Needs-based Design Makes the Models Work

Software and product design have been around for a couple decades now. The best software designers and providers take a very service-oriented approach to how they design and create software looking at things like:

• What does the customer want?
• What does the user need?
• What would they like in the product?
• What works/doesn’t work?

But uses of data? A lot of companies use data; they use models … but studies show that use of those models and data remains low. Thus:

If you build it, they might not use it.

However, the companies that are using data best and most efficiently take an approach that looks more like product design and best practices in service – listening to the customer, designing to the user’s needs, and more.

In an International Institute for Analytics article, Tom Davenport, a highly regarded name in the field, profiled Regions and the bank’s use of data products as an example of a company doing it the right way. In particular, his article, published on his blog and elsewhere, talks about the importance of taking an approach similar to software and product development when building and designing analytics and machine learning products and services.

“I remember reading his book, ‘Competing on Analytics, The New Science of Winning,’ and being very impressed,” said Manav Misra, head of Regions Data and Analytics. “In his recent article, he highlights how we have applied a product development-oriented approach to our data products at Regions. It’s a great honor to be profiled by him!”

Excerpt from the blog post:

“Regions has been employing the product orientation for three and a half years. It has been quite successful, with over ten revenue-generating/cost-saving products (with incremental impact in eight figures) and several more for internal support functions.”

Read the full article here.