A long time ago, around 2009, I wrote about killing same store sales as a comparative measure for tracking retail performance. My argument at the time was that store sales were no longer an accurate measure of a retailer’s health, thanks to the strong growth in online. I think same store sales (SSS) lingered even in 2009 due to a stubborn perception that eCommerce was a “different” channel and that sales gained in that channel should theoretically be additive to in-store sales.

That view of the world has grown increasingly muddied. Digital is now a major growth opportunity for large retail chains – and even those looking to open new stores in new territories (i.e., international growth), are increasingly looking to their digital strategy both as a way to mitigate the risk around making such a major investment, but also as an important part of the go-to-market strategy, particularly in established markets.

Some retailers have hedged on the challenge of balancing SSS against online growth – by rolling online sales into stores numbers, for example, usually based on the postal code of where the order is delivered. Retailers still ask us about how to best credit stores for cross-channel transactions, and as you’ll see in May when we release our latest store benchmark, not even retail Winners have come to any conclusion as to how best to account for cross-channel sales.

So what does that leave? How do you measure retail performance? The original idea of SSS was that retailers should be able to grow sales, not just buy them. SSS made it difficult for retailers to hide sluggish sales in their existing stores by rapidly opening or acquiring new stores – that’s why you don’t look just at top-line sales growth in retail. If retailers are growing their sales at existing stores that have been open for more than a year, and doing so at a rate that exceeds inflation, then those retailers are “winning” in the market.

In an omni-channel world, that no longer works. Especially when no one has definitively decided how to account for omni-channel sales. You could do what a lot of retailers have done – just roll digital channel sales into stores based on geography of the purchase – but I think it loses a lot of the finesse that SSS captured when all there were was stores. And it could hide a lot of store cannibalization behind very rosy online numbers.

A couple of weeks ago, Jacob Roche at the Motley Fool proposed a new metric to get a better sense for retail performance – employees per thousand square feet, or EPKSF, used in conjunction with sales. According to FierceRetail, which jumped on the story, retailers who have growing EPKSF and growing sales are doing well. Frank Hayes at FierceRetail disagreed, arguing that technology should increase retailers’ leverage, and decrease the need for employees. He also argued that different retailers have very different EPKSF’s, which makes it impossible to compare retailers to each other – something that SSS makes it easy to do.

I don’t necessarily agree with the logic. Retailers have been ruthlessly hacking away at store labor budgets for years, and while Costco, Walmart, and Safeway (three of the retailers used in Roche’s analysis) are not exactly high-touch customer service retailers, even low-touch retailers have been looking for ways to add labor dollars back into stores (well, okay, maybe not Walmart), and looking for ways to do it so that they don’t get hashed by Wall Street for rising labor budgets. Technology isn’t going to solve that problem, and it alone cannot solve the problem of the need to change the role of the store in an omni-channel world to a more service-oriented venue, both in terms of selling and in terms of the post-sale transaction. For that, you need people.

Hayes doesn’t propose an alternative metric, but it’s a real problem. It’s problematic to hold retailers to a standard of rapidly declining relevance, like SSS, but it’s also a problem not being able to compare retailers to each other without putting a huge number of caveats around the comparison. Did they grow or not? Really grow – not just hide stagnant sales with new stores. That’s the question.

Given how much omni-channel is opening up the way retailers think about inventory, I wonder if the solution to this challenge is to look not at stores, but at total square feet – retail and distribution in particular, and look not just at sales, but at the inventory levels required to generate those sales. I’m still thinking about it, but it seems to me that inventory leverage is going to become more and more important – those retailers who can sell inventory from anywhere in the chain, and not just primarily from expensive retail square footage, are the ones that stand poised for success in an omni-channel world. If that’s true, Wall Street should be looking to reward retailers who grow sales faster than square footage growth, whether retail or distribution. And they should be looking to retailers to grow those sales without growing the inventory required to support those sales.

I’m still pondering, but I have a feeling there’s an answer in there somewhere. It just must be an answer that one, is easy to calculate, two, provides a “real” assessment of growth, and three, is easily comparable among retailers. And takes into account omni-channel sales and fulfillment. A piece of cake!