Guest Post: 5-Steps to Lean Startup Prioritization That Actually Works

-5 Steps to Lean Startup Prioritization

(This is a guest post by Dan Toma, an Accelerator Manager and Coach, and author of the upcoming The Corporate Startup Book. You can find Dan on TwitterLinkedIn or his blog.)

Like with so many other good ideas that looks trivial on paper, but people struggle applying, lean startup prioritization using a Hi-Lo diagram will be classified, by the ones that tried it and failed, as exactly that: a good idea that even though on paper makes sense, in real life it’s creating more hustle than it solves.

Being a big fan of this way of prioritization ever since I’ve read David Sibbet’s Visual Teams and Visual Meetings, I’ve tried it in many circumstances and actually wrote about it on a couple of occasions.

Looking back on all of my attempts of using the method, it is safe to conclude that, my success with the method had more to do with the fact that I was using it alone and the number of things I tried to prioritize was quite limited, than with me actually understanding how to build and use a HiLo.

illustration - HiLo Method 1 - prioritization matrix

I’ve spent this past summer in Vietnam being involved, as part of a team with Nick and Ryan, in a Train of Trainers program for an Official Development Aid program of the Finnish Ministry of Foreign Affairs.

Part of the Train of Trainers Program we, together with our trainees, had to, on a regular basis, prioritize things such as: upcoming ecosystem development events, future government entrepreneurship policies, university curriculum modules, and more.

To be able to prioritize such complex topics, in a large group like ours (close to 20 people), we employed the Hi-Lo method, only to realize that we were failing miserably with every attempt and we were creating more arguments than we were solving.

The problem was that everyone in the room had a strong opinion and everyone wanted to see their ideas implemented, hence whenever someone was asked to walk to the board where we’ve created the Hi-Lo, their sticky-notes (at least 5 per person) with neatly written ideas ended up in the upper-right corner of the diagram (the higher and righter the better J).

As you might imagine our Hi-Lo looked more like a square (the upper-right quadrant) full of sticky-notes than an actual prioritization tool. Honestly, we were better off prioritizing alphabetically than with the Hi-Lo method. At least there wouldn’t have been that much follow-up work.

Since it is human nature to consider everything vitally important, some of you might end up experiencing the upper-right quadrant phenomena even when prioritizing alone a handful of items. Clearly this problem is bound to happen with any team larger than 4 members generating more than 2 sticky-notes per person, unless a new way of looking and creating Hi-Los is developed.

Talking with Nick and Ryan we realized that the problem was our process of creating and using the Hi-Lo, and not the people being actually honed in with themselves, and the place on the diagram their sticky-notes belong to.

Our biggest mistake – and the one I’m sure everyone that tried the method did – was to create 2 axes from the get-go.

We realized that by creating only one axis at a time we are going to have a fairly evenly distribution of items on the final diagram.

So here’s step-by-step process of creating a Hi-Lo that actually works and helps you prioritize items (lean startup prioritization):

Step 1: Create and name the OX axis and mark its middle point.

illustration - HiLo Method 2

Step 2: Start creating writing all items on sticky notes.

illustration - HiLo Method 3

Step 3: Start placing the sticky-notes on the OX axis.

illustration - HiLo Method 4

Example: if you try prioritizing assumptions around your business model, you might want to start by naming the OX as ‘knowledge line’. Assumptions that you know more about go on the left side and assumptions that you have less knowledge about go on the right side.

Step 4: Create and name the OY axis. Make sure it passes through the middle point of the OX axis.

Step 5: Without changing the relative position of the items on OX axis start moving them up and down relative to the OY axis.

illustration - HiLo Method 5

Example: Create the OY axis and name it: impact on business model, with high impact on the upper part of the axis and low impact on the lower part of the axis. Now start vertically moving the assumptions based on their perceived impact on your business model. Remember: you are not to change the relative position of the items on the OX axis.

A couple of months after developing this 5-step process to creating a HiLo diagram I had to deliver a workshop on experiment design at Estonia’s largest startup accelerator, Tehnopol.

Before I taught the participants how to create relevant experiments that will move their businesses forward, I made them prioritize all their business model assumptions using the 5-step process heretofore described.

From what used to be a tedious task, now I have entrepreneurs walking up to me, face smiling, saying that didn’t know prioritization is such a fun and easy thing to do.

Discussion (8 comments)

  1. Alex says:

    Hi Dan!

    Great simple method and article. Did the X axis now change character, though? There is still a difference between impact on sth. vs. how much I know about it, isn’t it?


    1. Dan Toma says:

      Sure Alex, impact reflects the level of risk you are going to take if you move forward without validation. How much you know is basically telling you if you need to sped time validating and how much time – here is important to be humble and true to yourself (don’t be overly confident about how much you know.)

      1. Alex says:

        I don’t see why impact of something directly correlates with knowledge about something. If you don’t know it, how can you even guess its impact?

  2. Sam McAfee says:

    Great article. The question I have is how do you ground the “perceived impact on your business model” in any kind of real data? It seems that placing the stickies on an axis based on some kind of gut feeling is more an a priori prioritization based on what you already think is important, whether it’s by an individual or group discussion, rather than using something like revenue or some other non-priority metric to drive to priority as a posteriori outcome of the exercise. What are the the ways to handle this problem?

    1. Dan Toma says:

      Sam you are right some of the things were based on pure intuition but another thing I found useful (and you might give it a go if you have time – kind of time consuming) is to use the Scrum Poker method.

      Basically you number each axis from 0 to 100 with the middle point being 50 and than you start playing Scrum Poker for each sticky-note, one axis at a time. Makes sense?

      This way you involve the entire team, you will reach consonance and it’s a more scientific way than pure intuition. (I know that this is not solving much of the ‘you don’t know what you don’t know’ problem but I think is more accurate)


  3. Pingback: Importance = Business Impact * Uncertainly -

  4. Kevin says:

    “We realized that by creating only one axis at a time we are going to have a fairly evenly distribution of items on the final diagram.”


    Is it because you “hide” the intent of the exercise by drawing one axis at a time?
    If that’s the case, wouldn’t the exercise work only the first time with a given team? Why wouldn’t the team put all items on the right hand side of the x axis the second time they use this HiLo version, the same way they will do with the 4 quadrants?

  5. Agen Sbobet Online Terpercaya 50rb says:

    Hello, this weekend is pleasant in favor of me, for the reason that
    this time i am reading this impressive informative post here at my residence.

  6. Pingback: Guest Post: Market Opportunity Navigator by Sharon Tal-Itzkovitch – Thoughts from a Lean Startup Coach & Innovation Ecosystem Designer

Got something to say?

This site uses Akismet to reduce spam. Learn how your comment data is processed.