Is Consumer Demand Really That Unpredictable?
I had occasion to spend some time with the results from our benchmark report on omni-channel order profitability, and there was one thing that struck me in a new way. According to our survey respondents, unpredictable consumer demand is their top business challenge. I’ve read that finding and talked about that finding a dozen times, and it was only this week that I realized how silly that is. After much consideration, I would offer that consumer demand isn’t really that unpredictable. Instead, retailers are just really bad at predicting it.
Part of the issue is one of “garbage in” – as in, garbage in yields garbage out. There seems to be a lot of garbage in the inputs retailers use to decide what to stock where. First, a lot of them start with last year’s plan, which means whatever bad assumptions you made last year are automatically carried over into this year. If no learning happens over mistakes that are made, you are doomed to repeat them.
Then there is the level of granularity used. If you’re looking at demand at a category, or even sub-category level, you’re still going to miss a lot. I remember seeing a retailer’s toy aisle in January one year where it was immediately apparent that the retailer was not looking at the category at a low enough level of detail. How could I know this? One whole side of the aisle was supposed to be Nerf – Nerf guns, balls, etc. Half the aisle was made up of stocked out holes, and the other half was over-stocked, to the point that employees had put some of the overstock into the holes, just to have a place to put it. The retailer clearly saw a category that was moving “well”, but couldn’t see it at level of granularity that would’ve told them guns were going like gangbusters, but balls were not.
Another area where I see retailers ending up with garbage inputs is around estimating missed demand. I’ve heard from plenty of demand forecasting vendors about how they can impute missed demand from stockouts. A good solution will, prior to forecasting next year, add back in estimates for missed demand as an overlay on the retailer’s reported demand from last year, to make sure that the forecast doesn’t undercount before using that information as the basis for next year’s forecast. Remember, what you sold is not “demand”. It’s just what you sold. There could’ve been plenty more demand out there to be captured, but you didn’t get a chance to capture it because you were out of stock.
I know that retailers are not good at estimating lost sales – all I have to do is listen to their shock and surprise when they start showing store inventory online. The eCommerce DC can only hold so much inventory and assortment, right? It inevitably stocks out of something. Most retailers take those products out of circulation on the site, because they don’t want to show consumers products that they can’t buy. But when you can show that a product is available in stores – you don’t even have to promise that you’ll ship it to them, only that it is available in a store near them – you no longer have to take those products out of circulation just because the eCom DC is out of stock.
For fashion retailers, the impact was immediate. “We had no idea the depth of demand for some of these items” was a phrase I heard over and over again.
Really? You had no idea? But if you are doing a good job of predicting demand, wouldn’t you know this already? I’m not sure exactly where this issue rests – are the algorithms that predict missed demand hopelessly flawed? I know many of them operate on the idea that there is a certain rate of sale that happens across some time period, say, a week, and if that rate of sale isn’t reached, and the product is shown to be out of stock, then you can make a pretty good guess that the rate of sale that week should’ve theoretically been higher.
Or maybe retailers are abusing this information, either by ignoring it, or fiddling with it and adjusting it when they should be leaving it alone (also a bad retailer habit when it comes to these things).
Or – and I would put some money behind this one – maybe the challenge comes from having inaccurate inventory counts – if you expect a certain rate of sale and look against inventory, and it shows as in stock, then it’s going to be really difficult for a system or algorithm to parse out that the rate of sale is not actually a true reflection of the possible rate of sale, because even though there was an inventory number, there was nothing on the shelf. And retailers readily admit that their inventory is inaccurate, which is becoming more and more painful as they increasingly need to be able to promise specific items in specific locations to specific customer orders.
It Shouldn’t Be That Hard
I go back to my opening statement: Consumer demand is not really unpredictable. It’s just that retailers are really bad at it. When you don’t have granularity, when you aren’t starting from clean inventory data, and thus clean demand signals, and thus consistently under-estimate the demand that you’re losing out on, then yeah, I guess it could look to you like consumer demand is just outright unpredictable. Especially when you add the increased complexity of more channels, more inventory and demand crossing channels, and endless aisle assortments that make managing the granularity of your total assortment ever more difficult.
But the reality is, you must get better at it in order to win at retail. Unpredictable consumer demand is an excuse for bad processes and data – and a lame excuse at that.