Try a New Model for Branch Remodeling

Register now

Banks spend a great deal of money remodeling branches. Many fork over tens of millions of dollars on remodel initiatives every year, and some of the largest banks will spend over $100 million.

We commonly hear two typical responses when asking banks about the return on this massive investment: "We never really measured it," or "It is hard to measure, but our best guess is that the return is low."

In both cases, the underlying assumption is that investing in the branch network is simply a cost of doing business. Not investing at all would, in the long run, result in out-of-date, run-down branches and would surely lead to the ultimate demise of any bank. Many banks are not currently employing three easy strategies to realize significant returns on remodel investment:

Avoid a common remodel frequency across branches. A limited number of branches can be remodeled every year. Which branches should be selected for the next wave of remodels? Commonly, banks employ a set frequency, such as every 8-10 years, and branches that have gone the longest without an upgrade are slated for the next wave of remodels. This approach, typically combined with operational considerations, is often the driving factor behind when and which branches are remodeled.

However, using a set remodel frequency makes little sense. The needs of each branch are different, and how often a remodel is required will vary greatly based on branch traffic, competitive presence, and many other factors. An urban branch in a highly competitive market may need a remodel every 3-5 years, while 12-15 years may be just fine for some rural branches. Similarly, there may be urban branches with a specific customer profile that responds more powerfully to remodels than other similar urban locations. Selecting which branches to remodel should be based primarily on the projected benefit of the remodel without focusing as heavily on a specific remodel frequency.

Don't spend the same amount on each branch. There is a lot of buzz around the "Branch of the Future," and some banks are spending big to upgrade large numbers of branches. The fact of the matter is that this level of investment simply won’t make sense for a large portion of your network.

Many branches perform as well as they can and don't have the growth capacity to make a massive remodel worthwhile. Banks need to determine where a major remodel would yield sufficient benefit and target only those branches.

This isn't to say you shouldn't invest in other branches. Remodels are still important and can improve performance for those branches. However, spending less can yield the same benefit and much higher ROI.

Figure out which remodel investments make sense. Many banks have a "remodel package" that all branches receive. There are a variety of components that may be included — signage, flat screen TVs, revamped teller line, layout redesign, etc. Most banks do not know which of these elements actually drive remodel success.

Some components are good ideas that provide high ROI and improve performance for the branch. However, others will be dead weight cost or will yield too little benefit relative to the cost. Eliminating these components from the remodel can significantly improve ROI.

Using a cookie cutter approach to remodels leaves a lot of money on the table. To build the best remodel strategy, banks need to look at recently remodeled branches. Using these remodels as "test branches," banks can answer critical remodel strategy questions — i.e. which branches should be remodeled next, how much should be spent on each branch, and which components should be included in the remodel.

The common theme here is to not make uniform investments across the network and instead to target spend, generating the most profit while still taking into account the practical concerns of maintaining a branch network.

Will Weidman is a vice president at Applied Predictive Technologies, a provider of market testing services.

For reprint and licensing requests for this article, click here.