Home / Banking Strategies / How the best banks optimize branch hours with predictive analytics

How the best banks optimize branch hours with predictive analytics

Jun 18, 2018 / Consumer Banking

So you’ve closed some branches and reduced staff at others. But you still feel serious pressure to reduce operating expenses without disrupting sales and customer service value in the branch channel. So you start to evaluate open hours as your next pocket of opportunity. Sounds like a fine idea but beforehand, ask yourself: Do you know the pitfalls of the traditional approaches for tweaking branch hours? Here are seven pitfalls to consider:

  • you look only at teller transaction volumes and ignore the relationship between sales and open hours
  • assume that demand patterns during the week and on weekends are the same
  • ignore the value small businesses place on convenient branch hours
  • fail to address open hours for high-volume branches
  • neglect network impact
  • overlook the impact of competitors’ open hours in the nearby branches
  • bypass scenarios where increasing open hours to align selling capacity with market opportunity makes sense.

That’s quite a few ways to get it wrong—common ones, in fact. What happens if you do?

Tom McDermott, managing partner of Inver Consulting Group, lists the potential business consequences: “You may inconvenience loyal consumers and small business customers. You may be out-positioned by competitors. You may lose sales during impactful hours. You may amplify the impact of consolidations and closures.”

All of this adds up—literally—to loss of revenue, as well as reputation and customers. Which steps must you take to get it right?  The answers lie in looking three ways: inside, outside and around each of your branches.

Look Inside: When do sales occur?

Collect internal data and not just transaction data. Consider branch activity from a sales/ growth perspective. Does this branch deserve investment? Fixing? Rationalization?

Put an equal or greater focus on sales. “Day-parted sales data is key,” says Dr. Abhyuday Desai, vice president of analytics at Kiran Analytics. “You will generally see that sales data is heavily influenced by neighborhood demographics. In business areas, sales demand is heaviest on weekdays. Demand in the suburbs will skew to the weekend.  If you have the data, use it.”

And if not, “Look at analytics-based clustering versus peer groups based on what drives differences in demand distribution,” Desai advises. “Organizational groupings are irrelevant, as well as sales potential groupings: saturated, high potential, etc. Belonging to one of those categories doesn’t mean they have anything in common as to when customers walk in.”

Look Outside: What are the other guys doing?

As McDermott points out, “Competitive analysis for each branch based on its market type is an important consideration. You may observe that other banks in the market open later or close earlier without impact to their performance. You may discover that they expand branch hours in one location and reduce hours in a nearby branch with less opportunity.”

While it is important to observe what other banks do, open hours and position don’t necessarily correlate in most markets.

Look around: How are your branches related?

Examine your network to determine nearest branch adjacencies. Then, define your market types—urban, suburban, rural, etc.—to determine your market clusters. Desai explains: “There are models out there that look at proximity or driving time. But the relevant metric is determined by type of market. For example, in Manhattan you need to talk about walking distances. In that case, four blocks may be the edge of the cluster. That’s very different from a rural town in Connecticut where five or ten miles is reasonable distance.”

Also crucial: “You need to understand peer groups: Manhattan versus rural Connecticut, for example. Only then can you estimate how much demand will move to a neighboring branch based on relative proximity for that peer group.” This provides the analytical foundation to estimate sales impact if you close a branch on Saturday or reduce hours during the week.

Measure, start again, repeat

“The most accurate measure of success in this process is to measure your reduction in full-time employee expense versus any reduction in sales numbers,” Desai explains. “You have to be careful about using measures such as customer satisfaction to evaluate the impact of changes to branch hours.” That factor is influenced “by so many things that it’s difficult to directly link to open hours.”

Speaking of hours, optimizing branch operation time frames is not a one-and-done project. Every bank needs to continuously evaluate and adjust their open hours. Some large banks have already instituted quarterly evaluations driven by advanced analytics.

Rather than adjust branch hours based on teller transactions or one-size-fits-all approaches, analytics-savvy retail strategy executives make it a part of their overall distribution capacity optimization process.

Here predictive analytics dominates and the doors, at least in this aspect of the branch world, must always remain open. In growing bank business, that’s the ultimate path to opening doors.

Want more Banking Strategies? Sign up for our free newsletter!

Deb Stewart of Charlotte, N.C., is an independent consultant working for the financial services industry.

BAI Consumer Benchmarking programs allow your organization to receive key metrics for more than 118 million consumer households, and $4.5 trillion in combined product balances.