Not So Fast: The Problem With Starbucks Distribution Model
Starbucks is closing stores after years of aggressive growth. Will banks face the same fate?
Starbucks re-wrote the playbook on distribution strategy by looking beyond traditional demographics to highly detailed neighborhood-level information and street traffic flows to determine the best store locations. It worked like a charm for 15 years and set a new standard for the industry. So when Starbucks announced on July 2 that it was closing 600 underperforming stores, the magnitude of their store shuttering was a surprise.
What went wrong?
Lesson 1: Stick to the model. Starbucks didn’t. They had a model that worked and applied it rigorously through most of their expansion effort. But under pressure for continued growth they strayed from their proven process, relied more on sales judgment and less on disciplined analytics, and ended up accepting locations that turned out to be sub-optimal.
Lesson 2: Don’t lead the market. Starbucks built aggressive growth projections into their location decisions, especially in hot markets like Florida, Nevada, California and Texas. When real estate development tanked, so did the household growth and store traffic that was key to their revenue forecasts.
What are the implications for financial institution site selection and distribution strategy?
The analytics work. The models aren’t perfect, but they are about 70-80% accurate in projecting deposit growth – certainly better than local market judgment or the recommendations of real estate brokers alone. Our own BankPower® process has proven to deliver 189% higher deposit growth vs. similar branches that did not use this methodology. We believe there is a balance between using proven, objective data combined with market experience – just as there is a balance between credit risk models and loan officer judgment. But in our experience, the models are usually more accurate than judgment alone, and certainly we have learned that locations where the models project low ROI are unlikely to be successful, irrespective of local management enthusiasm.
Leading the market – land banking for the future – usually is not a good way to invest your capital. The logic behind reaching out to new market areas is that real estate is cheaper and there is a “first mover” advantage. There is a reason real estate is cheaper: there’s less potential. As for the “first mover advantage”? Time and again analysis has shown it that the return on investment just isn’t there. It’s more efficient to wait for sufficient growth to occur and pay a higher entry price to get the right site than lead the market.
Why is this important?
Will banks suffer the same fate as Starbucks?
During the past 9 year period, the number of banks and savings institutions has declined 20% while the number of branches increased 17% to over 97,000 (not including Credit Unions). More importantly, the rate of branch growth is faster than household formation or the rate of growth in domestic deposits. Those are not sustainable trends.
As the FDIC noted, “The growth in physical branches is all the more striking in that it occurred during a period of rapid technological advances that would appear to have diminished the need to use branches.”
We can joke about a Starbucks on every corner but the fact is, there’s a bank branch on every corner – and sometimes 4 competing branches. The growth trends for bank branches aren’t sustainable, just as they weren’t sustainable for Starbucks. But there are growth opportunities and bankers that maintain a disciplined branching strategy can, and do, achieve higher than market gains.
As for Starbucks, there will still be about 10,000 domestic offices after the projected closings, so don’t worry — there are still plenty of places to get your ‘bucks!