How healthy is your decision supply chain?
We’ve all heard or uttered some variation of the phrase, “It’s not about the data, but the decisions you make with the data.” But still, so much of our emphasis seems to be on coping with data, not on converting that data into better business decisions.
One construct that helps maintain focus on the ultimate prize – better decision making – is the decision supply chain. At Mu Sigma, we use this artifact extensively in our work with clients.
A short primer: How decisions get made
In a business context, the impetus for a decision begins with a trigger. That event leads us to define a business problem, using questions and hypotheses, then to gather data, from which we synthesize findings, from which we strive to generate actionable insights. We then evangelize those insights, and once communicated, make decisions and take actions, ultimately measuring and reporting on their success. That’s a decision supply chain (Figure 1) and each day thousands of such chains operate continuously and simultaneously in large enterprises – whether we recognize them or not.
Decision supply chains are similar to their physical brethren. There’s a need to retrieve raw material (data), respond to demand fluctuations, distribute insights across an organization, all while seeking to continuously improve the flow of the chain.
But in light of the volume of signals and noise bombarding businesses today, I’d argue that the decision supply chain now trumps the physical in terms of potential business impact. Just like we aspire to optimize physical supply chains we must optimize decision supply chains, and that requires us to recognize them and manage them as formal constructs. Doing so can help accelerate your organization’s journey from data- or analytics-orientation to true decision sciences.
Decision supply chains help diagnose analytics illnesses
When we first helped apply the decision supply chain model in a client environment, it became readily apparent where data and analytics initiatives were exposed. That alone justified our use of the framework. In fact, by mapping out all of the steps in the decision supply chain, you’re likely to uncover suboptimal situations like these:
- A creation bias. A large retailer used data analytics services to come up with a sound CRM framework. They procured the right customer behavior data, engaging consultants to mine the data and build predictive models. But they didn’t align their models with the campaign technology that would ultimately connect to customers. As a result, their investment languished, their credibility suffered, and their budget was cut.
A proper decision supply chain would have forced the team to give early consideration to how the insights would be used and consumed downstream. - Analytics islands. A corporate finance team led a project to uncover the biggest drivers of risk in their 3 year strategic plan. What they didn’t realize is that the sales operations team had already spent more than a year surfacing similar metrics, by region, to inform quota setting and de-risk their annual plans. The two work streams could have been integrated to the benefit of both.
A proper decision supply chain would have identified early on the interdependencies with other decision supply chains in the organization. - Metrics massaging. A CEO is passionate about her new dashboard, and her weekly operating meetings commence with a review of its contents. The initial design effort involved her executive team selecting the right metrics, but their subsequent involvement deteriorated into weekly scrambles to validate or dispute the data, then massage the stories around those numbers to suit their agendas.
A proper decision supply chain would have enforced transparency in the production phases of the supply chain, minimizing the risk of disputes later on.
You’ve likely experienced these and other illnesses, perhaps without realizing that a decision supply chain construct could have helped avoid them.
Five attributes of a healthy decision supply chain
Assuming you’re game to set up a decision supply chain, here’s a starting point for ensuring that it’s a healthy one. A healthy decision supply chain will be:
1. A harmony of consumption and creation. When groups create new big data analytics they need to obsess over how those analytics will be consumed. Unfortunately, most organizations overweight to the creation side of the ledger (Figure 2.) You should aim to industrialize your creation activities, but because real differentiation will come from consumption, be smart and greedy about the right-hand side of the figure.
2. Fully visible. Every problem area should have a plan that spans all 9 links in the chain. But just as important, you need to enforce transparency across the nodes. For instance, rather than a black-box approach to the “Perform Analysis” step, open up the approach so that business partners (your consumers) understand the “How” behind the “Why” and “What” of your work. This will add credibility to the chain.
3. Interdependent. Business intelligence and analytics is complex because the signals, expectations, and perceptions aren’t just stored in one individual or one job function, but spread across multiple stakeholders and systems in the organization. Making a business decision is not a localized activity, but rather a set of many interconnected chains – what we call decision markets (see Figure 3.) One idea is to take a page from Journey Mapping methods used in Customer Experience design, and apply those same principles to the construction of your decision supply chains.
4. Robust. When you’re repairing a node in a decision supply chain — e.g., the reporting environment — does the chain fall apart? The best decision supply chains have backup processes and methods available at each node. That’s one flavor of robustness, perhaps more akin to resiliency. But a truly robust decision supply also flexes and adapts quickly to changes in the business environment. Walmart’s much publicized response to Hurricane Katrina – telling its employees to simply “do the right thing” – was an example of a company with a rigorous decision supply chain, who changed its rules quickly in the face of a major event.
5.Efficient. Results measurement is already a key link in the supply chain. But the most successful analytics groups will also measure the cycle times of their decision supply chains: The time it takes from seeing a trigger to taking action, and each step in between. Doing so helps identify decision “stock outs”, points where leaders either lack what they need, or are delayed because they seek too much consensus.
And as a counterweight to minimizing decision stock outs, identify points where you have high inventory holding costs in the forms of stagnant or unused insights. Flooding a decision supply chain with too many artifacts or competing insights will slow it down, diminish its business impact, and hurt its credibility.