Real Estate / Analytics / Finance Models and Store Closures

Most large retail brands have a Real Estate department. This department finds the best places to build new stores, and works with Finance and Strategy to determine markets to move in/out of.

In recent years, Real Estate departments have partnered with Analytics departments to understand the impact of online sales on existing / new store locations. The Finance folks appreciate this partnership, because it is becoming harder to justify keeping open B/C locations.

Let's review (at a very high level) the factors that come into play when determining if a store should be closed.

First, most large retailers maintain a "strategy worksheet" for each existing store and each potential new location. The strategy worksheet lists a series of metrics about each store (demographic metrics, performance metrics, competitive metrics, and channel-centric metrics). When evaluating new locations, each new location is compared against existing locations - finding "comps" where applicable.

Second, forecasts for each of the next five years are produced for existing stores. The forecasts take into account trends across the customer base (i.e. 40% of 12-month buyers will purchase again next year, spending $200 each, while the store will attract "x" new customers spending $130 each - yielding a forecast). The forecasts take into account online cannibalization (i.e. as the web grows, the web pushes customers into stores - and as the web grows, the stores push customers online at a faster rate). The forecasts account for competitive issues. If Macy's is closing at a location, then future forecasts are adjusted by "y" percent. When I worked at Nordstrom, we loved it when Neiman Marcus moved into a market because existing Nordstrom stores performed better due to Neiman Marcus traffic. This kind of adjustment is baked into the forecast.

Third, profit-and-loss statements are run for each store for each of the next five years. The p&ls are run for each store, and more often these days are run for the "trade area" where a store exists, including online sales.

Fourth, profit-and-loss "what if" statements are run assuming a store is closed. Large retailers know what happens in a trade area if a store is closed. If a store generated $2,000,000 in annual sales and the store is closed, it is common to see 15% of sales move online ($300,000) and 15% of sales move into nearby stores ($300,000) and 70% of sales to simply disappear. Based on historical store closures, "what if" scenarios are run. Between Real Estate / Finance / Analytics, a list of future "unprofitable" trade areas are published. C-Level Executives and Board Members then determine what to do with "unprofitable" trade areas.

How about we work through an example?

Let's assume that we own one store in Portland, Oregon. The store generates $1,500,000 in annual sales, with 25% of sales flowing through to profit, and $150,000 in fixed costs associated with the store. In addition, the online channel in the trade area generates $300,000 in sales, with 25% of sales flowing through to profit and $30,000 of advertising costs (10% of sales) to generate online sales and $30,000 of fixed costs allocated to the online brand in the trade area.

• Store Profit = $1,500,000 * 0.25 - $150,000 = $225,000.
• Online Profit = $300,000 * 0.25 - $30,000 - $30,000 = $15,000.

Again, we'll walk through stuff at a very high level. There is a lot more detail that goes into this stuff than I am sharing here.

Forecasts are produced for this store - the models use customer repurchase rates, new customer acquisition count assumptions, and channel crossover assumptions (among other attributes). In our example, let's assume that the store will weaken because sales move online over time.

• Year 1 Store Sales = $1,500,000 * 0.98 = $1,470,000.
• Year 2 Store Sales = $1,470,000 * 0.98 = $1,440,600.
• Year 3 Store Sales = $1,440,600 * 0.98 = $1,411,788.
• Year 4 Store Sales = $1,411,788 * 0.98 = $1,383,552.
• Year 5 Store Sales = $1,383,552 * 0.98 = $1,355,881.

Similarly, those sales will move online - and better yet, two things happen. First, online sales will grow at an organic rate and second, online marketing expense will increase as online sales are associated with Facebook Advertising and Paid Search and Affiliates and Regargeting. Let's assume that online sales will grow by the amount that in-store sales decline, and let's assume that online sales grow by 5% organically above-and-beyond current rates. Here's the online forecast for the trade area.

• Year 1 Online Sales = $300,000 * 1.05 + $30,000 = $345,000.
• Year 2 Online Sales = $300,000 * 1.10 + $29,400 = $359,400.
• Year 3 Online Sales = $300,000 * 1.15 + $28,812 = $373,812.
• Year 4 Online Sales = $300,000 * 1.20 + $28,236 = $388,236.
• Year 5 Online Sales = $300,000 * 1.25 + $27,671 = $402,671.

Ok, this is where things get interesting. Your Real Estate team and Analytics team will measure the impact of, say, Macy's closing a store in this mall. When Macy's closes, let's assume that sales drop by 3%. And let's assume that other stores will close in future years, hurting retail sales by 3% in the first three years, and then by 5% in years four and five.

• Year 1 Store Sales = $1,500,000 * 0.98 * 0.97 = $1,425,900.
• Year 2 Store Sales = $1,425,900 * 0.98 * 0.97 = $1,355,461.
• Year 3 Store Sales = $1,355,461 * 0.98 * 0.97 = $1,288,501.
• Year 4 Store Sales = $1,288,501 * 0.98 * 0.95 = $1,199,594.
• Year 5 Store Sales = $1,199,594 * 0.98 * 0.95 = $1,116,822.

Technically, we need to adjust online sales growth because we have less retail sales, but we'll skip that step for now, as it won't fundamentally change our story.

Make sense so far? Good!

Let's run a profit and loss statement for each of the next five years. We will assume that retail fixed costs and online fixed costs will increase by 3% per year. Ready?

Yup, this market is dying. The store is losing sales to the online channel, and the store suffered because of other store closures in the mall. As a result, profit is in decline.

What does the forecast look like if we close the store? Let's assume that 15% of sales move online, 15% of sales move to other nearby stores, online marketing costs increase as sales move online, and retail fixed costs disappear without the store. What happens?

Look at that! If we close the store, the trade area / market is less profitable. So we're stuck, aren't we? The store is going to perform worse and worse - and yet, profitability is still there, so we have to keep the store open.

This is the process that retailers go through to determine when a store should be closed.

Are there factors that change the relationship I illustrated above, causing stores to close faster or slower? You bet there are!

• Retail Fixed Costs:  Retail brands that are "debt heavy" have greater fixed costs, however, if the store closes, the fixed costs don't go away - they simply hang there and drag on profitability. If a company has minimal debt, then fixed costs disappear, causing the p&l to look better when a store closes, accelerating store closures.
• Online Marketing Costs:  If a retail store closes and sales shift online, and if those sales are then tied to Paid Search and/or Facebook and/or Affiliates and/or Retargeting, then there are added costs that result in a better situation by keeping the store open.
• Online Fulfillment Costs:  Free shipping causes an odd dynamic - variable costs per order increase and as a result the existing store (without the variable cost) becomes more profitable by comparison. In other words, free shipping accelerates customer shifts to the online channel while causing the online channel to appear less profitable and therefore causing Management to keep the existing store open longer.
• Shift To Online Channel:  The faster sales shift to the online channel, the faster you'll close stores. The issue isn't the percentage of sales that are generated online, but instead the rate that sales shift online.
• New Customer Generation:  If the retail store can generate new customers faster than the online channel can generate new customers, then the math will dictate that the store must stay open longer so that the store can fuel online sales growth. This is a little-understood dynamic that only the smartest retail brands take advantage of.
• Flow-Through to Profit:  When a store can generate 40% to 50% flow-through to profit (on a variable basis), the store can stay open longer even when sales shift online quickly. When a store has to discount heavily and can only generate 20% flow-through to profit, then fixed costs quickly overwhelm the p&l causing more profit to be generated by closing the store.
• Market Cannibalization:  The rate that sales move to other stores and/or the online channel dictate whether a store should be closed or not. In a saturated market, it is common for 70% of the sales to remain after a store closes ... half moving to other stores and 20% moving online. Brands that are closing stores (Macy's, JCP, many others) frequently close stores where market saturation is high. When there is only one store in the market, it is common for 15% of the sales to move online and 15% of the sales to move to other stores. Traditional Retail is closing stores because of the interaction between online sales shift and market cannibalization.
• Product Mix:  Certain product lines sell better online than in stores, and vice versa. If the product mix yields comparable rates of sales online vs. in stores, stores are going to close faster. If the product mix skews so that some items sell online better and some items sell in stores better, then there is a place for stores and closing the store will result in the death of product lines, causing long-term online sales to decelerate.
• In-Store Experience:  When you have a Bass Pro Shops or Cabelas, you have an in-store experience that is fundamentally different than walking into a J. Crew. This means it is easier to close a J. Crew store and not lose sales - whereas Bass Pro Shops will get killed when they close a store because those sales are not moving online. If you want to keep your stores open, you need a better-than-average in-store experience.

There are so many other factors as well - factors dependent upon each individual brand. Closing a Nordstrom Rack store cuts off new customers for Nordstrom Full Line stores, which then cuts off Online growth, so you have to put that information into your model, right? Each individual company deals with their own individual challenges.

Does the explanation make sense?

Do you have any questions?

And sure, I do this kind of work, so if you need help give me a holler (kevinh@minethatdata.com).