Phantom Demand – It’s Not What You Think

Today is one of my favorite topics…phantom demand. No, it has nothing to do with superheroes or Andrew Lloyd Weber or anything else of a non-business ilk. It’s a wildcard in forecasting that can have the biggest impact more than you think possible. There are two significant uses of phantom demand when it comes to forecasting. The first, and most often used, relates to asking and answering the question, “How many would we have sold if we didn’t sell out?” Let’s face it, too many companies run their forecasting efforts utilizing the “Last year plus ten percent” rule. And that may be fine, as it may be fairly accurate at a macro level. But a mentor of mine from my days with Staples, The Office Superstore was known to say, “Retail is Detail” and forecasting is exactly that – on steroids.

Example. You have a Top 10 item that you’re selling out of every year. The Last Year plus 10 rule is flawed here - Ten percent of what? You’re going to continue to chase the sales potential of this item, selling out every year. Phantom demand – providing you with a rationale of what you reasonably should have sold, would provide you with a much better forecast. And that’s just the one item. I’ve found that almost a third of companies sell out of a third of the offering before the season is 75% complete. What these companies are leaving on the table is a much bigger number than they suspect.

The second significant use of phantom demand relates to how you forecast new items. Let’s face it… New items in your business represent the backbone of your business growth. I like to use a rule of thumb that every company should have 15% of its offering new to the customer each and every year. That is a daunting number to most - but if you found a way to put that into practice, you simply will have a better business. New items give existing customers a reason to continue to shop with you, whether it is through your catalog or on your website or reading through your email contact strategy. New items are what drive customer retention. And customer retention drives profitability. Especially when you are selling something that is not purchased year after year. Like perennial plants, trees, shrubs, or mattresses. Let’s face it, getting forecasts for new items correct really matters.

Think about your own garden. You’ve already bought what you thought you needed. But what about something new…the newest echinacea, the newest genus in your offering…the season’s newest color of the year. And the year you introduce these new products often is their best year. So, you need to get that forecast right. 

The basic process of figuring out phantom demand is how you organize your data. All forecasts need to start with history…the more, the merrier. And, the more variables appended, the smarter the forecast will be. i.e., genus, color, pot size, etc. Each sku should have a “segment” appended, as well. Everyone has A, B or C skus. But do you know your AAA skus? Do you know your never out skus? With many clients, I refer to some skus as their “never out of stock ever” skus. The acronym is eerily appropriate.

Ideally, one of your best, Microsoft Excel people will do the heavy lifting on this. You need to organize your sales history by year and over time in the best way possible. And you should build a “percent in” curve per item. If you break the year up into two 26-week portions, then you start week one on 0% and you move towards 100%. The problem is if you sell out by week twenty then you’re at 100% by week twenty and you’ve left six weeks of business on the table. Wrong, wrong, wrong. Having a data-driven process to forecast phantom demand - using your history, your averages, your data – will make sure you’re fulfilling what you could’ve sold for future years.

Now, I know many of you are saying yes, but they didn’t buy that item that sold out. They bought another item. That’s probably true. And so, we recommend a way to counter that when building out your phantom demand numbers. Line up all your data. Figure out what your percentage is of items you don’t sell out. And, put every other item that you do sell out on that curve. Measure the DELTA in units. And then make assumptions for how many of them you could’ve picked up. We use 50%and are usually, on average, correct. 

The first thing you will notice is that your last year plus 10%, while utilizing a layer of “found” phantom demand, might actually forecast a total increase closer to 20%. (!!!)

In addition to running his own business consultancy since 2003, Ken Lane is a Senior Business Advisor for PivotPoint Business Solutions. Connect with Ken Lane at ken@pivotpointbizsolutions.com or hathawayandlane.com.

Ken has been a phantom demand savant for the better part of the past thirty years and is happy to help your organization maximize your sales potential by Building Better Forecasts.

Ken will be co-chairing a workshop at Cultivate 24. Click here for details.