James Hawkins

I'm a little unclear on this topic with myself, but have some personal and messy thoughts at least....

I've been debating that "one north star metric" may not actually be the way to run a business. Or at least as treating other metrics as a constraint perhaps:

Ie hooray! we're growing 10x, but we're losing $100M a year.

hooray! we're increasing revenue 3x a year, but our sign ups are way down so we're screwed in anohter year

Another way I've thought of it is creating metrics that tie to our mission and high level product strategy, and then simply tracking those.

That'd be something like:

Mission: Equip every developer to build successful products

Metrics for example - how many WAUs / how many products of ours being used by each WAU / how many products are being tracked

and a few constraints ie:

Are we default alive / revenue growth over time

That makes sense, I had the same perspective and wanted to debate a little bit too.

Products are an holistic being. Cool then let's see how it goes, ty<3

howdy Toni

probs not the answer you want, but we totally self built it specifically so (i) our website would stand out more (ii) we felt long term we'd want loads of clever functionality like being able to offer merch / points and prizes etc to users for being helpful, and integration our users here with our product users too somehow - we didn't want to get limited basically

if you're curious, we did actually open source an older version of it (https://github.com/PostHog/squeak) - but that's an abandoned project now as we use ie Strapi for our CMS

Wow, good job, it's really nice! Thanks for the quick reply!

Yes, startup credits can be used for the Teams plan and any PostHog product or add-on(1). The $50,000 in startup credits can be applied to any mix of PostHog products over a 12-month period(2).

However, there is one limitation - teams in the startup plan cannot use credits towards a BAA (Business Associate Agreement) due to legal risk(1). If you need a BAA, you should contact the CS & Sales team to find a solution(1).

Once the credits are used up or expire after 12 months, your account will automatically move to a standard paid plan(1).

Max AI is right here! You can use startup credits for the team plan.

'An Elegant Puzzle' is where this term comes from - it's a book by Will Larson about engineering at Stripe.

Check out "the four states of a team" in this post for a good articulation of the points he makes. I'm referring to this!

Both should just be full time and 50-50, it’s such a discrepancy otherwise. Else they’re not really a cofounder they’re just dipping their toe in

thank you! i am always interested in this stuff. this is a different style. i like the concept of just having a "thing" for how you do feedback as a business.

similarly I use the phrase "genuine question" before asking something like "why are we building this" to try to highlight - this question is coming from a place of good intent, it may sound direct or leading, but i genuinely want to know why something is happening.

A free trial is a standard way to solve this problem so id start there, and I’d get people into a shared slack to be hyper responsive. We got much higher quality feedback from paying or potentially paying customers. I’d default to this unless there are good product reasons why freemium would be better.

I was pretty bad. I had some fairly good times, like I've done a 25 mile time trial in 50:57, or 100 miles in 3hrs43min (and something seconds...), but I lacked skill in road races.

As things get more competitive, your fitness to other people gets more and more similar so ie risk taking/skill differentiate you more. So I couldn't get anywhere above the level of full time but just about scraping by.

Hi James, those are pretty good times!

I used to be a half decent cyclist myself, was on the Irish national team for a few years, but the competition was so fierce... we were never that good compared to the other europeans.

I love what you have done with the company branding/culture/ethos. You are right, there are so many samey samey SaaS companies out there and none of them stand out. Well done on being different and taking a very different path.

I love the pricing ethos too... very well done.

The

(instance_url)/organization/billing

URI is an empty blank page. :\

Nice write up, looks simple enough.

Startups themselves are risky because they usually don't have product market fit or are unable to get it.

Once you're at a level of product market fit, then staying in touch with reality and not doing dumb stuff as a result is the basic risk.

Since we have product market fit, I'd argue the greatest risk we have is losing contact with reality. I'd then argue that this happens broadly for three reasons:

1. Ego gets in the way of seeing the truth
2. You don't have the right perspective to see the system you're in
3. You don't see the results of your decisions

For each of these:

1. Ego. We believe in product engineers making decisions for each small team. I suspect someone that loves building stuff as an individual engineer will likely have lower ego than the founder trying to build a $100bn business.

2. Perspective. It's hard to see what is going on without zooming out sometimes. We regularly challenge our small teams (quarterly planning, direct feedback, growth reviews) with people present from outside like me who have a different perspective.

3. Seeing results of decisions. The small team interact directly with visitors so have the best feedback loop. Much better than I have for example.

Favorite food?

Wood fired pizza, curry, cheese on toast and spaghetti bolognaise are all up there!

Drink?

Flat white in the morning, alcohol free beer in the evening.

Snacks to keep you going during long days?

Cheese, biscuits and pickled onions = yum

Two things, just to check you've seen these first:

• We have groups functionality for this explicit purpose
• This is an add on, you'll need to switch it on in billing to be able to use it

Hi James , thank you for you response.

Actually I am using the addon, so I just to make a Cohort of users that belong to a Group (not specific) . Example: Users that belong to companies to have 2 different cohorts: B2C users vs B2B users.

Yes - see our docs on identify

In your frontend, use something like:

posthog.identify(
  'distinct_id',  // Replace 'distinct_id' with your user's unique identifier, for example an id in your database
  { email: 'max@hedgehogmail.com', name: 'Max Hedgehog' } // optional: set additional user properties
);

Thanks James. So as far as I understand it, I have to manually link each user to their email address? So there is no automated solution such as "link all users from strapi database to posthog"?

yes - you need to create a fresh account at eu.posthog.com to do this!

(sorry for slow reply!)

Congrats on writing down your philosophy as you grow - that's an important first step.

This is gatsby but it's all custom. The reason we went custom is because we want this website to stand out, so it simply won't if it's the same as other people's. We've put a ton of energy into it the whole lifetime of the company, it's one of those things where doing it to the last 0.001% has given an outsize return because that's why people talk about it a lot - it's exceptional!

I couldn't replicate the issue on our end. If it persists can you report it via our in-app support here so we get more information? Thank you!

nope, this is normal - simply we track them separately until they're identified them merge them together so they'll pick up lots of extra ids along the way!

If you merge them together, does this merge propagate correctly to BigQuery if you've set up an export destination with a daily cadence?

hey Marta

we have an open source product (github.com/posthog/posthog) that doesn't include this functionality

separately, we have a cloud product (ie if you login at us.posthog.com or eu.posthog.com) and within this we offer your first 250 survey responses a month for free

does this make sense to you?

separately, the docs wording "free/open source" i can see is confusing so I'll make an issue to make that clearer