The 3 Horizons for strategic planning
The future you write down is the future you create
For almost a decade we’ve used a strategic planning framework called the 3 Horizons. We use it to share our view of our current strategy and the future we hope to create, in a format that’s easy to digest without extra context.
The 3 Horizons framework was introduced to us by our chairman, John. He didn’t invent it, but if you Google it most of the descriptions aren’t great (or: they don’t match how we do it). So consider this the official definition.
What is a 3 Horizons?
A 3 Horizons is a document that you use to describe your current view of the future of your company. The content of a 3 Horizons is not assigned to specific people, is not directly accountable, and doesn’t need specific metrics. The purpose is to align everyone’s view of the future and encourage big thinking.
A 3 Horizons looks like this:
When writing one, make the categories whatever names make sense to you, but they should be high level and broadly encompass everything the business does. These are ours: how do you make money (revenue)1, what do you make (product), and what sort of company are you (organisation).
H1 is things you want to protect and grow. That is, things you’re doing now but could be doing better, or that are within reach but need intense focus to achieve.
H2 is opportunities to capture. Things where the path to achieving them is not totally mirky, but you need a bit of creativity and a lot of work to get there.
H3 is for when you create the future. This is where the big time ideas go, where if you get all the H1s and H2s just right you will be on the path toward.
3 Horizons example
Here’s an example of a 3 Horizons that I made up for Github. I have no insight into their actual metrics or strategy, but it should give you a feel for the sorts of things that might go into each section.
Even if it feels really obvious, it’s worth writing a few sentences to explain each point. You should be able to give a new starter - or even someone you are interviewing - a copy of your 3 Horizons plan, and have each point make sense without you having to say anything.
Horizons are not time bound
Nothing on a 3 Horizons is time bound. Something could be in H3 today, but it’s just missing a great idea or a key hire to make it H1. Or something could be in H2 for years while you chip away at it. A common mistake is to try and move through horizons on a set timeline - that is not correct.
That said, horizons tend to be sequential. It’s gratifying to write a H3 down, and come back a year later and see that it has moved to H1 - or be totally done! - as the business has evolved.
Because horizons aren’t time bound, they can be paired with other tools like OKRs that are tied to specific tasks with specific dates.
Running a 3 Horizons workshop
We have gone through many different approaches for writing a 3 Horizons over the years. Here I’ll describe how we did it most recently. I’ve also helped a few other companies run this playbook and it seemed to work well for them too.
Part 1 - Brainstorming
If you can, do this in person. If not, use Miro.
You should not have more than 20 people involved.2 In a small company, that can be everyone. In a bigger company, that’ll be department leads or senior staff.
If you’re doing it in person, you’ll need a big room, lots of butchers paper, a lot of post it notes, a lot of pens, and a roll of voting stickers.
For each category of your horizons, stick 3 sheets of butchers paper on the wall. So if you have 3 categories, there will be 9 sheets on the wall.3
Everyone has 60-90 minutes to write down any idea they like on a post-it note and stick it on the butchers paper. The aim is to generate a lot of ideas. Every sheet of paper should be overflowing a little - keep nudging people to add more stuff until it is.
Then everyone gets about 4 stickers, and they can use them to vote on any idea they like. This part can take 10 minutes or more so that everyone has time to read all the ideas. The purpose of the votes is to highlight ideas that resonated with lots of people.
You’re done! (Go to part 2.)
If you’re using Miro, make a grid with your categories as rows, and a column for each horizon. Split into groups of 5 people, and give each group a grid, and put each group in its own video call. Then repeat the same exercise but using Miro’s post it notes and voting stickers. I’ve found this works best if people are using two screens - one for Miro, and one for the call so the group can discuss ideas as they write them down.
Part 2 - Condensing
You have to do this part in person. Get the co-founders (or equivalent) together, and run the exact same exercise as part 1. You can use the output from part 1 to help inspire your ideas, but you can also add whatever new content you like.
The finished product will be the 3 Horizons grid, with a few items in each grid, and a short description for each.
Don’t rush this bit. If you don’t feel good about any specific items on it, go back to the drawing board. When the horizons is done, you should feel good about giving a copy of it to every employee, new hire, or candidate you interview. You should feel good about framing it and putting it on the wall in your office.
3 Horizons cadence
Because horizons aren’t time bound, there’s no required cadence to update them. Change them when they are out of date or when your view of the future changes.
We’ve found that reviewing the horizons once a year works pretty well. You may not want to do the full exercise every year, but you’ll definitely want to make substantial updates.
What makes a good 3 Horizons?
The first time you write a 3 Horizons it probably won’t be very good. This isn’t a unique problem - everyone who uses OKRs complains that the first few times they didn’t make sense. Any tool that forces you to articulate complicated abstract concepts in a succinct way is going to be challenging to do well, but you get better by sticking at it.
The rule of thumb I use for success is: do people refer to the contents of your 3 Horizons plan when making decisions? This happens a lot in the days after you ship the document (particularly if you also make it a poster and stick it on the wall); the real test is if it’s still happening 6 months later.
In the past we’ve had a 4th category, “Go to Market”. “Revenue” was more about numeric targets, and “Go to Market” was more about tactics.
Maybe you can, but I’ve never done it with a bigger group so can’t speak for it.
To make it fun I sometimes add an extra sheet for H4 ideas. H4 ideas are ideas that are too wild for any real horizon. In practice most of the ideas that end up on this sheet are actually good H3 ideas.