How do I know if it’s right?

A mouse with colorations and stripings of a tiger. Underneath it, a quote: "Worrying Selectively. Since all models are wrong the scientist must be alert to what is importantly wrong. It is inappropriate to be con- cerned about mice when there are tigers abroad."
A different George Box quote than you're used to hearing.

In a welcome first for us, we have a response to our last edition! Yoga writes:

I have in my head a model that Wardley mapping is useful only after you see how it can help you. And for that, I think you only realize it after it’s been sold to you in a very particular way. One where you generate the insight that it can be helpful.
If you see one map, you may not understand why it’s useful. Two maps? Three? How many do I need to see?
I think that getting the gist of mapping really only takes shape if you’ve seen a few examples that you can relate to. And most importantly, if you can see the pattern that emerges.
I think it’s difficult to see the pattern that emerges of you’re stuck looking at one specific map. You probably need to see 6 of them – and see them animated – to see what’s going on.
Then you might have an insightful moment where you realize that certain states repel or deplete actions, certain states multiply actions… And the natural next question that people will ask themselves is, how do I get to that multiplicative state for my problem domain?
Until you reach that insight, it’s just someone rambling about a flowchart. (Which is hard to hear, when you have the insight but the other person doesn’t)
– Yoga Naraine

Thank you, Yoga! Your response read like poetry to me.


Up next, “Newbie mapper” writes in:

I’ve read a bit about mapping and am struggling with my first maps. One problem I’m encountering is that I don’t know if they are correct or any good. How do I best validate that?

Sometimes people ask me to look at their maps.

Then they ask me, “Is it right?”

I’ll let you in on my poorly-kept secret: I have no idea. 🤷‍♂️

But here’s what I can do.

I can check for structure.

“Users, needs, capabilities, depends-on relationships, evolution… Yep, that looks… mappish!”

I can also listen to how you talk about it, to see if what I see in the map matches what I hear as you explain the situation or share your argument about what to do.

“Ah yes, that’s the sort of thing I was expecting you to say about a map like this.”

And I can ask basic questions to test your map.

“What did you mean by that word?” (Language)
“What else does that thing depend on?” (Relationships)
“Does that thing fail a lot, or is it super reliable?” (Evolution)

But in the end, there’s very little I can do to confirm it’s “right.”

Partly, that’s because I don’t have the expertise to know. I haven’t been doing your job as long as you have! Only you (and others like you) will be able to know.

“Right” isn’t a good goal for a map.

I know, I know. You’ve heard it before. Maps need to be “useful” more than they need to be “right.” I won’t bore you with all that again.

And our newbie mapper still has a question that needs answering. I don’t think saying, “make your map useful, not right” is going to help.

So let me modify the question, in the hopes that I can give a better answer:

How can we make sure a Wardley Map is a good enough representation of the situation?

For this, I have a few things to try.

#1: Rubber ducking

A method useful for debugging software, rubber ducking is also great for Wardley Maps.

Just find a rubber duck (or other suitably inanimate object) and explain your map to it, bit by bit.

Just talk out loud, hear what you say, and confirm that it makes sense.

You’ll catch all sorts of surprisingly obvious mistakes this way.

(And if anyone asks, you’re not talking to yourself. You’re rubber ducking! It’s completely different!)

#2: Write it out in words and share

Once your rubber duck has gotten tired of listening, try sharing with a human.

First, write down the explanation you verbalized for the rubber duck. Check that it makes sense.

Then, share it with someone who is glad to tell you what you get wrong. (You don’t have to share the map, just what you’ve written.)

See what commentary they offer, and adjust your map accordingly!

#3: Define your terms

A slight extension on writing it down, try to create a glossary that defines all the components on your map.

Share the glossary (but again not the map) with others for feedback, perhaps one term at a time. (Less is more here.)

Expect to get contradicting feedback from different people, like one word meaning many different things, or many different words that refer to the same thing.

(Reality is still super messy like that, regardless of whether you manage to make a useful map of it.)

Revise your map as you learn from the feedback. Make bold decisions about what to do in the face of contradiction.

Evaluating your map’s stability

A map is a piece of knowledge (or perhaps data, depending on how you look at it) so that means you can also evaluate how evolved it is. (Yes, you can use Wardley Mapping on your Wardley Map!)

Any evolving thing has an applicable market — a scope within which we can review the thing’s evolutionary characteristics. In this case, at least at first, that applicable market is just you!

Per the table in the link above, Knowledge progresses through the following labels, stage by stage:

  1. Concept
  2. Hypothesis
  3. Theory
  4. Universally Accepted

And meanwhile, Data progresses through the following labels, too:

  1. Unmodelled
  2. Divergent
  3. Convergent
  4. Modelled

Given these ways of describing each of the four stages, you might be able to see where your map fits.

It might help to ask, “How often do you learn things that cause your map to change?”

If the answer is “all the time,” then it’s unstable and early in its evolution (stages 1-2).

But if it hasn’t changed in years, and you still think it’s representative of the situation, then it’s probably more highly evolved at that point (stages 3-4). So, maybe it’s “universally accepted” (if that universe only has one person in it… you! 😜).

From there, you might decide to reduce bias (hello, doctrine!) by expanding your applicable market beyond yourself to include other players, comparing your assumptions to theirs to see how they stack up. (This may change what stage you think your map is in, and that’s normal.)

I’m not sure that was very clear. Did it make sense? Hit reply to let me know if it didn’t!

It gets easier with experience

You’ll never truly know if your map is “right,” but with time and practice you’ll certainly get a better sense for it.

Remember that we’re just aiming for our maps to be “not wrong” enough to use them to do some good thinking.

Above all, keep an open mind, and be ready to change your map when you find out you got something wrong. Tigers, not mice.

A mouse with colorations and stripings of a tiger. Underneath it, a quote: "Worrying Selectively.Since all models are wrong the scientist must be alert to what is importantly wrong. It is inappropriate to be con-
cerned about mice when there are tigers abroad."
A different George Box quote than you’re used to hearing.

Leave a Reply