How has the recent turmoil within the OpenAI offices changed your plans to use GPT in a business process or product in 2024?
Increased uncertainty means we are more likely to evaluate alternative AI chatbots and LLMs.
No change in plans, though we will keep an eye on the situation.
With Sam Altman back in charge, we are more likely to go all-in with GPT and LLMs.
What recent turmoil?
DevOps / Tech Life

Simon Wardley on Mapping Our Way to a Common Language

We talk to Simon Wardley on the power of Wardley mapping to find the landscape and to talk in a common language.
Aug 15th, 2021 10:00pm by
Featued image for: Simon Wardley on Mapping Our Way to a Common Language
Feature Image par Darkmoon_Art de Pixabay

Simon Wardley is a researcher who covers technology and business culture at the Leading Edge Forum, where it’s his job to spot signals of change. He is most known for his application of “Wardley Maps” to business and global problems alike.

We talk to the Brit about just what maps really are — versus their imposters: graphs — and why they are being applied to complex issues ranging from distributed systems in the cloud to billions in cost savings to the climate emergency to each of the United Nations Sustainable Development Goals.

Maps very efficiently center us all in one direction and clarify the nuances of meaning and language, so we can unlock patterns. And maybe they also put some checks on that very human habit of storytelling. We talked to Wardley to find out more.

The New Stack: How did you go about developing Wardley maps and how has that become a movement?

Simon Wardley: I used to run a company called Fotango. We had brilliant engineers and everything else, but the executive in me was completely clueless. And I realized over time that part of the problem was that I couldn’t actually perceive my landscape at all.

If you go back to Sun Tzu’s “The Art of War,” you have purpose, moral imperative, you need to understand your landscape, and you even need to understand how it’s changing, the rules of the game, climatic patterns of the heavens. Then you get into principles of organization, doctrine, and then you get into leadership.

Sun Tzu talked of five factors that mattered:

  1. Purpose / Moral imperative
  2. Landscape (i.e. situational awareness)
  3. Climate (i.e. the patterns of change, rules of the game)
  4. Doctrine (i.e. the principles of organisation)
  5. Leadership (i.e. gameplay)

The most basic thing is: understand your landscape. I didn’t understand my landscape. I had no maps. I had loads of graphs. Most things in business which we call maps are in fact, graphs. They’re not actually maps. So I had to create a way of mapping. And I used that from 2005 onwards.

Then I ended up running strategy for a company called Canonical, with a friend of mine Mark Shuttleworth, CEO at Ubuntu [publisher of open source operating system on Linux]. I used that to map out the cloud space, and we went from 2-3% of the operating system market to 70% of cloud in 18 months. It cost us about half a million.

Then I wrote something with others called “The Better for Less” paper [subtitle: “How to make Government IT deliver savings”] which led to transformation in the UK government, helped create something called spend control and helped support something called GDS [UK Government Digital Services, when all online public services went “digital by default” in 2011.]

It was about that time I started discovering in 2011 that actually people didn’t learn how to map at MBAs. They didn’t have maps. And it was like, well, how are you making decisions? It turned out most of it was luck or copying others — 67% of generals were bombing a hill, so bomb a hill. Or 67% of generals are doing AI or big data or blockchain — “Oh look, they’re doing that, we should do that.”

So it just grew from there. It turns out that looking at your environment is quite a useful thing to do.

Why a map and not a graph? What’s the difference? [Note: Here Wardley starts screen sharing many maps, which will be sprinkled throughout.]

People like to talk about maps all the time because maps invoke this idea of strategy. But actually what we mostly have are graphs.

three identical graphs that don't change until become maps with a due north

These three images at the top are all graphs, and they are all identical. So it’s three different places — Nottingham, London, Dover, connected by two roads M1 and M2. That’s not specifically accurate, but they are three identical graphs, while the three images at the bottom are all completely different. And that’s because they’re maps. And the fundamental difference between a graph and a map is that in a map space has meaning. If you move a piece on a map, it changes the meaning.

It’s a bit like a geographical map. If I move, say, Australia, and put it next to England, that obviously changes the context, the meaning of the map. But that’s not true with a lot of the stuff we call maps in business.

A systems map

Let’s move the CRM.

The exact same systems map

How’s the map changed? It hasn’t. Because it’s not a map. It’s a graph. And so pretty much everything we call a map in business — value stream maps, systems maps, mind maps — they’re all great, really useful, but none of them are maps. They’re all graphs.

If you’re trying to explore a landscape and learn patterns from it, it’s really useful to have a map. But our maps are not maps, what we have are graphs. So it becomes really difficult to learn patterns and anything else.

“In a map, space has meaning.”

Simon Wardley

If you want to create a map, you need three basic characteristics:

  • The anchor or magnetic north
  • Position of pieces (ie: North, South, East, West and distance)
  • Consistency of movement

Position visibility X and Movement evolution Y with the anchors being business and the public, going to the cup of tea which needs a cup and tea, which then hot water needs both cold water and a kettle and the power for the kettle

This is where I started back in 2005. I wanted to map a cup of tea. I’m going to choose two — the public who consumes tea and the business selling tea. They are my anchors. Public needs a cup of tea and business needs to sell cups of tea.

Now, what does the cup of tea need? It needs tea, cup, hot water, and hot water needs a kettle and a kettle needs cold water. So what I’ve got is a chain of needs, and the further I go down that chain, the less visible things can become to me. So if I’m a member of the public and I’m buying my cup of tea, the power that heats the kettle is far removed — so that gives you the idea of distance. So now I’ve got anchor and position.

And I need movement. It turns out that all of these things are actually evolving forms of capital. You can map physical things, you can map data practices, even things like ethical values, but these are all forms of capital. And, it turns out, there’s a common pattern by which capital evolves.  And this is a map.

Adding cost increases if you change staff to robots, and if you use standard kettles

Understand the details: “Why is that there? I don’t know why that’s there.” It’s because somebody else can see you’re missing something.

“You’re missing staff.” “We don’t want staff. We want robots.”

“Why aren’t we using standard kettles?” “We’re using brand exclusively.”

And then somebody else can put metrics on there, so we can do profits and losses. But the point is, it doesn’t matter whether you’re marketing, you’re business, you’re finance, you’re engineering, we have a common language that we can talk about the same environment.

A process flow graph, very linear: Compute Need > order Server, then goods into server > modify > mount > rack, adding a robot between server and modify, which creates a bottleneck for compute needed.

This is about ten years old. This is a process flow for an insurance company that they had. They had a bottleneck around the modification of servers, and so they came up with a plan to invest in robotics — at about 10 million or thereabouts in capital investment — to invest in robots to get rid of this bottleneck to improve their process play. Sounds sensible. It’s been six months working on this, seen all the vendors’ RFPs, they got their ROI, calculations, everything else.

But now they asked me to have a look at it. And the thing is, I can’t “Why are you going to invest in robots?” because they’d built up a big story about why they were going to do this. And the problem with stories is — as we tell everybody — to be a good leader, you have to be a great storyteller. I.e. your idea didn’t succeed because you sold it in the wrong way. Which means when you challenge nobody’s story, you’re actually saying “You’re a bad storyteller.” You’re challenging them, which means it becomes highly political really quickly. So I can’t turn up and say ‘Why are you doing robots’ because defenses go up.

So I said, “Can you just quickly map this?”

A Wardley map of same info that places modify and rack in custom-built; the mounting, user, and compute in the product, but then order server and goods in under commodity/utility

So I said to them, “Why do have rack and custom-built servers?” “Well, we had this company build our own racks.”

“So what are the medications you’re making to servers?” “Well they don’t fit our racks, so we have to take cases off, drill new holes, add new plates in order to get them to fit our racks. And that’s where you need robotics.”

And somebody in the room went, “Why aren’t we using standard racks?”

Now, the problem here is that these people aren’t daft. They’re very, very smart. But they’re trapped by the stories, that context they live in, because they can’t see the environment. And so what they’re doing is optimizing process flow, which would lead you down “Let’s spend 10 million on robotics,” but actually what they want to do is focus first on evolutionary flow: The thing is evolved. It’s no longer custom-built, it’s more of a commodity.

A Wardley Map showing the evolutionary flow that the servers should become a commodity

We’ve seen loads of people getting this wrong. Like cloud, for example, the shift of computer from product to utility. It’s amazing the number of people who decided to build their own.

You just learn basic partners from this. And it’s always quite spectacular. In government, we reckon it’s going to be about £1.5 billion [~US$2.07 billion] in savings. And all it does is come from using a map.

What’s the wildest use case that’s come out of your mapping community?

One was combating illegal fishing and slavery. One is looking at global poverty, with the U.N. It goes all the way down to the individual national statistics organizations because things like all-weather roads really matter. All-weather roads we use for transformation from people to wherever they’re working, so we need to know where all the all-weather roads are. Each different country has their own national statistics organizations collecting information on this but they all do it in completely custom ways which is fairly normal. In fact, the U.N. book on their data platform is all just full of maps as well.

The RNLI [Royal National Lifeboat Institution charity], there was a bit of work done by James Finley where he had a reduced communication time. Maps helped him there, so that saves lives.

There was the carbon map with the NASA satellite [to pinpoint methane and carbon dioxide emitters.]

It’s lovely to hear that people find them useful in these different areas. But the main thing about maps is that they are imperfect representations of the space — even geographical maps are imperfect. If you wanted a perfect map, say of France, it would have to be a one-to-one scale, which means it would be the size of France, which means it would be France, and therefore it will be useless.

They’re all imperfect. And they’re all models. This one’s based on evolution — all models are wrong. They’re imperfect and wrong, but they tend to be quite useful for getting us to challenge things and discuss things as well.

Let’s talk about the Map Camp community, which started from a tweet in 2017. The next Map Camp will be Oct. 13, 2021. How has the community formed and grown?

I made it all Creative Commons [open sourced] — there’s about 600 pages of book there. There’s now an entire community. We now have something called Map Camp. Last year 1,300 people from all over the world turned up. [Wardley mapping] is now being used in places all over the world from NASA to Amazon’s book “Reaching Cloud Velocity” has about 15-17 pages of maps, there’s about 13 different books on it now. Now it’s just spread. There are entire workbooks on Wardley Mapping and training courses.

People are teaching each other how to do this sort of stuff, which is great.

A lot of this rings true with Behavior-Driven Development (BDD) and Domain-Driven Design (DDD). How much is anchored to the user behavior?

Mapping is used with BDD and DDD too.

There are 40 basic principles of mapping [also called Doctrine Patterns], beginning with:

  1. Have a common language — the map
  2. Have focus on the user. (Anchor #1 at the top.)
  3. You need to focus on the user need. (Anchor #2 at the top)
  4. Understand the details – like the value chain
  5. Understand what’s being considered, how evolved the components are
  6. Challenge assumptions — it can be imperfect [and mapping feels neutral or external to storytelling]
  7. Remove duplication and bias — custom build stuff, while long became a commodity [most legacy software.]…

People think duplication is bad in government. The worse example in government I’ve seen is about 118 workflow systems. The worst I’ve seen in the private sector is, well, we stopped counting at a particular bank which had built risk management systems over 1,000 times. And it was constantly complaining it couldn’t innovate.

And somebody said “How can we survive against others if we’re that bad?” Well, the reason is the others were just as bad. What you’ve got to remember about competition is that it’s OK to use the wrong methods, to not understand user needs, to get everything wrong, as long as you’re competing against people who do exactly the same thing. It’s fine to be useless as long as everybody else is useless.

The only problem is where somebody comes into your space who isn’t quite like that and you’re in all sorts of trouble.

So, the disruptors?

You say disruptors, but Amazon is a company which focuses mostly on industrialization. They shoot from product to more commodity-utility. And it takes on industry after industry.

When we talk about disruption, there are actually two different types — which you can’t see if you don’t map:

  • Predictable disruptors: product-to-utility substitution, highly predictable in terms of what’s going to happen roughly when, like old hardware servers vendors selling you servers versus utility compute with Amazon
  • Non-predictable disruptors: product-product substitution, like Apple versus Nokia

With this movement towards psychological safety and diversity, equity and inclusion, is the user ever the employee, as the focus and the anchor of the company?

You can put the employees, or whatever you want as the anchor. All maps have perspective and that will change according to what you set as the anchors at the top.

A map of the interconnectivity of the 40 Wardley principles when applied to the emerging, convergent and accepted principles of humanity: Phase I - Stop Self Harm 1.1 Know your users 1.2 Challenge assumptions 1.3 Use appropriate methods 1.4 Use a systematic mechanism of learning 1.5 Focus on user needs 1.6 Focus on high situational awareness 1.7 Remove bias and duplication 1.8 Use a common language 1.9 Think small 2 Phase II - Becoming More Context Aware 2.1 Be transparent 2.2 Focus on the outcome, not a contract 2.3 Think aptitude and attitude 2.4 Strategy is iterative, not linear 2.5 Move fast 2.6 Use appropriate tools 2.7 Think small 2.8 A bias towards action 2.9 Be pragmatic 2.10 Manage inertia 2.11 Use standards where appropriate 2.12 Distribute power and decision making 2.13 Think Fast, Inexpensive, Restrained, and Elegant (FIRE) 2.14 Effectiveness over efficiency 2.15 Manage failure 3 Phase III - Better for Less 3.1 Provide purpose, mastery & autonomy 3.2 Do better with less 3.3 Be the owner 3.4 Set exceptional standards 3.5 Optimise flow 3.6 Strategy is complex 3.7 Commit to the direction, be adaptive along the path 3.8 A bias towards the new 3.9 Be humble 3.10 Think big 3.11 Seek the best 4 Phase IV - Continuously Evolving 4.1 Exploit the landscape 4.2 Design for constant evolution 4.3 There is no core 4.4 Listen to your ecosystems

You can take any of the labels from Stages One through Four, and apply to the evolution of physical things, of data practices and of knowledge. Including a very imperfect map of culture.

So anthropologists are the experts on culture. And they spent 140 years and they can’t define what that [culture] is.

Margaret Mead talked about how language was part of culture. And this is a really important concept because of mathematics’ Gödel’s Incompleteness Theorem. If something is part of the model, you cannot describe it. If language is part of culture, you will not be able to describe culture with language alone. And that started me down a journey of mapping out values and ethical values and systems, looking at things like universal basic income, workers’ rights, civil rights, etcetera. And these were values or beliefs and I expanded out from that.

So it’s interesting when you talk about safety and psychological safety because a lot of psychological safety comes from memories within the organization, stories, symbols, rituals, heroes, and it matters in doctrine.

You can only do [mapping principle] “challenge assumptions” effectively if there is psychological safety within an organization, and that depends on the memories of people within that organization. And that’s connected up to values.

If we want to be competitive as a society, we need people to challenge assumptions, which means we need people to feel included and psychologically safe to do so.

Simon Wardley smiling net to an illustration of the Trolley Problem -- hit one or five people? -- with the text "Answer should vary with the context of the collective."

If you want to go on a tangent on why “Agile is a Cult” with Simon Wardley, he will join the keynote stage along with Esther Derby and Gene Kim at the Ninth Annual Agile Tour London, live and online Oct. 21 to 22, 2021. The author of this piece is a co-organizer of this event.

Group Created with Sketch.
TNS owner Insight Partners is an investor in: Velocity, The New Stack.
THE NEW STACK UPDATE A newsletter digest of the week’s most important stories & analyses.