The X-Matrix Strategy Deployment Model

There is a model for Strategy Deployment that sits behind the X-Matrix that is worth explaining in more detail as a way of understanding why it is designed the way it is, and how to use it. It is built around describing four types of elements – which I call results, strategies, outcomes and tactics – and how they fit together.

Before we start, lets get the George Box aphorism out of the way:

All models are wrong; some models are useful

Results
Results represent the organisational impact you want to have. They are lagging indicators, success or failure only being declared at the end of the journey. They usually reflect the nature of the business and its economics.

The Results are implemented by Strategies

Strategies
Strategies are constraints which guide how you achieve the results. They are enabling, allowing a range of possible solutions (as opposed to governing, limiting to a specific solution). Thus they guide decisions on where to focus attention (and hence also where not to focus attention).

The Strategies lead to Outcomes

Outcomes
Outcomes provide evidence that the strategies are working. They are leading indicators of whether the results can be achieved ahead of time. They describe the capabilities that the organisation requires in order to be successful.

The right Outcomes will generate the successful Results.

Screen Shot 2016-05-25 at 20.47.46

Of course the Strategies don’t directly lead to Outcomes. Some form of action has to take place. Thus the Strategies are actually implemented by Tactics.

Tactics
Tactics are the activities that take place to implement change. They are experiments which test hypothesis on how to achieve the outcomes. The represent the investments in the improvement work that is being done.

Therefore, it is the Tactics that generate the Outcomes and ultimately lead to the Results.

Screen Shot 2016-05-25 at 20.48.11

For change to be successful, there should be a correlation between the various elements in this model (and it should be remembered that correlation is not causation). Each element will have some level of contribution to another. This will range from strong or direct, to weak or indirect, or there may sometimes be none. You could also say that the correlations are Probable, Possible, or Plausible. All together there should be coherence (albeit messy) to the way all the elements fit together.

Screen Shot 2016-05-25 at 20.48.32

By starting with Results, moving on to Strategies and Outcomes and leaving Tactics until last, there is a greater chance that the Tactics chosen are ones which do implement the Strategies and generate the Outcomes. The intent is to avoid premature convergence or retrospective coherence when identifying the Tactics. It is very easy to hastily jump to the wrong conclusions about what the Tactics should be, and then justify them based on the Strategies.

Even if you don’t use the X-Matrix explicitly, understanding this model can be useful for asking questions about change and improvement.

  • What end results are you hoping to achieve?
  • What are your strategies to deliver them?
  • What intermediate outcomes will show you are on the right path?
  • What tactics are you using to move forward?
  • How do all these pieces fit together?

If you can answer these questions, then you should be able to populate an X-Matrix. I will work through an example in an future post.

X-Matrix Simple

Alignment and Autonomy in Strategy Deployment

Following on from my previous What is Strategy Deployment and Dynamics of Strategy Deployment posts, there is a model I like which I think helps to show how the mechanics and the dynamics work together.

In The Art of Action, Stephen Bungay describes how Field Marshall Helmuth von Moltke, Chief of Staff of the Prussian Army for 30 years from 1857, had an important insight regarding Alignment and Autonomy. Previously these two had been viewed as extremes at the end of a single scale. Having high alignment meant having no autonomy because alignment could only be achieved through defining detailed plans which everyone should follow. Equally, high autonomy meant having no alignment because autonomy would result in everyone doing their own thing with no regard for each others actions.

Von Moltke’s insight was that alignment and autonomy are not a single scale requiring a tradeoff between the two ends, but two different axis which can actually reinforce each other. Thus not only is it possible to have both high alignment and high autonomy, but high alignment can enable high autonomy.

Alignment and Autonomy

They key to making this possible is differentiating between intent and action. Alignment is achieved by clearly stating intent centrally, such that autonomy can be achieved by allowing action to be decentralised in support of the intent. This requires mechanisms to both clarify and amplify intent, and enable and encourage local action. Thus using the definition of Strategy Deployment as “any form of organisational improvement in which solutions emerge from the people closest to the problem”, solving the problem is the intent, and the emerged solution is the action.

Using this model we can now describe two mechanisms necessary to make this happen. Alignment can be achieved with the X-Matrix, which enables the conversations about intent and summarises and visualises the results of those conversations. In other words, the X-Matrix shows how results, strategy, outcomes and tactics align and reinforce each other. Autonomy can be achieved through Catchball (Bungay describes the equivalent as back-briefing), which enables the X-Matrix to be passed around the organisation such that everyone can reflect, give feedback, and improve it, helping focus action on meeting the intent.

X-Matrix and Catchball

Viewing Strategy Deployment in this light also highlights a symmetry with the Autonomy, Mastery and Purpose model of intrinsic motivation described by Dan Pink in his popular book Drive. Autonomy is a direct match in both models and purpose is equivalent to intent. Mastery is then the result of improving capability autonomously with strong alignment to intent.

Drive

What this way of looking at Strategy Deployment shows is that both the X-Matrix and Catchball are necessary components. Just using the X-Matrix with out Catchball will probably result in it being used as just another top-down document to command and control employees. Similarly, just using Catchball without an X-Matrix will probably result in collaboration around local improvements with no overall organisational improvement.