Let’s tell the truth about customer profitability analyses

  • Richard Thelwell
  • July 3, 2015

truth customer profitability analysesSimplifying a concept makes it easier to explain and understand. But occasionally such simplifications are misleading, raising expectations that can’t readily be met. Which tends to be the case with one commonly-encountered hypothetical illustration of Business Intelligence in practice—namely customer profitability analysis.

Because, time and again, you see Business Intelligence books, blogs, and marketing materials blithely talking about running reports and queries in order to establish a business’s most profitable customers. Which—erroneously—suggests that such an exercise is something that you can dash off on a whim.

You can see why such examples are used, of course. It’s precisely that sort of profitability information that is guaranteed to get ‘C’-suite executives salivating, and presumably eager to sign off on a Business Intelligence project.

But it’s far, far too simplistic. And yet, as with the fabled emperor’s lack of clothes, this is a truth that’s rarely acknowledged.

Output, not input

One immediate problem is that ‘customer profitability’ isn’t a searchable field. It isn’t a field that even exists in an ERP system.

You want to know your biggest customers, in terms of sales revenues? Easy. Oldest customers? Easy. Your customers located in Scotland? Easy.

Such information can be obtained by either selecting on particular fields, or carrying out calculations based on the results of various fields. In short, they are straightforward reporting tasks.

Selling price minus discount?

To get a real handle on customer profitability, the required calculation workload—not to mention the amount of database I/O activity—is of an entirely different order of magnitude.

One approach might be to look at each customer’s gross purchases—based on invoice value, either in aggregate, or at line item level—and then deduct any discount given.

customer profitability analyses discount
Discounts are an important factor in customer profitability analyses

If most customers buy mostly the same sort of things, then this is an approach that gets you part way there. So if discounts from a standard price list are a significant feature of the pricing model of your business, then you may feel that you have a fairly good proxy.

But if discounts from a standard price list aren’t a significant feature of the pricing model, or there are significant variations in product mix, then you’re really still stuck at the revenue level, rather than having any true clues as to profitability.

Taking a margin-based view

To close the gap, you’ll want to look at the standard cost of products sold, which provides insight into gross margin, and then factor in the selling price and any discount given.

But by now, you’re operating at the line item level, the whole thing is starting to get very calculation-intensive, and there’s an awful lot of database I/O activity going on.

Yet even then, how informative will the output actually be? Because a standard cost, as any accountant knows, makes significant implicit assumptions about overhead absorption, supplier pricing, volume levels and product mix.

And your profitability answers are only going to be as good as those assumptions.

Payment discounts and other credits

So far, we’re operating at the ERP level, with a view of customer profitability that reflects what the ERP system sees.

But depending on the business in question—and the ERP system in question—that may or may not be the full picture.

How about discounts for prompt payment? Or—which isn’t quite the same thing, but close to it—pricing with an element of retrospective dynamics discounting? Or the impact of retrospective volume-based credits, which reward major customers when they reach specified volume levels?

customer profitability analyses discount
Factors such as prompt payment discounts must be considered

These could significantly skew the profitability picture—and the relevant data might not be in your ERP system, but in a separate payment system, potentially one administered by a third-party.

Cost to serve

Finally, remember, too, that standard costs are product-related, and make no explicit assumption regarding a given customer’s cost-to-serve.

So you’ll obviously want to consider the impact of distribution costs, for instance, which again may not be reflected in the ERP system.

For that, you’ll perhaps want to look at your Transportation Management System, if you have one, or examine data on carriage costs—which may be held on a carrier’s platform, or remitted from the carrier in a monthly data file in a spreadsheet or other intermediate format.

Simple query, complex answer

In short, a supposedly simple analytics task such as assessing customer profitability has quickly become a complex many-headed beast. And one which, needless to say, is going to have a significant impact on your business’s IT performance.

If, that is, your business doesn’t take steps to offload that computational and database I/O workload.

And here, a Cloud data warehouse and a Cloud reporting and analytics tool would be a good place to start—but those over-simplified Business Intelligence books, blogs, and marketing materials don’t always say that.

In our view, they should.

To learn more about Cloud reporting and analytics, download our free bumper Ebook below