Strategy Management’s Chicken or the Egg Dilemma

What comes first – the strategy map or the validated cause and effect model?

As you may know, the strategy map and balanced scorecard approaches to business strategy and performance management provide you with the ability to map out AND demonstrate the cause and effect relationships that exist in your organization. I call this cause and effect model the “theory of how your business works”.

The theory behind it all is that once you map out the cause and effect relationships in a value stream, you are in a better position to work these relationships in a targeted way to produce desirable business results. One of the things that is so powerful about a cause and effect model is that it shows how focusing on objectives lower down in the strategy map/balanced scorecard hierarchy (i.e. the learning and growth or organizational capabilities perspective) translates into business results. When we can do it, this is a good thing because, though we intuitively know that interventions in the more intangible parts of our business (i.e. our people investments) translate into better business results, it can be hard to demonstrate the ROI (return on investment). Cause and effect models theoretically (and, in some cases, actually) help us do this.

There is some debate about the best way to realize the benefits of the strategy map’s cause and effect model. Some argue that you should complete a comprehensive data mining/business intelligence exercise to identify the cause and effect relationships in your business and then use this information to create a validated strategy map. Others take the opposite route and create the strategy map, identify the presumed caused and effect relationships between strategic objectives, and then collect information (i.e. data, commentary, and learnings) to test and, perhaps, validate the causal model.

So, which should come first – the proven cause and effect model or the strategy map?

My preference is to begin with the strategy map – here’s why:

Remember what you learned in stats class

I don’t remember much from my university statistics classes but what I do recall is that is takes a lot of data to prove causation for anything – including cause and effect relationships within your business.

The cause and effect modeling “poster child” is Sears. In a 1998 article in the Harvard Business Review called The Employee-Customer-Profit Chain at Sears, authors Rucci, Kirn, and Quinn presented a pretty tight cause and effect model validating that a significant relationship existed between employee attitude, customer impression, and revenue growth at Sears. In fact, such a high level of precision was created regarding the causal nature of these relationships that Sears knew that a 5% improvement in employee attitudes resulted in a 1.3% increase in customer impressions and satisfaction, and a 0.5% improvement in revenue growth. It was causal model nirvana!

As you know, Sears is a big company (it was probably a lot bigger back in 1998) and being a big company with lots of stores doing the same thing means lots of data points. I’m not sure how long it took for Sears to collect the data they needed to validate their causal model but I do know that it would have taken significant amounts of data. Collecting this much data might not be that hard for a huge corporation but for the rest of us, it would probably take a very long time (years) to collect enough data to PROVE the cause and effect relationships in our business. No company should/can afford to endure years of data collection before creating their strategy map. For that reason I say just get on with it – get started creating a strategy map for your company NOW! And besides, how will you know whether you are collecting the right data to prove causality if you haven’t defined your strategy map first?

Even if you have the data, looking backward may not help you move forward

Many organizations have CHANGE in mind when they decide to develop their strategy map. It may be a change in their customer value proposition or a different focus on the way work gets done to produce results. If changing the way your organization does business is at the foundation of your strategy map, different aspects of your business will naturally become important and, as a result, it’s quite probable that you will end up measuring different things to tell you how your organization is performing. Even if you have lots of data available in your organization (often not the case in many organizations), you may not have data that is relevant to the new focus of, and cause and effect relationships within, your business. Knowing this, I wouldn’t plan to use existing data as the basis for creating your new strategy map. I say just take the plunge because…

Your people are wise and knowledgeable

I will bet that your organization is full of people who know a lot about how your business, industry, and competitors work/could work. They know your customers and what they need and expect from you now (and in the future). They know how the key parts of your organization work and how they could work in a more integrated way to produce better results. In fact, knowledge about the potential cause and effect relationships in your business is resident in your company right now – all you need to do is tap into it! Why not get a cross-functional team of people together to discuss your company’s change agenda and value proposition, and then define the value-creating cause and effect model (aka your strategy map) the group believes will produce the business results you want. While it’s not a hard-core, data driven approach, I would argue that creating your strategy map this way is still fact-based – not a “shot in the dark”.

I appreciate data as much as anyone but what I value more is information (a combination of facts, analysis, knowledge and understanding, and learnings and wisdom). I believe that the strategy map should come first in the strategy management process and should be built on collective organizational information about the projected value-creating causal model that will produce the performance results your organization wants. Once you have an aligned balanced scorecard indicator set in place, you can begin to collect information (including data) that can help you see whether your cause and effect model looks right/works.

If you are a really big company, you may be able to collect enough data points to validate your causal model. But the rest of us shouldn’t be discouraged because I don’t think that proving the cause and effect model on your strategy map is the REAL objective of strategy map use anyway.

Ultimately, the strategy map plays a powerful role in (1) getting everyone in your organization on the same page about how you think your business works in an integrated way to produce results and then (2) allowing you to learn TOGETHER how it really works and where to take actions that are likely to produce results.

Many of the strategy execution problems I see in business happen because people within the organization don’t have the same picture of what the business is trying to accomplish and how it will get there. Essentially, people end up working at cross purposes while they genuinely try to implement what they think the business strategy is. The strategy map can clear this problem up almost immediately – it’s really that simple.

For this reason alone, I say create your strategy map first and then let small “v” “validation” of your cause and effect model be part of the conversation you have across your organization as you use your strategy map to execute your strategy and fine-tune the way your organization works to produce better results.


  1. Phil Roach
    Feb 3, 2011

    Encouraging words Sandy, thank you! Feel free to add more thoughts on value generation in the public sector and how strategy maps are implemented in a mission-driven organization.

  2. Sandy Richardson
    Feb 4, 2011

    Hi Phil – glad you found the post helpful! I will write about this topic in my next blog. Thanks for the great suggestion. Sandy

  3. Jesús Sánchez
    Feb 4, 2011

    If the purpose is NOT TO DEVELOP THE STRATEGY MAP, you can follow the second approach. You can fall in the “analysis paralysis” trap with the second approach. Consider, for example, that you will not have all the information (data) for new indicators… then you can´t do data mining.
    First, you develop the Strategy Map. Then, with data and “strategic dialogue”, you can validate cause and effect relationships (strategic hypothesis).
    We have used data mining and dynamic simulations (with IThink software) to validate deep cause and effect relationships between variables (indicators). But this is a step that we have implemented after two or three years using the BSC.
    Best regards, JS

  4. Sandy Richardson
    Feb 4, 2011

    Thanks Jesus – well said! Sandy

  5. Project Management Company
    May 1, 2011

    Nice article. Kudos for sharing!

  6. Sandy Richardson
    May 1, 2011

    Thanks – glad you liked it!