different after release #X.
* GitHub issue backlog: Your PostHog js GitHub project has 185 open issues. (And I do understand it can be hard to maintain issue backlogs), but when I submitted a valid issue, it took weeks to finally get a response with a fairly simple fix in the next release. Transparency helps, a message should have been created as soon as the issue began being worked on by the team.
* Communication: In yesterday’s SHA1 HULUD attack, it took a coworker to send a link to the PostHog tweet saying to downgrade packages above a version number. A comment on this post (and I can confirm) mentioned that an email should have been sent out to PostHog customers immediately with this info, the odds of devs seeing a tweet are super low for a critical security issue.
* The ingestion downtime last week
Of these issues, several were occurring while you were updating your website to include a page on employee feet pics.
And sure, maybe you value the quirkiness that comes along with hiding things like this on your website to maintain company culture, but it’s not necessary, it’s harmful to product image and experience.
I think there’s a disconnect in your approach to company culture. Instead of a monolithic unit, company culture is something that should be split between internal and external culture.
I’ve worked at big tech that had no company culture. I’m also incredibly lucky to currently work at a company with amazing internal culture. We’ve gone on multiple incredible retreats, and I will gladly hangout with every single one of my coworkers outside of work. The work we do is meaningful, and we have a lot of fun developing it, even using it in our spare time to identify new features that we would personally benefit from, even if they’re not the most serious. But: At no point did we ever prioritize uploading employee feet pics to our website over fixing issues with our product.
Is PostHog truly not afraid of “losing company culture”, while prioritizing unprofessional Easter eggs over fixing actual product issues? Even if you make the argument that the teams are siloed, and those working on the new website UI don’t interact with product fixes, I think this means you need more product (and infra) engineers!
TLDR: Take feet pics off your website and fix your product. In your own words, you are losing your exceptionality by focusing on external image. Your product prioritization hurts your image with those implementing your services, and really makes our team wonder if your bundling-style benefit truly outweighs the reliability that other established products have, even if it means using multiple at once.
tell it to em Josh 😤