One year of building a product analytics tool

Thanks George, glad you enjoyed it :)

With momentum it's hard to avoid stating the obvious, which is that you have to be consistent...there's no momentum without that. That doesn't mean working like a dog, just that you need to figure out the amount of work you feel confident you can do each week over a sustained period of time. That also means weighing up the impact on your mental health, family, day jobs etc...and obviously you should still take time off and vacation etc too! You're more likely to succeed doing 5-10 hours per week for a year than trying to do 40 hours a week alongside your 40 hour a week day-job and burning out after 2 months.

That said, consistently doing the wrong thing is worse than doing nothing at all. You need to be regularly talking to users and driving their feedback into your product. If you're talking to enough users often enough then you'll start to see common patterns in their needs - for me those are the feature requests to prioritise. Sometimes those won't align with what you thought the product should be. We didn't intend to end up with customer feedback widgets, but it's what people were asking for and so far our biggest customer is primarily using those features - and there are a bunch of expansions to that functionality lined up! Anyway, the point is that your momentum is also partly sustained by the knowledge that you're working on the right problems, and the satisfaction of making your users happy.

I also think the phased approach we have taken is important. It's not like we planned it super rigidly, but so many people set goals that sound like PMF goals, when really they should be setting POC or MVP goals -> if you're chasing goals that just aren't achievable during that phase of product development then you're going to lose motivation, and then momentum, really fast.

Hopefully there's some useful advice in there. I still get this stuff wrong all the time, Squeaky could still fail, we could still lose momentum. I've experienced that before, and learned some hard lessons from past projects e.g. burning out from over working around day job/family life commitments, not talking to users enough and validating my ideas and direction repeatedly, hoping to go viral on PH and HN like all those stories you hear and then getting disheartened when projects died because really I needed a proper go-to-market strategy and lead-gen tactics etc...no matter how good the product was.

I can say from my experience as a German, yes, especially German companies (but also others in the EU) don't like to use anything that is hosted outside the EU, just to be on the safe side regarding GDPR

In relation to where the business is based, the friction for customers is never especially high, it's as a business owner it sucks if you get this wrong.

Privacy/compliance related matters aside, it helps if you're based in the EU as the tax and VAT stuff is a little more streamlined for all companies within the EU. We have customers in multiple EU countries and it's considerably less painful than if we'd registered our business in the UK (just ask all the UK entrepreneurs that have opened Dutch companies to make it easier for themselves post-Brexit).

All I was really getting at in the incorporation part of the article was that it made sense to incorporate in NL in the long term, but in the short term is was really painful and expensive because it's not as straightforward or affordable to incorporate in NL vs UK (and even US, though it's pretty bad there too). I've incorporated in all 3 at one time or another and it ranks:

1. UK is the easiest to incorporate and has excellent (modern) banking options too, but it's now the worst place of the 3 to do business internationally from because of Brexit.
2. NL is very expensive to incorporate, and the banking is a bit painful to set up, but it's actually pretty solid and is super helpful for selling within the EU. I also live in the Netherlands, so obviously I'm biased.
3. US is not quite as expensive as NL, but the legal and banking sectors/services are a decade behind the EU and especially the UK, so the whole thing feels incredibly painful every step of the way.

At least all 3 countries have really favourable corporate tax compared with many other countries in the EU - whether you agree with it ethically or not - though I'm curious to see how this universal minimum of 15% corporate tax plays out...not that that has anything to do with VAT obligations. Likewise all 3 countries have great tools and services to support bookkeeping, accountancy etc.

thanks so much
I appreciate your answer

Using react or not won't impact your SEO. There's lots of great articles about technical SEO, I suggest to google that term :)

The 4 sources were:

• 3 people from my direct network -> people I knew who run businesses and used or needed similar tools.
• 5 people from different online communities I'm part of (a couple of IHers, and some people I know from winning Pioneer a few years back.
• 2 people who used to be part of a small community we ran before starting Squeaky. Basically we emailed the 250ish people that were part of that community to (a) let them know we were fully shutting down the community and (b) what our new venture was ...and a few of them were curious enough to try it out.

Their feedback was amazing, but only 3-4 of them are still actively using the product. That's okay though, at the MVP stage we didn't really think that our product was going to be worthy of them sticking around, especially if they had other tools already in the mix.

The first 10 users are probably trying your product out of a mixture of kindness, curiosity, eager to share their expertise or input, or they like to help other entrepreneurs etc. Obviously you should be reaching out to people that want/need (or already use) the sort of product you're building though, but you're probably not going to make these 10 people fall in love with what you're offering if you MVP is as shitty as it probably should be.

We connect via WebSockets to a Go microservice that streams in the events data, compresses it and writes it to Redis as a buffer. When the session is complete a message is placed on a Sidekiq queue. After a period of time, a Ruby background worker will pick up the message from the queue and fetch the data from Redis, decompress it, process it and store it in ClickHouse and Postgres.

haha, hey Shane - we don't have a page like that I'm afraid! I think the cookieless aspect is overblown anyway...it's not hard to make a cookieless analytics tool...cookies are just one of various storage options after all.

That is actually one thing we are thinking to address though - not the 'having a page about the cookieless approach' per-se - but more that we might consider going open source because otherwise people just have to take our word that we're privacy friendly and that's arguably not good enough (even though we are as privacy friendly as a say!). To expand on that a bit:

• Cookieless
• No IP tracking/storage (no hashing or anonymisation of IPs either...the only time we check IP is to screen out your own IP so it doesn't make your analytics data skewed).
• Data input into forms is suppressed by default
• You can hide/surpress any element on a page from our script, so really you can avoid capturing any PII whatsoever (form data isn't the only PII display after all)
• EU data storage, also private instances available on request
• We don't have access to people's analytics account unless they expressing grant permission
• We have a DPA (Data processing agreement) that anyone can sign.
• We're launching a consent widget at the end of this week to help users better manage consent of their visitors when required (quite often it isn't required though).

I do kind of think this should just be the bare minimum we expect of analytics tools nowadays though. What's the name/URL of your tool?