Unprofitable Customers: Naughty or Nice ?

Customer value helps identify the top tier customers we need to focus retention activity on, but what of the other 80% of customers who generate nominal or even negative contribution? Are they naughty or are they nice to have in your portfolio?



This question has been troubling strategists and marketers since customer profitability measurement first became viable in the early 1990s. And with good reason: depending on what you are measuring and what your goals are customer value can suggest very different actions. It is imperative to measure customer values using a model appropriate to the decisions you want to make. More often than not you will discover that a variety of measurement models are needed to support different kinds of decisions. (See CMA article How should we measure customer profitability). Even if you’ve got the models you need, however, it is inevitable that 60% of your clients are going to be somewhere in the middle and 20% at each of the top and tail of your list.


Let’s consider the bottom 20%. Are they “bad” customers that we should de-market? Are they “abusers” of our services? Customers in the bottom quintile of value rarely have an “average” customer profile. In this tier you will find customers with a wide variety of business relationships with your bank, most of them fairly substantial in terms of balances and activity. If you dig deep enough into the numbers, you are likely to find pricing at the root of their negative value. Some will have their value depressed by shrewd negotiation of rates and fees, others by strategic discounting and others still by irrational market pricing conditions.


Negotiated discounts in fees and rates certainly need to be taken into account when assessing customer value. But pricing anomalies driven by market conditions or strategic discounting have little to do with the customer, and should not be included in customer value. The extreme case of this is when the market prices entire business lines at negative spreads, which happens from time to time in periods of crisis. Whole segments of customer values can turn from gold to brass in a matter of months when this happens.


Obviously one cannot switch customer relationship strategies with these shifting winds of chance. You need to look past the numbers to manage customer strategy effectively. Clearly we need to reprice relationships where excessive discounting is negotiated. It is equally clear we should not penalize customers for aberrations in market or strategic conditions. There is no substitute for wisdom and understanding when working with customer value !

Best holiday wishes to all.
David McNab

Unprofitable Customers: Naughty or Nice ?

Current value, historical value, present value, lifetime value....

One of the more interesting aspects of customer value measurement - the precedent to customer value management - is that it can have so many different temporal definitions. There is a vast difference between each of these terms:

• customer current value
• customer historical value
• customer present value
• customer lifetime value

 and even within those terms there are variants in meaning. For example current value may mean last week | month | year or it may mean the value of the customer's current business over the next week | month | year.

Getting basic definitions right is crucial to successful acceptance of your customer value measurement techniques. Each of the value metrics identified above has validity - there is no one right answer or single version of the truth that everyone seeks. The reality of customer value measurement is that the techniques and policies used to define the metric must be tailored to the decisions that are to be based on this information.

For pricing decisions one set of definitions might be ideal, while for customer retention strategy another might be more applicable. You need to get the relationship between definitions and decisions right if you want to encourage the right behaviour...and that is what it is really all about.

- David McNab

Current value, historical value, present value, lifetime value....

The illogical logic of credit card rates

We have been hearing a lot of talk in Washington and Main Street about how credit card issuers are managing risk and rates lately, so I thought it would be a good idea to offer up a quick analysis of the so-called rationale that supports raising rates on deteriorating-credit clients.

The pricing mantra goes like this...

the expected credit losses on the portfolio is rising so we need more revenue to pay for those losses. And we raise the rates more on people with the worst credit scores because they are the most likely to go under.

Well, this is not very sensible, folks. There are several problems with this way of thinking.
First and foremost the issuing card company (that is, issuing the credit i.e. making the credit card loan) is supposed to be in the business of adjudicating credit and taking managed amounts of risk. When they oversell by issueing cards to people who are bad risks the issuing bank/ card company is supposed to take a hit - it's their fault for making the bad loans i.e. failing to adjudicate credit responsibly. It is not the responsibility of the vast majority of credit card holders to "make it up" for them through higher rates. Allowing credit losses to be built into the portfolio rate base promotes bad credit card sales and shifts the burden of credit adjudication cost to cardholders which is patently unfair.

Second, raising rates on the least-capable-to-pay customers is also inappropriate. The right thing to do - for both the card company and the customer - is to limit and reduce the amount of credit available, not raise the rate. Deteriorating credit status that occurs after after a loan is made (credit authorized / limit granted) is something that was supposed to be assessed and factored into the cost of credit before credit is granted to the customer. The risk of deterioration in credit quality should be taken as a pooled risk at the time the card is issued, not after-the-fact. And even then, does it make sense to push the customer with the worst risks into bankruptcy faster by raising rates ? The answer is clearly NO. Individual credit risks need to be assessed and predicted and factored into credit underwriting before lending to people, not afterwards. It doesn't even make any sense from a collections perspective, because the high risk=high rate paradigm makes credit quality worse, increasing lending losses. Reducing the available credit amount is the only sensible way to work out a deteriorating risk situation.

And on another note, has anyone noticed that merchants are getting whacked with horrible charges to pay for excessive "loyalty" programs ? My local restaurant gets charged one rate for Amex, another for regular cards and a different rate for premium cards. To him it is all the same - payment for a sanwich and a cup of coffee - but the interchange he gets charged differentiates between cards. So my local restaurant is paying for the bells and whistles on their customers' cards. That isn't fair either. So I don't use a card with my local vendors, I want them to prosper - I pay them cash and that is only fair to them.

The illogical logic of credit card rates