Pretotyping Challenge, Step 2 — Construct your XYZ hypothesis

Robert Skrobe
Dallas Design Sprints
5 min readFeb 19, 2021
“If I just put all of my previous failed experiments in this machine, I’ll surely get one that won’t fail!”​

If you went through Step 1 of our pretotyping series and have your initial Market Engagement Hypothesis (or MEH, for short) written up, it’s time to pair that down and get specific about what kind of engagement you want to measure. In other words, you’re going to work on what Alberto Savoia calls your ‘XYZ’ hypothesis.

If you’re not familiar with it, here’s what each letter stands for:

X: The percentage of your target market that you can engage/capture …
Y: who your target market (or audience) is and …
Z: what action your target market will take in regards to your product/service/idea

Normally when people try to transmute their MEH to an XYZ hypothesis, they immediately get stuck on the ‘X’ part of the equation.

  • Do I choose a random percentage?
  • What’s the default sample size I should use?
  • Can I switch out the percentage with a particular number of people?

All good questions… and there are some options to choose from.

You can try Joe Marasco’s ‘get to ten’ technique that simply aims to get ten qualified sales/commitments from your intended audience while tracking the number of people you have to talk to. The aforementioned Alberto Savoia has a great video explaining the approach.

You can also think of the X as the threshold you need to cross in order to make your product or service worth designing, developing and launching in the first place. What moves the needle for you? If it’s a high number or percentage of commitments (sales, opt-ins, signups, etc.), then your tolerance for failure should be high as well. Conversely, if you’re more risk averse and are willing to accept lower numbers or percentages of engagement to build upon, then micro-experiments with frequent iterations may work for you.

But if you still feel unsure about which way to go, I have an alternative approach that’s much easier to work with. You’ll even be able to build out multiple XYZ hypothesis for even more testing later on!

Here’s how you go about it

First, start with the ‘Z’ and list out all the ‘intent’ transactions you want your audience to take. Your audience can probably interact with your product and service in multiple ways, so why not list them all out? Let’s take the Global Virtual Design Sprint MEH example from yesterday as a starting point.

  • If (I offered an international event where practitioners could collaborate, co-create and network with one another while practicing the design sprint methodology in an online environment)
  • Then (facilitators, design and research professionals would pay $249 USD)
  • To (participate in the Global Virtual Design Sprint to learn how to do virtual design sprints)

If I just take both the ‘Then’ and the ‘To’ part of that MEH and combine them to define “what action my target market will take in regards to my product/service/idea”, then I can start creating different actions they might do.

The first line item (would pay $249 USD to participate in the Global Virtual Design Sprint to learn how to do virtual design sprints) was my original, but the other four alternatives showcase different types of ‘skin in the game’ that my audiences might give for what I’m offering.

Now that I have a list of possible actions my audience might take for my products and services, let’s switch to the ‘Y’, or who our target market (or audience) is. Here’s what that looks like:

If we were to expand upon that list, we might include other types of professionals who might be interested in the Global Virtual Design Sprint

Then, when we combine both our Y and our Z, we have a lot of different ways we can build the latter half of our XYZ hypothesis. It might look something like this.

When you look at this table, you might see some audience types in the ‘Y’ column that could potentially engage with one, two or several items in column ‘Z’. In fact, you could do any number of combinations to set the stage for market engagement. Here’s just a few:

  • Designers, facilitators and researchers would pay $249 USD to participate in the Global Virtual Design Sprint to learn how to do virtual design sprints.
  • Designers would pay $59 USD to attend a 90 minute co-creation session to learn how to conduct a four-step sketch
  • Facilitators would pay $399 USD for a ‘Featured Facilitator’ program to learn both virtual design sprinting and unique approaches to become an advanced practitioner.

Finally, you’ll introduce your ‘X’, or the percentage of your target market that you can engage/capture to quantify your audience engagement. And remember, don’t worry about getting this right. Just put something that would align with the amount of experiment failures (or mismatches, which is what I like to call them) you would be willing to tolerate.

  • 10% of designers, facilitators and researchers would pay $249 USD to participate in the Global Virtual Design Sprint to learn how to do virtual design sprints.
  • 10 Designers would pay $59 USD to attend a 90 minute co-creation session to learn how to conduct a four-step sketch
  • 20% of facilitators would pay $399 USD for a ‘Featured Facilitator’ program to learn both virtual design sprinting and unique approaches to become an advanced practitioner.

Now it’s your turn!

Take the Market Engagement Hypothesis you created in Step 1 and curate both your Y and Z portions of your XYZ hypothesis. Add a number/percentage at the beginning to quantify your audience engagement, and you’re all set for Step 3!

If you need help or have any questions, please ask. Otherwise, we’ll see you next time when we work on our ‘xyz’ hypothesis by narrowing our scope and focus.

Thank you very much for reading the article! 😁

--

--

Robert Skrobe
Dallas Design Sprints

I run Dallas Design Sprints, The Design Sprint Referral Network and Talent Sprints.