Why In-Store Supermarket Branches Are a Bad Idea

In-store branch opportunities may look appealing in theory — high foot traffic, with lower capital and operating expenses — but compared to traditional branches, they are rarely successful.

By Jon Voorhees with

Subscribe to The Financial Brand via email for FREE!For years I have shared with colleagues in the financial industry my perspective that supermarket or in-store branches are a bad idea. It’s one of those topics that often sparks debate, and it tends to get me ranting.

For instance, ask CEOs and senior executives in the banking industry about their supermarket branch strategy, and here’s a sample of what they have to say:

  • “We can open eight in-stores for the cost of one traditional location,” a CEO of one regional bank said, as he lauded the affordability and profitability of supermarket locations. “The annual operating costs are slightly lower than a traditional branch.”
  • “You’ve seen us bring online a number of in-store branches which we think are going to be very advantageous for us in the future in terms of optimizing the cost of distribution,” boasted a chief credit officer at a regional bank. “Our customers are migrating to those branches for their transactions, and we actually sell very, very well in those facilities.”
  • “We have a mixture of traditional branches and in-stores, and it’s the in-stores that are really growing,” said another regional bank CEO. “The cost of a traditional branch compared to an in-store branch is three, four or five times as much.”
  • “We have grocery stores that have 15,000 to 20,000 shoppers that walk in front of our in-store branch every week,” explained an EVP and director of retail banking for a large community bank. “We think the in-store model provides some in-store acquisition opportunities.”

As you can see, those who believe in the in-store strategy consistently like to stress two points: (1) supermarket branches are cheaper to build, and (2) supermarkets have lots of foot-traffic. Both statements are true, and that’s generally been the foundation for their appeal. But the problem isn’t cost or foot traffic. It’s lack of revenue.

Just think about what Christopher A. Williams of the Federal Reserve Bank had to say about supermarket branches over 20 years ago, back in 1997:

“Supermarket branches date to the 1960s,” Williams wrote in “Instead of growing into profit centers, however, these outlets generally became mere check-cashing centers. Accordingly, many failed and were subsequently closed.”

Okay, Let The Rant Begin…

Many banks — large and small — have experimented with in-store branches over the years. I’ve worked for several banks that deployed in-store branches, and in every case where the bank exited its in-store program, another bank was waiting in the wings to take its place.

But let’s look at some facts:

  • In its last update on branches (June 30, 2017), the FDIC reported that there were 4,684 supermarket/in-store) branches, representing 5.4% of all retail bank branches.
  • While getting comparable data on credit unions is nearly impossible, it’s safe to assume that about 5% of the 21,000 credit union branches in the country are located in supermarkets as well, so that adds another 1,000 supermarket locations.
  • According to FDIC reports, the average supermarket bank branch today holds $21.5 million in deposits. That’s just 27% of the traditional branch average nationwide.

The concept of efficiency is broadly used to evaluate how effective banks are in revenue generation. The efficiency ratio — the ratio of cost to income — is a quick and easy measure of an institution’s ability to turn resources into revenue. Remember that the theoretical appeal of supermarket branches is that they cost less to build and operate. The first part is absolutely true: the average supermarket branch is about 15% the size of a traditional branch, and only takes 25-35% of the capital to build.

Total operating expenses, however, are not reduced as much. Lease rates per square foot are higher, but since the size is small, the occupancy expense is much less than traditional branches. The real cost driver is staffing. These smaller branches don’t need a lot of staff, but they do need to cover longer hours and weekends to take advantage of in-store traffic, so bank and credit unions wind up having total FTE comparable to- or even higher than that of traditional branches. Add together the high lease rates and the long operating hours, and expenses end up running about 65-75% of traditional branches.

If the average branch deposit base (which serves as a good predictor of profitability) is only about a quarter of a traditional branch but expenses are 65-75%, how can this be a good use of capital — with the majority of the expenses, but only a quarter of the revenue?

Reality Check: In-store branches average $21 million in deposits, but the average is misleading. Digging deeper, 43% hold less than $10 million, and 72% less than $25 million.

There are some banks that make in-store branching a key component of their retail strategy. According to the latest FDIC branch reporting, there are ten banks that operate more than 100 branches where more than 10% are in-store location. You might assume that they have figured out how to solve the in-store branch riddle. For these ten banks, the following chart compares their average deposits per branch for traditional branches vs. in-store locations. There are wide differences in performance.

Comparing these banks to peers, 90% have efficiency ratios significantly higher than their national peer groups. Banks with a heavy reliance on in-store branches have efficiency ratios as high as 85%, meaning it costs $85 to generate $100 of revenue, compared to peer banks where it only costs $57.

Bottom line? I don’t like in-store branches because they are rarely profitable, and therefore are a poor use of capital and an inefficient asset. They also carry some element of additional risk from partnering with grocery chains that can (and do) make store closing decisions on short notice, forcing the financial institution to seek regulatory waivers or scramble to find ways to serve their customers.

I’ve had colleagues argue that “not all in-store branches are bad,” and that “they work in certain situations.” Both statements are true. There are some instances where an in-store branch is the only way to serve a market, especially in some low- and-moderate-income neighborhoods. But the facts don’t lie. Most in-store branches — as they operate today — don’t deliver adequate returns, especially for the additional risks they present.

Can in-store branches be successful? One reason why in-store branches mature at such a low deposit base has to do with the typical in-store branch design. The spaces are small and often follow a traditional branch model. That means they have an ATM, teller line of 2-3 windows, and an office. It’s hard to pack much more into that tiny space. Most of the space is dedicated to serving customer’s transactional needs.

If in-store branches can work, but usually don’t, what should banks do to be successful? I’ve been thinking about alternative approaches to manage in-store branches. Here are my four suggestions:

1. Think small. If your space is small, and the purpose of the branch is to serve as a convenient spoke for larger branches, then organize the branch so it can be run with only 1-2 person serving as a concierge. Automate the teller line with full self-service ATMs. Use the concierge to help customers when they need it, and to do business development during slow times.

2. Thank large. Small space is a constraint, so don’t build tiny branches. What if the space were designed to have two offices, a freestanding teller station with TCRs (teller cash recyclers), and a deposit-taking ATM? That way there is more room for selling.

3. Think automation. In-store branches will always be small (just don’t make them too small), and that means having traditional roles like tellers will create inefficiencies that make the branch unprofitable. Change the focus of the branch from being a transaction site to more of an advice and sales center by automating the teller line so you can use universal bankers and maximize the space, and your staff utilization, for selling.

4. Think carefully. Yes, there is a role for in-store branches and they can be a successful part of your mix. But be very careful with the right markets, the right mix with your traditional branches, the right size, and the right configuration.

What has been your experience? I’d love to hear your experience and feedback.

Jon Voorhees is an advisor at based in Austin, Texas, specializing in banking strategy. Before joining Peak, he was head of Distribution Strategy and Execution for Bank of America. You connect with Jon, please send him an email.

This article was originally published on March 14, 2018. All content © 2018 by The Financial Brand and may not be reproduced by any means without permission.


  1. You should seriously think about reccomending the NCR Interactive Teller Machine for in store branches. This allows for the automation of 95% of teller transactions and freeing the staff for sales and advisory. The efficiency that this provides is much better than even TCR’s and can help move these types if branches to becoming profit centers with a very small footprint.

  2. Nice one Jon.
    Its great to see the arguments being put forward based on facts and the data from the market. However, if we have to measure every outlet for a bank based on its efficiency ratio, then surely we would be banking on apps only (which have the smallest costs (relatively). There are other metrics, brand awareness, customer service etc.

    Your points on how to improve the efficiency ratios are all good, and the possibility of putting in Assisted Self Service as Jason suggests, all move the ratios based on people interacting with the customer base, providing sales opportunities, for sure… but critically, the idea of the in-store branch is to create the opportunity to have those face to face interactions, make sales, but also to solve problems, and perform customer servicing (change of address, order a new bank card, all of which may not be calculated into your efficiency ratios. And your 4 recommendations are really about maximising that interaction, which should be the same for every branch, the only difference being the size constraint for in store branches.
    Great to see data being used.

  3. Nice article. But I’m not clear about the problem. One of the persons quoted at the start of the article says, “we actually sell very, very well in those facilities.” You begin by contradicting this when you say the problem is “lack of revenue”. But at the end of your analysis, you shift the problem to low profitability / return on capital employed. These are different metrics. What is the problem, really?

  4. In-store branches achieve and help banks realize all of the benefits mentioned. Where they fail, like any business strategy, is in execution. You cannot design an in-store facility with a traditional banking mindset. The in-store Customer Experience is about convenience, access, and frictionless customer service – which equates to as much self-service as possible. Our firm designed and launched an in-store facility for a community bank that was more akin to a hip coffee house than a traditional bank. Associates with notebooks walked the floor “selling” the banks services. Unfortunately, “bankers be bankers.” They will do all they can to cling to traditional branch models.

  5. Agree with the 3rd point around automation, then they’re not a ‘bad idea’.

    Make the outlet cashless from the start and the space is better utilized for quality sales and advisory conversations. Particularly if the outlet is located within realistic distance of a traditional branch with tellers. The maintenance costs of such legacy outlets are also vastly reduced by being housed inside an existing structure.
    Finally, mirroring supermarket trading hours makes that bank a key differentiator. The social aspect of grocery shopping can be a platform from which to inexpensively promote extended banking hours.

  6. Hi Jon – Thanks for the excellent insight.

    I am not convinced that banks should measure supermarket branches by profitability. A couple of reasons:
    (1) Supermarket branches are built for convenience, need to be positioned as spokes of a larger hub, and should be measured accordingly. Sales is nice but is not the primary objective.

    (2) A branch in a supermarket frequented by a customer definitely makes the bank more attractive, but the customer may decide to open the new account in a traditional branch that provides a much more relaxing account opening experience. I know I would…

    (3) Finally, the old branch-based profitability model is outdated. Most customers rarely visit branches regularly and no longer associate themselves to a particular “home” branch. If I don’t know where my home branch is and have no relationship with anyone in the branch, why should my deposit/loan counts as his/her performance?

Speak Your Mind


Show Comments