Optimizing Your Branch Network

By Deb Stewart

The need for physical branches is in decline. Coping involves more than just figuring out which ones to shutter. It means figuring out the highest and best use for each remaining facility.

Here are the distressing facts: More than 2,600 U.S. bank branches closed last year, with most large banks contributing heavily to that total. These actions included the following:

Bank of America closed nearly 150 branches last year, according to CNBC television.

PNC Bank, Pittsburgh, has closed 240 branches since 2013. Most of those cuts came in 2014 when 190 locations were closed.

JPMorgan Chase Bank recently announced it is closing about 300 branches (5 percent of its total) in an effort to trim costs, Bloomberg reports. JPMorgan also slashed its number of ATMs by 12 percent in the fourth quarter of 2014.

European banks between 2013 and 2020 are expected to close 40 percent of their branches—meaning 65,000 fewer branches, according to the consultancy Bain & Company, Boston. U.S. banks are expected to have 20 percent fewer branches by 2020, with the trend accelerating after that, according to the white paper, “Retail Banking 2020: Evolution or Revolution?” by PricewaterhouseCoopers (PwC), New York.

And it’s not just the number of branches that is being cut, PwC cites that the average number of branch full-time employees has been reduced from 13 per branch in 2004 to an average of less than six today.

Is the branch still important?

As a part of PwC’s “Retail Banking 2020” study, top executives were asked that question. “Fifty-nine percent of respondents expect the importance of branch banking to diminish significantly as customers migrate to digital channels, and 48 percent expect branch banking to change significantly by 2020.” Yet, only 16 percent of respondents viewed themselves as “very prepared” for this shift.

Respondents globally view the largest banks as benefitting most from these changes, and smaller regional and community banks being the most threatened, PwC adds.

Jean-Werner de T’Serclaes, a partner at The Boston Consulting Group, says the future of bank branches wasn’t just about closing them. “Banks are looking to optimize their branch networks rather than eliminate them,” he says. “Relatively few bank customers want to do everything online, and most still want branches for face-to-face conversations about complex products such as mortgages.”

Optimize your branch network

So if optimization is more than simply closing branches, what is it? What should you consider as you approach a network optimization project? What are the outcomes you are working toward?

Here are five general steps identified by leading experts in the field:

–Have the right people guiding the process and building consensus.

–Validate your banks’ strategy—business/segment/geographic—optimization needs to occur in this broader context.

–Gather as much market and competitive data as possible.

–Analyze the data to arrive at several conclusions for each market/branch: close, consolidate, invest, reconfigure.

–Repeat—consumer behavior is going to continue to evolve—optimization needs to be approached as an iterative process.
Let’s view each of these steps in greater detail:

  1. Have the right people guiding the process and building consensus.

Darren Dewey, senior vice president, director of retail distribution at Associated Bank, Green Bay, Wis., offers this advice, “We have touched 200 of our 230 branches in the past five years—an executive committee was formed at the beginning of the process with our CFO, chief retail officer, commercial real estate manager and community market leader. We needed to make the right capital expenditure decisions with changing consumer behaviors and technologies. This balance of perspectives helped to ensure the right decisions and buy-in at the most senior levels.”

Denise Stokes, senior vice president, director of retail and marketing for Sandy Springs Bank, Olney, Md., describes a similar approach. “We established a channel committee about three years ago. Channel implies more than just the branch. We started broad-based, asking what is omni-channel about for us and what is the role of the branch in that context? The committee oversees all research and decisions around optimization. The retail/marketing head, head of commercial, CFO, chief technology officer, head of real estate and even our chairman were active members of the committee.”


  1. Validate your banks’ strategy—business, segment and geographic role of the branch.

“You have to look at the historical role of a branch and how roles have changed—if traditional roles include advertising, recruiting new clients, cross-selling and transaction management—ask yourself how that mix has changed,” says T’Serclaes. “Much more of the emphasis today is on sales and recruiting. As you begin the optimization process you need to decide what the implications are to your branch density and format?”

“Start from the perspective of how do I fulfill my target and current customers. It’s not about “What can I tweak,” it’s about “Where do I want to be in the future,” says David Kierstein, president of Peak Performance Consulting. “What are my target segments? What are my target geographies? You need to establish that framework before you start collecting and analyzing data.”


  1. Gather as much market and competitive data as possible.

Hal Hopson of Pitney Bowes suggests that you start with micro-market level data. “Know who lives in each geography—understand their demographics, channel preferences, psychographics and channel usage. What is your current share of wallet in the market? Survey branches in person, and collect attributes around these branches. Understand the competition and branch density in each area. Is it already overbanked? Look into the future. What level of growth in number of households and household income are expected?”

Hopson continues, “Understand market preferences and commuting patterns. What is the willingness to drive in a market? You may find that customers are willing to drive two miles for transaction servicing but five miles for a more complex transaction such as a mortgage or investment product. Take a look at population density, customers in more rural markets generally have a greater willingness to drive. Who wants to buy in person? Who wants to service in person? Who wants to do both? All of this data comes together to inform your optimization decisions. Do you exit or expand in the market? Which branches do you close, sell or reconfigure?”


  1. Process the data to arrive at several conclusions: close, consolidate, invest, reconfigure.

“It’s about deciding where and what—the way you approach data analysis must reflect your business objectives. Should you expand into new geographies? Is your future focus on attracting/retaining more of the same segments, or do you need to attract new segments? Analyzing data to uncover greatest opportunities may lead you to add experts to deepen existing relationships. Or it may lead you to expand your branch network into markets with higher concentrations of new target segments. We find that, typically, 80 to 90 percent of revenues are generated by 50 to 65 percent of branches in a given network, creating considerable opportunity for rationalization,” T’Serclaes continues.

“One bank’s analysis may lead it to withdraw from a market, but, as we’ve seen, this may create an investment opportunity for banks seeking to expand their presence in a high-opportunity market. For example, Huntington Bank took over 37 former Chase branches in Meijer (grocery) stores in various parts of Michigan. And Huntington plans to add branches in six additional Meijer locations. Washington Federal has acquired 74 Bank of America branches in Nevada, Arizona, New Mexico, Washington, Idaho and Oregon. Ameris Bank acquired 18 Bank of America branches located in North Florida and South Georgia. Optimization is about reallocating capital to the areas of greatest opportunity not just shuttering branches,” Kierstein adds.

“Reconfiguring branches to address a reduced consumer preference for in-branch servicing and to enhance productivity is another important piece of the optimization puzzle. Chase is reconfiguring 1,200 branches this year. Most of these have not involved complete redesign of the branch. Rather they have kept one or two teller windows open in their existing teller lines and put ATMs in the other windows. Many community banks have been less aggressive about making these changes than they should be, thinking that the investment will take a long time to recoup,” says Kierstein. “That is not necessarily the case.”

PNC has shown that the optimization process can be approached one step at a time. After closing 240 branches since 2014, Bill Demchak, CEO and president, described the bank’s approach at a 2014 Goldman Sachs Financial Services Conference, saying that 2015 will be less focused on closing branches and more on converting 300 locations to a teller-less, universal concept by the end of the first quarter. Universal branches require, on average, one less person to staff and are more productive, Demchak said. PNC has stated that by 2019, two-thirds of PNC branches will be teller-less with staff focus on customer service and selling products.

PwC’s study notes that leading banks are moving away from “managing branches” and instead are “managing distribution” across all the bank’s channels including evolving branch models to balance local-customer needs with the high cost of branch delivery. They are designing their branch strategies to deliver a differentiated experience, based on customer needs, the competitive landscape, brand promise and internal capabilities.

Banks choose an appropriate mix of branch models to support their desired customer experience. PwC sees various models being experimented with today—assisted self-service, in-store branches, full-service branches, community centers and flagship stores.

In Accenture’s study, “Banking 2016: Accelerating Growth and Optimizing Costs in Distribution and Marketing,” it states that networks composed solely of full-service branches with duplicate services and skills are no long sustainable. “The branch network model must be designed with careful consideration to differentiated formats and integrated in a multichannels views that is able to maintain territorial coverage, react to local customer needs and optimize skills and capabilities. This extends beyond full-service branches and hubs to include light branches (with an average of four to five employees), kiosks and cashless branches. By migrating low-value activities to digital channels and retail-based formats, banks can satisfy new customer segments and further overall branch network transformation.”

The Accenture study goes on to state: “An efficient ‘hub and spoke’ branch model can help enable a bank to decrease the branches in its overall network by 15 percent to 20 percent and reduce average branch staff by 25 percent with the proper mix of flagship, full-service light and kiosk branches.”

A potential mix of differentiated branch types are summarized in the study as follows:

–Cashless, kiosks: maintain for service; 50 percent of network; 100 percent service.

–Light branches: maintain for sales; 30 percent of network; 50/50 sales/service.

–Full-service, hubs: maintain for advanced banking; 15 percent of network; 40/60 transactions/sales.

–Flagships: innovate, attract; 5 percent of network; 30/70 transactions/sales.


  1. Repeat—consumer behavior is going to continue to evolve—optimization needs to be approached as an iterative process.

Hopson reminds us, “History is irrelevant as you approach optimization—make sure you are closely tracking how customers are interacting through different channels. Behavior has changed fast and is going to continue to change. You can’t put this project on the shelf.”

Boston Consulting adds that two of key factors in successful optimization are test and learn. “As behaviors continue to change banks need to be continuously evolving their delivery model. Test and learns at the branch, market and region level must become a way of life for banks moving forward. Be humble. This is going to be an iterative process.”


Case Study: Follow the Data


Institution: Sandy Spring Bank, Olney, Md. Assets: $4.1 billion

Project lead: Denise Stokes, senior vice president, director of retail and marketing.

Time working on focused optimization program: three years

Branches: 49 at beginning of program, 44 today

Project leadership:

Channel committee oversees all research and decisions around optimization. The retail/marketing head, head of commercial, CFO, chief technology officer, head of real estate and the bank chairman are active members of the committee.

Validate your banks’ strategy:

Channel implies more than just the branch. We started broad based, asking what is omni-channel about for us and what is the role of the branch in that context?


The bank knew that data was going to be critical to decision-making. It partnered with Pitney Bowes for data. It looked at both demographic and market data—for today as well as anticipated population, household and household income growth on five-year horizon to define demand. Competition is also analyzed looking for branch density for us and for our competitors—how many branches do top-five banks in any county (in the market) and how many do you have? In some cases we found markets that were already overbanked or markets in which our presence was so small we cannot realistically compete. We looked at the ratio between deposit availability and number of branches (county or smaller tract level).

There was a distance analysis that helped to understand how far away from a customer you can be before becoming irrelevant. We looked at this from both an account opening and a transactional perspective.

Analyze and conclude:

Five closings have occurred in the past three years, reducing network size from 49 to 44. Most of these were easy decisions—unprofitable, low-volume branches with a receiving branch within short distance. We have relocated a branch and are using the data to test new sites looking at 20 potential new sites from new build or acquisition.

Our traditional branch model has been a stand-alone with four tellers and two drive-ups. Given the dramatic consumer behavior changes, we have opened one new concept branch. It is more self-service oriented with no traditional teller line, universal bankers, image-enabled ATMs and WiFi. The concept as a whole is being tested as well as looking for opportunities to bring specific elements into the rest of the network.


We are testing things all along the way. We know that it’s a journey—adding and transitioning over time. We’ve concluded that half of our branches were fine and that half should have some change. As a part of the ongoing analysis, we have a rolling analysis of our leases to identify optimal timeframes for making some of these changes. This process is a long-term thing.

Words of advice:

Follow the data. It’s the prudent, fiscally responsible thing to do. It’s hard to do it yourself without a partner to help with market data/competitive data. That’s why we work with Pitney Bowes. The data helped set the level internally—for instance retail and commercial may have different “opinions” about a particular site. Returning to the data helps make decisions guided by facts.

It has been critical to get strategic alignment across partners. And moving forward we know that we need to keep testing and keep up with the trends.


Case Study: A Journey, Not a Destination

Institution: Associated Bank, Green Bay, Wis. Assets: $23 billion

Project lead: Darren Dewey, senior vice president, director of retail distribution.

Time working on focused optimization program: five years

Branches: Associated is an association of smaller community banks in Wisconsin, Illinois and Minnesota. Its branch network is one-third smaller than it was a decade ago.

Project leadership:

Regular check-ins with business lines—all decisions are approved by executive committee including the CFO, chief retail officer, commercial real estate manager and community market leader. There were regular check-ins with the business lines and business line scoping meetings at the onset of the process.

Validate your banks’ strategy:

This process is a combination of business, geographic and real estate investment strategy. This process takes a community market perspective, looking beyond retail strategy.


Get started with data—transactional, performance data, deposits, ATMs, total square footage, deposits/square foot, demographic trend data and behavioral segment data. Then look at performance versus opportunity.

Local market input helps you dig deeper into the data. For instance, Wisconsin has great infrastructure, but new by-passes in regional and middle markets have unintended consequence of shifting travel habits. Branches still matter, and customers still choose banks based on location, but we need to address that need in different ways.

Analyze and conclude:

In past five years, we have touched 200 or 230 branches—updated or remodeled, some relocations, some de novos, consolidated some sites (re novo). We have many different branch models out there today with a concentration of 4,500 square foot sites. By next year, any new “large branches” will be in the 1,500- to 2,000-square-foot range.

Many of these existing 4,500-square-foot sites are in the right place but are simply too large. We have been quite successful in demising many of these spaces—keeping the drive-through and cutting the branch in half. Other halves have become restaurants, dentist and medical offices. We’ve been careful with signage and directing traffic around the building.

We’ve added a transaction express branch type. The first is located at the University of Wisconsin, Madison. This is an aided self-service model at 1,000 square feet or less—reflecting changes in business strategy, staffing and technology.

Transactional support comes from 24-hour self-service zones with high-function ATMs and teller pods with teller cash recyclers. Technology kiosks allow staff to easily demonstrate online or mobile. The sit-down element of the kiosk can also be used for self-service needs such as online loan origination if the need arises.


This is a journey not a destination—changing market conditions and customer behaviors demand that we continue to learn and adjust.

Words of advice:

Don’t be afraid demising your oversized space. But always try and do it from a developer’s point of view so it is marketable in the long run.

This is a process that happens over time—stay connected to the local market for qualitative assessment. Get to the data as quickly as possible so you have something to base adjustments on and to measure success. This needs to be part of a holistic strategy—physical and digital together.


Deb Stewart of Charlotte, N.C., is an independent consultant working for the financial service industry. Telephone: (704) 759-1633; email: [email protected].