On April 29th, 7 years ago — I pushed publish on a Kickstarter campaign for half baked prototype called @Ghost — with some wild ideas.

Today, we're a team of 16 with a just over $2.1M ARR and some even wilder ideas.

Here are some of the highlights 👇🏼 [ 1 / ? ]
I've done various blog posts with lessons over the years, but 2020 is weird so I thought this year I'd just share some random anecdotes instead.

Let's begin at the beginning:
20% of the UI in the Kickstarter video was faked because we were up against a deadline and needed to launch. The dashboard was img tags inside some divs which had offset CSS transitions. Also there was no image uploader. That was done in Final Cut Pro. https://www.kickstarter.com/projects/johnonolan/ghost-just-a-blogging-platform
Was balance of showing real demo of the parts that actually showed me+ @ErisDS ability to ship (editor/app/content management) -- whilst not delaying shipping by messing around with things that were hard to build, but added little in substance!

Good mix in hindsight.
Lot of people complained about the "unrealistic and completely unnecessary fake bird sounds" that had been added to the demo video.

I wish they were fake:
At the time Ghost hit Kickstarter we were the 3rd most funded open source project of all time. Now that Diaspora and Ouya are both gone/inactive - I think we might be the oldest crowdfunded open source projects still around(?)

"BuT WhAT aBoUt ThE DAsHboArd???"
Dashboard was a technically unviable photoshop mockup that was very easy to kill. We had to choose between shipping on time, and making a fancy dashboard. We chose shipping.

Some people are *still* upset about it today.

Those people will never ship.
Our first retreat was in Egypt and cost less than $1,000 for 4 people. We did the first, I think, 3 retreats in my house. I used to cook my secret chili recipe on the first day when everyone arrived. We did about as much drinking as working. And we worked a fucking LOT.
For first 5 years we had only 1 sysadmin running all of Ghost(Pro) - with 600million requests for many of the largest sites in the world, see: https://ghost.org/customers/ 

This obscene ratio *only* possible thanks to @digitalocean & their amazing product + support over the years ❤️
For the first 6 years we had only 1 person on support. We like to joke that @SarahLFrantz doesn't rest, she just charges over USBC.

Srs though, support for a technical product is a VERY difficult and tremendously undervalued job. Quite literally wouldn't be here without Sarah.
Our first $1M in ARR took 5 years.
The next $1M of ARR took 1 year and 9 months.
We're on track for the next $1M to take 8 months.

Compounding growth really works as long as you stick around for long enough.
Weird thing about growth: Almost nothing actually moves the needle. Vast majority of growth hacking, AB testing, and advertising is complete shit.

We tried a LOT. Only 3 things that worked:

- Improve product
- Raise prices
- Better plan limits/upgrades

Simple really.
Best move in last 7 years:
Choosing not to raise money and be profitable instead.

Worst move:
Overcomplicating technology. Constantly. Simple code is better. No code at all is best. Always try to make it smaller.
Thing that nobody really talks about: it's not all fucking good.

I've burned out hard multiple times. So has Hannah. Many sleepless nights and absorbing an insane amount of pressure for literal years. Nothing comes for free, and it isn't always fun. It often isn't.
I'd say first 5 years were mostly pain with occasional glimmers of hope. Last 2 have been just as stressful, but with fewer undertones of impending doom.

There is real sacrifice involved in achieving anything, no matter what anyone says.
Remember: Most people only tweet good things, which makes it very easy to feel like you're doing shit and everyone else is great all the time.

It's not true though. You're probably doing better than you realise, and others doing worse than you imagine.

Perspective.
We're rapidly approaching the end of what might be considered a reasonable length for a tweet storm, even though there are so many stories I haven't been able to fit in here.

So I guess I'll try to finish up with what we're planning for the next 7 years:
Ghost has always been about enabling independent publishing, starting with creation/distribution.

Next, we're solving the business model. Here's how: https://ghost.org/members/ 

In beta RN, but a lot more coming.
Last month we also opened up fresh opportunities to support Ghost (yes, we're still a nonprofit org) via GitHub Sponsors: https://github.com/sponsors/TryGhost

Or if you'd like contribute with code, we could always use the help :)

Thanks for everything
You can follow @JohnONolan.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: