Strategy maps
A technique for visualising future scenarios to help us collaborate, understand and make better decisions in line with anticipated value and desired outcomes. A game of options, or “how do we make the best move”. Presented at Agile Tour London 2019.
This may be useful for some lean agile situations and more relevant to Cynefin and complex adaptive systems than existing strategy maps which are little more than to do lists. Several examples of existing “strategy maps” will hopefully make you think about if there is an opportunity for improvement.
There are many organisations that have persisted too long with a strategy before they found out too late it was the wrong one (“the Kodak, Yahoo, Blockbuster, Nokia bucket list”).
We often talk about alignment and autonomy in agile, however alignment without being aligned to the right thing is wasteful. It is also important for organisational outcomes and strategy to be easily understood in context to minimise cargo cult.
Professor Kaku “Intelligence is our ability to map and simulate the future” (link has the exact quote, full video below)
We seem to have a canvas for everything. A canvas is useful but so is context and being able to plan and track across time, whereas a canvas is more of a snapshot view.
If you only wait until you have done an experiment before you consider alternative paths, then you may already have missed the boat. I suggest we should consider alternatives early and use appropriate empirical data to support this but proper analysis of data is necessary in a complex scenario.
Corporate strategy is not a static canvas, it is action as a system. This part of Wardley doctrine is most relevant here, especially the “complex” and “iterative”.
A few potential use cases for this
Strategy maps are a use case of event maps, here’s a few applications
- Delivery map (figuring out how to deliver a product roadmap including risks). Early application of this. Another early application.
- Historical map (working out what happened in the past, e.g. historical interpretations, crime solving).
- Personal improvement map (setting a few life goals as part of a personal OKR).
- Understanding arguments and the origin points in order to visualise the origin and possible solutions. Emotionally embedded arguments need more than facts.
- Complementing 5-whys root cause analysis but doing it from a future perspective rather than analysing the past.
- Knowledge mapping what do I know, what is known, what is knowable, what is unknowable and its relationship with Cynefin domains. See: Liz Keogh “Who’s done this before” (i.e. where does the knowledge exist?).
- General tool to externalise thought processes and the problem in order to facilitate collaboration and common understanding. An external cognitive map. “This is my understanding of this situation, how does it compare to yours?”
(Explanation articles for the above to follow, this is an MVP. Also to do: a shorter summary version of this article and a real example with a company’s strategy)
What’s the point?
- To maximise the probability of achieving an outcome by appropriately considering the advantage of intermediary tactical positions to shift outcome probabilities.
- To get a shared understanding of the journey to make that happen.
- To get a better understanding of where we should go and how to get there.
- To involve the relevant people in regular reflection rather than pursuing a possibly out of date strategy in a changing landscape and to minimise confirmation bias.
We need to do this. We are poor at strategy, brexit has made a laughing stock of the UK, the US has an endemic guns problem that it’s in denial over, people are being bankrupted over US healthcare bills and climate change is a real and dangerous problem we ignored for far too long. In 1990 I campaigned on Channel 4 for more smoke free places in the UK, it took 16 years for common sense to catch up and a law to be passed. That cost lives. Being able to see the future and react effectively to dangerous problems is critical.
We need to do this and it’s coming anyway. Let’s get ready.
We’re also a good 5–10 years away from being able to effectively anticipate change, to use maps to target the landscape and to learning how to manipulate the market in order to write strategy. Simon Wardley: Rebooting GDS
Introduction and context
Mapping as a practice
The use of Wardley maps, and more recently maturity maps are examples of effective visualisation to understand the context of work and strategic decision making.
Agile practices often make use of visual techniques, the first practice of Kanban is “visualise the work”.
Mapping is another example in a long line of visualisation practices which go back a long way, including UML activity diagrams, business process modelling, event chain diagrams and even the early days of flowcharts.
This is strategy, we don’t currently visualise the why of movement effectively.
What is a map
In the introduction to Wardley maps “On being lost”, Simon Wardley gives these criteria. Bold items from Simon, explanations mine
- visual — you can see it and use this to share a common perspective
- context — the map is situational to what you are doing
- position — position on the map has meaning
- anchor — there is a specific direction or orientation of the map
- movement — movement on the map has meaning
- components — different elements of the map have specific meaning
Existing strategy maps are therefore not maps, they are to do lists. Here’s an example, a to do list. No context of movement, no feedback, no outcomes. A list of goals is not a strategy; a target is not a strategy; a long term platform is not a strategic platform, it’s a long term platform; an outcome is not a strategy either, the strategy is how you achieve the goals.
Examples of actual strategy
Battle of Thermopylae
Simon Wardley: https://blog.gardeviance.org/2014/09/if-wardley-mapping-is-so-powerful.html
Strategy: “The great general Themistocles devised a strategy where the narrow pass of Thermopylae would be used to hold back the Persians whilst the Athenian navy would block the straits of Artemisium”.
Target(?): Win 3 battles this year and get a bonus
Goal(?): Win the battle.
Outcome(?): We expand our territory.
Strategy: Block the straits of Artemisium, then fight the battle on land where the landscape will give us maximum advantage.
Robert the Bruce and Bannockburn
Overcome the superior English numbers by taking advantage of the landscape (bog), build traps (pits in the ground with stakes) and use of schiltrons in order that the superior English numbers do not confer an advantage.
“Win the battle” is not a strategy. It is the objective. The strategy has why and how of how the objective will be achieved.
Summary
Let’s move on from PowerPoint goals. We need a why of purpose, a why of movement and an understanding of the interactions which influence outcomes.
Why maps?
Maps are a form of visualisation and appropriate mapping can help us to understand what’s going on. In the same way that a kanban board is a useful visualisation tool to understand the flow of work, so a map is a practice for seeing the landscape and context of the work in order to observe connections, opportunities and where the next move makes the most sense in that context.
This is a map (thanks to Simon Wardley)
If you understand the rules of chess, looking at the above board probably makes more sense than this equivalent:
f3 e5 g4 Qh4 (Chess moves, not a map)
The more complicated the game, the easier it becomes to see what’s going on by looking at the board (the map) than reconstructing the state of play by looking at the move sequence.
Event map introduction
An event map shows interactions over time against outcomes, strategy maps are a type of event map. Intermediary steps provide points for reflection. Whilst event maps can be used by teams, especially Product Owners, like kanban they can be used at all levels of an organisation and are not just confined to product delivery.
This article explains why techniques which get us to think about the future are important and how we can overcome bias.
“Delivery personas” for organisational state.
An event map helps us to visualise intermediate future steps and pathways on the road to achieving a goal. Just as user personas can help us to visualise and make real the people and their needs which we are building for, so the intermediate steps, the delivery states, on an event map can help us better understand the impact, probabilities and pathways on the roadmap. These are scenarios for exploring possible futures — the stones in the “crossing the river by feeling the stones”.
At each intermediary step we reflect and revise. This is covered in the Management 3.0 Change Management game as:
- Consider the system
- Consider the individuals
- Consider the interactions
- Consider the environment
Other feedback considerations include: feedback from customers, feedback from stakeholders, risk, team happiness, team engagement, organisational impact, contribution to long-term goals, complexity, technical debt, opportunity creation, tactical advantage etc.
Adjust accordingly based on feedback. Many roadmaps are too simplistic for this and are often focussed on product rather than change. Organisational strategy is about systems, processes, interactions, shifting landscapes and of course people. Many organisational strategies focus on process, methods and frameworks whereas really they are about people and culture.
Culture > Strategy (not just for breakfast!).
Constituent parts of a strategy map
With two axes to work with and one being time (horizontal axis) there are a few useful things we could put on the vertical axis, such as
- Alignment to outcome. In The Three Laws of Performance (Steve Zaffron) he talks about a predestined default future. This is our destiny if we choose not to act. By acting in line with strategy we can shift outcomes in our favour. Alignment to outcome is about understanding how much our actions make the intended outcome the default and how much do we need to change direction to make it the default.
- Probability of success / risk levels
- Cynefin Domain (often compared to how many people know it)
Map contents
The items which will be useful on the map include (pick what is most suitable for your context)
- Cynefin domain. A map with time against outcomes could have the stages labelled with the cynefin domain expected because excessive experimentation and hypotheses in the complicated space might not be efficient.
- The effect of the actions and intermediate actions including on the people doing the work, the organisation they are working in, customers, competitors.
- Option pathways (avoid the cul-de-sac options or those with limited flexibility)
- Important timings (e.g. last responsible moment)
- Probabilities arising (and numbers of possibilities)
- Secondary effects (of course there is).
- Connected sequences of events.
- Identification of hypotheses, assumptions or empirical data.
Existing strategy maps really don’t do this. Here you go: Existing strategy maps don’t do this. Slideshare? No. Oh go on, more bad examples please! (That’s at least 5 links on “strategy maps”, feel free to research this further…)
Often they are corporate “blahs blahs” or “fake CEO” slide decks showing blobs on a page. No why of purpose, no why of movement, no alternative courses, no meaningful context and often top down with no feedback loops. The “why of movement” is lost when presented like this. Pass these onto people not involved in strategy creation and they lose their context and become fossils.
The journey towards maps
The agile manifesto, evokes a journey
Individuals and interactions over processes and tools
Working software over comprehensive documentation
Customer collaboration over contract negotiation
Responding to change over following a plan
The italic items were what people were doing before the manifesto came along, so the manifesto shows the journey from the old state (italics) to the desired new state (bold items) and priority. Working software, Customer collaboration and Responding to change having more value. It’s a journey showing 4 useful steps.
My journey with this started from product roadmaps in 2018.
Product Roadmaps
When building a product, it helps to have an overall context and purpose for the product, rather than turning up each fortnight to refine your product backlog with no idea of the long term purpose. A suggested approach to this is:
Product Vision: Purpose, ultimate goal. (perhaps 5 year timescale). Aligned with organisational vision.
Product Strategy: Overall approach, path to the vision, product life, cycle stage. Aligned to organisational strategy
Product Roadmap: Actionable plan, product journey. Perhaps around 12 months. Informed by the strategy
Product Backlog: Details of how the roadmap will be built in the upcoming sprints. Prioritise in line with the roadmap. Typically user stories.
(Credit to Roman Pichler from a talk at the BCS Agile Specialist Group, 8th November 2018)
Roman gives these examples for a product roadmap
(neither of these are maps, however they are useful)
I struggled a bit with the bigger picture here and for me this felt too much like a summary canvas rather than a mapped journey in context. I was looking for something that worked alongside this to help us make decisions, weigh up different options and figure out the path most likely to deliver the desired outcomes.
To me, product roadmaps were useful, but lacked context. I felt there was something missing, especially from the change, culture and people perspectives when we know these are critical and central to modern leadership approaches especially Agile and Jurgen Appelo’s Management 3.0.
Mapping Brexit to understand strategy
Using Brexit as an example to illustrate strategy maps because if it can handle something as (chaotic | complex | complicated | disordered | random | dysfunctional ) as Brexit then maybe it’s robust enough to stand a good chance of working in other scenarios. Following this there’s a few principles to help you get going from scratch.
Here’s a map pathway. This is an out of date map, where predicting further ahead had a spread of options, at the time of writing during a Tory leadership election the ability to plan even a few months ahead is much harder.
Brexit seems to be using waterfall practices when agile inspect and adapt early might have been better. If there was a strategy to start with (!) then it’s way out of date.
Applying the Brexit example
Step 1: Visualise the Brexit map above on its side. Time is on the X axis going left to right. Position has meaning, the map shows time mapped against the probability of achieving an outcome with intermediary steps and resultant pathway probabilities for both the outcomes you want and the outcomes that are favourable to the competition (not unlike business, or playing chess).
Step 2: the seven Brexit outcomes above are scored in line with how they align to what you want to achieve. I’ll pick remain as the strategy here (that’s how Scotland voted). Other strategies are available depending on the outcome you seek.
Step 3: Evaluate the pathway options leading to the desired outcomes and avoiding the worst outcomes. Discover the risky points which you need to traverse but which could also be favourable to opponents if they go the wrong way for you
Step 4: discover alternative pathways.
Step 5: using this information you can formulate a context specific strategy like this one:
The strategy on the left evolved from looking at the outcomes (on the right), understanding possible paths and associated risks, then formulating the strategy from this taking into account the risks and multiple pathways. A strategy in context. This will then be updated again (and the map updated) when there is an important change, such as reaching one of the decision points and learning what happens at that point. Each decision point is in effect its own causal loop model.
In the Brexit scenarios above there are seven outcomes. Depending on which outcome you prefer you would view each of the decision points differently in terms of how that decision point supports the likelihood of reaching the outcome you want. Something which is a big risk requires an intervention to see how it can be mitigated. In seeing the appropriate pathways and risks, you can act accordingly. Within a corporate context, seeing changing levels of risk over time can inform appropriate governance rather than one-size-fits-all. Anyone for lean, automated governance?
A suggested approach for getting going with this
We will likely end up with a series of connected events, some diverging and converging again and perhaps different ways of reaching the desired outcomes, like the Brexit map. It is also useful to consider what the challenges are and what the competition is doing which might make reaching the outcomes harder. Our goal is to understand what we do in the present in order to open up favourable outcomes and make them more likely via tactical advantage steps.
- The vision forming phase — what is the future outcome?
- Understand the present — where are we now?
- Plan to a realistic horizon — plan what is useful, too deep or far is waste.
- Connect the present to the future — primary path and alternatives
- Refine the model — evaluate probabilities, risks, impacts on people and the system, who are the actors?
1. Vision Forming
Consider the organisation’s why of purpose. Techniques such as impact mapping, open space [PDF], Agendashift discovery and future search can help here. An envisioning from right to left — what is the end state and what might be the predecessor states which are required or enable this end state to happen. These predecessor states will be useful in step 4 “connect the present to the future”. See also Future Backwards.
2. Understand the present
What are the current organisational structures for work? Is the current work known? There may be gaps, overlaps, work with no clear purpose.
What is the current culture? What are the primary impediments that prevent the future from being realised? Are silos getting in the way? Is there fear in the workplace? Is the technology for collaboration adequate? What skills do people have and are they being used? Developing an improvement backlog can help here. Any change initiative is people led and employee engagement is a key leading indicator how willing company colleagues will be to become involved.
Agendashift exploration can help here.
3. Plan to a realistic horizon
In the Brexit map above, one of the outcomes is “no clear route ahead”. Maybe leave the planning for that for a bit until the landscape is clearer. Too much planning too far in advance, especially in a complex landscape, is wasteful. Detailed enough to be useful and high level enough to be easily understood. In a chaotic or stressful situations and planning might only look a few hours, days or weeks ahead.
4. Connect the present to the future
Map any known future events onto the map. Consider how the present could evolve into the known future events or the predecessor events from step 1.
Useful tactics here include Agendashift’s 15 minute FOTO, Theory of Inventive Problem Solving (TRIZ),
Start with what you might consider the primary path, the backbone and then spread out from there to discover alternatives.
Consider also what competitors might be doing. You might want to be #1 in a sector — what might that entail? What sorts of products/services might that involve? You won’t be the only company wanting to succeed, what are your competitors likely to be doing? How will you overtake them? Are there opportunities on the horizon where they might fail?
In considering experiments along the route, the lean startup approach is useful but remember that in a complex space multiple experiments are usually needed and the results might take a while to become obvious. Many thanks to Dave Snowden for his talk at Agile Scotland (Kevin Austin) hosted by the Royal Bank of Scotland in March 2019:
Genuinely complex work requires multiple experiments, a drug won’t just be trialled on one patient to determine if it is safe for general release. Limited data gives limited insight.
5. Refine the model
Evaluate probabilities, risks, impacts on people and the system, who are the actors? Iterate on model using techniques such as impact mapping, Toyota’s A3 model. The model will probably look like a network map, similar to the brexit model above, reshape the layout of this so that it makes visual sense, perhaps aligning intermediary steps against risk or future outcomes.
Get the right people in the room to understand the situational knowledge. This is how we think some organisations are set up
But this might be more typical of the real lines of communication!
To get things done, it’s important to understand the people of influence and the well connected, rather than the people with impressive job titles.
Envisage right to left, then plan left to right.
Quote from Simon Wardley: “ When you examine a map, you need to go beyond just the landscape, the why of movement (i.e. this choice over that), the why of purpose (to be this or that) and to consider your role and that of others. There are many actors in a map and they have different perspectives. Even the consumer’s view of the landscape can be different from that of the producer. Mapping simply shows you a landscape, you have to apply thought, you have to balance conflicts and you have to strive for your maximum advantage”
There are likely multiple inputs to each situation, (potentially from other maps) and there are secondary effects from each situation. Companies can also play one strategy in public and another in private. Multiple strategies are often useful. Maps can connect with each other into an increasing understanding of the present. An effective predictive model, systems thinking, works for weather forecasting and playing chess, so why not corporate or global outcomes?
A Template
Here’s a suggested template.
Conclusion
I believe strategy maps can help teams understand the why of purpose and the why of movement better. Strategy maps are an evolving item, they need to be updated regularly
Strategy maps can help us to agree a shared perspective of what needs to be done and why in context. Practices such as lean startup are useful but there is more to organisations than product and sometimes experiments take a long time to evaluate, so thinking ahead before the results are known is useful. I have felt that many product backlogs or portfolio kanbans had lots of things to do but not much context on why are we choosing to do it this way here and now.
Strategy isn’t the whole answer; culture is more important but a good strategy will help an organisation to succeed more than a bad strategy.
In scrum we inspect and adapt the evolving product. Here we inspect and adapt our understanding of the evolving landscape.
Please give me feedback. I can also be found on LinkedIn. If you like this, a few claps would be awesome!
Craig Cockburn (“Coburn”)
Enterprise Agile Coach, based in Edinburgh, Scotland
Many thanks to Mike Burrows, Scott Seivwright, Liz Keogh and others for the feedback.
Update: 18th October 2019: I spoke at Agile Tour London 2019 on this topic, the slide deck is available here: https://www.slideshare.net/siliconglen/visualising-a-context-based-agile-strategy-18-october-2019