Applying Data Science to Make Smarter Decisions

Since their invention, computers have always played an integral role in the business world, and in today’s digital economy, data now provides the new “diesel” that fuels competitive advantages. Business managers are empowered to replace gut-feeling presumptions with data-driven decision support.

Many technologies have served as catalysts for the data science that’s behind these smarter decisions—from big data to artificial intelligence, machine learning, the cloud, mobility, and data mining. It’s imperative to find a way to leverage these technologies to manage the enormous data sets your business is already collecting.

That means extracting it from multiple data sources, processing it, and presenting it to business users so they can make smarter business decisions. And it all has to be done in real time while making sure the way the data is handled complies with data privacy regulations. Data science covers the spectrum of all these requirements.

As you embrace data science, know too that it will be vastly different five years from now. Decision-makers and the IT teams that enable them to make smarter decisions won’t become obsolete. But responsibilities will change as managers transition from tactical decision-making on what needs to be done today and this month to more strategic decision-making on what needs to be done next quarter and next year.

In this sense, managers will become more effective, making more intelligent decisions than ever before. Let me illustrate with three examples:

Smarter Spare Parts Deployment

To support normal maintenance procedures, capital equipment manufacturers often have to deploy spare parts to geographically-dispersed distribution centers. In the past, the spare parts would be deployed to regional areas for staging based on engineering estimates and perhaps the number of machines in each location.

But consider if data was available regarding how past equipment had performed in each of the specific geographies. Perhaps the maintenance manager could determine that certain parts are consumed more frequently in certain geographies due to the local climate. And as a result, parts have to be sent from other depots for replenishment—at a further expense.

Using this data, spare parts could be staged in order to support this tendency and avoid future shipping expenses. This would also lower overall maintenance expenses.

Smarter Direct Marketing

Within the commercial real estate sector, agents have to find multiple properties that will be bought as bundles by potential investors. It isn’t enough to cultivate a user-entered profile of potential buyers; their interests can change.

In addition, a bundle that interests one investor will not interest another. This makes it difficult to automate bundling and marketing efforts.

However, agents can use an investor’s purchase history to create similar bundles as those made in the past. This improves the propensity to purchase by offering smarter bundles of properties with similar characteristics. It also reduces the time properties are in inventory and lowers marketing expenses.

Smarter Medical Treatments

When new medical technology enters the patient care arena, the only information available to doctors are the training materials relating to the treatment, which are based on the patient testing phases. However, the data set grows at each point of treatment, providing medical practitioners a richer set of data from which to drive their decisions regarding treatment options.

Data science makes it possible for doctors to use appropriate, anonymized data acquired within the patient medical information system. They can then analyze how a treatment has worked for specific kinds of patients with similar health.

The knowledge base driving the decisions will then continue to grow in accuracy—rather than remaining static. That same knowledge base can even provide computer-aided decision-making and patient treatment options. This will increase the accuracy of treatment selection, accelerate the delivery of care, and improve the lives of the affected patients.

Data Science Adoption Rates Rising Sharply Across All Business Function

These data science examples from the manufacturing, real estate and healthcare sectors aren’t isolated. Similar trends are occurring across all industries. For example, Statista projects the market for big data will soar to $85 million by 2026—an increase of nearly 42% compared to 2020.

And the chart below from Clarion shows the impact that data and analytics are having on core business practices across 10 industries:

Change to Industry's Core Business Practices Brought About by Data and Analytics

These trends demonstrate how decision-making powered by data is no longer a “nice thing to have” but rather a requirement to remain competitive. If you don’t find ways to capitalize on data to drive customer experiences and improve the efficiency of internal workflows, your business could be in peril.

Common Themes for Driving Your Data Science Project

If these industry-specific use-cases and data science trends whet your appetite, perhaps you are ready to consider developing a data science project for your business. Here are some common themes to help guide you as you begin your journey:

  1. Be clear about what decisions you think will make your business operate smarter.
  2. Look for data that can replace presumption or fixed-decision matrices and provide continually-evolving and improving decision support.
  3. Model the impact on the business bottom line of being able to make those smarter decisions.

As you consider these three themes, you may realize you do not have a data science team within your organization that is up-and-running and ready to go to deliver on your project. So what should you do?

That’s what we will talk about that in the next article, So You Want to Do a Data Science Project: Now What? (Part 1)