Over the last twenty-plus years working with executives worldwide, we observed how one of the most crucial steps in educating leaders on the benefits and use of Decision Management Platforms is to help them navigate what’s possible and what is still esoteric.
To an untrained eye, a decision is a simple process of gathering data, turning it into insight, and coming up with a defensible hypothesis. However, as soon as we embark on a Decision Management discovery journey, leaders quickly learn how not all decisions are created equal, and the technology-processes mix differs dramatically.
This article is about our portfolio company LithiumAI.
While our Decision Readiness Maturity Model comprises many steps, it starts with two simple questions.
Question 1: Do you know the logic behind your decision, and do you need to find it out?
Question 2: Can you measure the outcome of what you want to accomplish?
For the first question, examples of clear logic are processing FS&I contracts in supply chain management or administration in healthcare. The common thread is that their components are an “if-then-else” and need to execute quickly; they are very repetitive and occur multiple times a day.
By way of example, “if” a Decision Platform recognized that, due to a product defect, a shipment was delayed, “then” the platform might suggest to two other trucks to stay one more day in Customs, saving the company thousands of dollars in taxes. At the same time, the platform would inform the customer of the delay offering alternative solutions based on the business rules set by the company.
Decision technology, in this specific case, is used to link platforms that previously wouldn’t talk to each other, set parameters to activate and trigger decisions, and be plugged into enterprise systems, like an ERP, to execute those approvals and decisions.
On the other end, when we are unsure why some things are happening, such as why Customers Churn, a Decision Management Platform relies on a slightly different set of technologies such as AI, specifically machine learning, that can highlight behaviors otherwise invisible to a naked eye.
In the following example, a well-known subscription-based entertainment service improved its customer retention by 4% – leveraging a piece of insight from a machine learning model – the company then added an email subscription reminder to their sales funnel instead of a phone call.
For the second question – setting a clear, measurable goal – is crucial. Can you measure the outcome of your decision?
A Decision Management System is inherently allergic to statements like “let’s plug the Decision Management Platform into our data lake and see what it finds out,” and leaders will find the result of such exercise frustrating and unedifying.
Setting up a goal such as “Reduce churn by X%, “Optimize fuel consumption by Y%,” or “Decide the most effective discount rate and product recommendation mix to increase T-shirts sales by 10%” is what a Decision Management Platform is craving to hear.
As mentioned, the Decision Readiness Maturity Model includes many other steps that touch on Teams, Software, Servers, Accesses, and many other elements that will be discussed in our future articles.
