I’m describing the three main attributes of an analytically driven, agile business. Let’s assume you read my last blog, “Define the Strategy.” Now it’s time for the second key trait.
#2: Culture of integrating quantitative and qualitative into business decisions
Wow! This cuts right to the heart of a company. What is culture? Can you manage culture? Sounds easy but it’s not. Out of all the management levers, managing culture is the most difficult. Yet culture is the key to becoming an agile company.
At a minimum, the culture should support: ingraining quantitative data in all of the business processes and nurturing the desire to experiment.
Quantitative data and the business process
Since a strategy, by definition, involves targets and measuring methods, the culture needs to appreciate quantitative information. Quantitative information such as financial information has always been appreciated by the leadership team and the finance team. But financial figures alone are not enough. An analytically enabled company uses quantitative information in all business decisions as practicality allows. How is that possible? Does everyone need to think quantitatively?
The senior leadership team already thinks quantitatively about the financial or the sales numbers. They also need to understand market entry events, timing, and customer changes. At large companies, there are teams of market researchers, business strategists, business operations managers and other groups all designed to manage information for the leadership team. A commitment to these teams is a commitment to agility at the most basic, data-gathering level. Without supporting data, without the “evidence,” a CEO should be hard pressed to approve and guide a business unit’s business plan.
What about a project manager or an individual contributor? Whether you are the business leader, a project manager (PM) or an individual contributor, you should integrate quantitative information into your job. A PM tracks budgets but should calculate the Estimate to Complete (ETC), a forward-looking measure. A PM should understand the variances to plan after the project ends in order to improve estimates on future projects. An individual contributor should understand the “hours” budget allocated for a task and notify others if the task cannot be completed in the time allocated. Integrating quantitative information into everyday decisions, into business processes, is a hallmark of an analytically driven culture. Fact-based decision making is also a giant step towards accountability and transparency.
Most information we gather in the real world is biased or “dirty” in some way. We cannot rely solely on quantitative information. Management judgment has to be integrated and is sometimes the dominant source of decision making. That’s okay as long as it’s a conscious trade-off.
Integrating quantitative information into the business process is more challenging but incredibly valuable. Setting prices by customer segment, calculating an optimal route for package pickup, providing a discount to frequent customers, estimating the delivery time of a drop-shipped product, creating loyalty programs, or making a real-time product recommendation on a service call are all ways in which analytics can be embedded into the business process. Each day, a company makes hundreds if not thousands of decisions around serving customers, all driven by analytical information. In today’s world, not using analytical information in the business process is a recipe for failure.
Desire to Experiment
A hallmark of an agile company is the desire and the ability to experiment. How do you know if the best course of action is the one you took? Maybe reaching customers through an indirect sales channels lowers sales and raises sales costs. Maybe the new tiered service model actually raises short-term costs for the small to medium size customer segment. Maybe the customer experience is improved if you enact cross-SBU branding effort. Or maybe not.
Maybe. Maybe. Maybe.
Dangerous words. Managers manage. “Maybes” have to be converted into knowledge and action. Maybes can only be converted by trying new approaches, making changes and observing the results. A company has to be set up to try and fail multiple times. If you are an online company, this is often easier. Change ads on your sites. Or change the layout. Or change the navigation.
But what if the cost of changes, of addressing the maybes, is far larger? It’s harder to try out a newly-optimized cross-sell up-sell process in an expensive call center channel than it is to change a banner ad. It’s easier to change the letterhead or colors than it is to change a disease management program and the frequency at which healthcare professionals reach out to at-risk individuals with diabetes.
There is no best answer here. Your strategy should decide what needs to be tried and changed. It should anticipate it. It should also plan for experimentation and testing various hypotheses. If you need analytics to answer these questions, you need analytical systems to help you judge the success of an experiment.
That’s why you need to think about your analytical style—what type of analytic approaches you use for different situations. That’s the subject of my next blogs.