RSR Research: Retail Connections – Predictive Analytics in a Down Economy
By Paula Rosenblum, Managing Partner, RSR Research
Through a special arrangement,
presented here for discussion is a summary of an article from Retail
Paradox, Retail Systems Research’s weekly
analysis on emerging issues facing retailers.
At an analytics roundtable at RetailConnections’
Third Annual Business Executive Summit, I had a really interesting exchange
with the CIO of a major jewelry retailer. I asked, “What good could predictive
analytics possibly have done for you in 2009? After all, no one had a clue
what demand was going to be.”
The answer I got was fascinating. In this
CIO’s view, predictive analytics had been completely helpful to his company:
sales were trending low, so retailers bought based on those trends, and
essentially sold out to the walls. In many ways, 2009 was a very profitable
year, with little excess inventory.
That really gave me pause. All
year we thought retailers were under-buying. And no one really predicted
aggregate holiday sales with any degree of accuracy. The NRF thought “flat”;
PwC: “up one percent”; me: “up two percent.” I think the final number was,
in fact, up two percent, but not driven by the categories I thought they’d
be. So if we couldn’t predict aggregate demand with any sense of confidence,
how could an analytics engine predict sales by SKU?
I suppose the difference is in the desired
sell-through rates. Average seasonal sell-through rates are typically about
65-75 percent. Apparently retailers gave themselves the opportunity to
raise that rate into the 90’s. And it was good. Yet this year, port traffic
is back up, and retailers appear to be buying up again. Even so, no one
seems to have their arms around aggregate demand yet. So have we decided
to drop sell-through rates again? If so, why? If you have some thoughts
about this, please do let me know. It’s one of those retail paradoxes that
I can’t quite figure out.
Discussion Questions:
What do you think of the value of predictive analytics? How does such
analytics overcome uncertainty in demand?
While I can’t predict sell through rates, I can say predictive analytics are the cornerstone of shopper marketing ROI. Illumination of demand metrics helps set appropriate investment levels, and works well on the manufacturer side as well as in retail. When done well, year two should be even better, with applications of lessons learned…and so on.
That said, its also essential to keep close tabs on the shopper to understand emergent desires early on, enabling some opportunity of inventory to surprise and delight recession weary shoppers!
Retailers need to use all of the tools available to them to look into the future. This is essential for managing inventory levels and controlling costs. Predictive analytics may not always be right, but they can make a great difference for retailers.
Predictive analytics are always important and in a down economy, they may be even more critical. Whether forecasting demand, inventory, or marketing expenditures, the current environment requires use of more multidimensional thinking.
Traditional predictive analytics leans heavily on history and established patterns to forecast the future. Today, it is necessary to incorporate more external and environmental variables. These may include macroeconomic conditions, key consumer trends, and competitive activity, just to name a few.
Of course, each market or category may also have unique variables such as the supply of a key ingredient, the timing of certain events, or just the weather.
If you are trying to prepare beef bourguignon, you had better have a good chef, or you just may get beef stew.
Forecasting applications can be put into three categories: the historical simulation method uses past performance, the variance-covariance method attaches the series of interest to some other variable, and the Monte Carlo or stochastic simulation method uses probabilities and random number generators to generate a range of possible outcomes. All of them assume some type deterministic characteristic for the future. None of them plan for the “Black Swan Event,” the event that is so rare no one would expect it to occur but when looking back seems so obvious. The general economy suffered just such an event in September of 2008 and despite all the warnings that seem so obvious now, we were not ready. Only a few wise people had pulled away from the punch bowl.
I don’t think it is wise to make any long-range decisions based on the results of 2009. Because the circumstances of 2008 were so extremely (we hope) unusual, the accurate predictions for 2009 can be thought of as nothing more than good guesses. As the economy returns to the “new normal” and economic events follow predictable patterns again, forecasting algorithms will adapt to the new standard. These will be helpful in forecasting category demand and planning purchasing activity. That’s good because the channel options have added a level of complexity that did not exist before. It is no longer “how much will people buy” but also “how will they buy” that must be answered. So if we get the demand right, then we have to know which channel will be used.
Forecasting algorithms are important, but they will not make good merchandisers great. Just like any tool, the algorithms will keep you from making horrendous mistakes but the really great merchandisers combine the algorithms with a dose of wisdom. The great merchandisers had a good 2008 holiday season.
If there is anything about the new normal that will look like the old normal it will be about freshness of selection.
Consumers are understandably cautious, but they want to be optimistic. When they see sales floors and shelves sparsely stocked or even empty, it sends a depressing signal. The psychology and emotional side of the sell is very important, as we all know.
The merchant in me loves seeing 90% sell-throughs; the consumer in me wants more selection and inspiration from a merchandise mix. The consultant in me questions where the balance is.
Retailers and consumers alike need to be more efficient in their buying habits. Hopefully the tension between the old and the new normals will make us more innovative on behalf of our customers.
There is a tendency to look at the issue as a macro one, rather than a truly granular challenge.
The issue is not how retailers manage their inventories, although that appears to be the issue. The real question is–what do Best Customers need from your company, and can they get it when they want it. If you do not have what they need, they will seek it elsewhere, which provides opportunity to experience other brands and potential attrition. If you provide enough Best Customers with enough opportunities to experience enough brands, they will leave your business, and you will be left with fickle, promotion-sensitive customers who will leave you in a heartbeat if you do not give away your profits to them.
Is that loss balanced by the benefits of lower inventory and higher turns?