Monday 9 April, 2007

How do you measure loyalty?

Humm…

How do measure love? Or hate? Or interest? Or indifference? Perhaps psychologists can. Maybe they can measure loyalty too. But can brand-building business executives make sense of their methods and measures?

More to the point, who is loyal customer? In an ad agency, he can be very different persons to the account executive (trusts me, asks me, doesn’t order me, wants ideas and second opinions) and, say, the account manager (lets me raise bills), not to mention the accountant (pays bills, on time). Such differences are bound to exist in every company and industry. When you talk about ‘the loyal customer’, are you sure that everyone who needs to understand understands, more or less, the same thing?

Also, why do you want to measure loyalty? Because the past will show the future? Does it? Will it? In your company, in your business? Do you want to know how many will respond to your next marketing scheme? Or who should you be extra nice to? Or who’ll form the 20 of your 80:20 four years from now?

Do you really run your business so systematically that you can obtain, leave along use, information of that nature? Can anyone? An astronomer may be able to predict how the heavens will behave a million years from now? But ask him to do it for a business, and he’ll almost certainly recommend you see an astrologer instead.

We are not, of course, saying something as stupid as ‘Burn your plans,’ merely suggesting that the plans may have to depend more on observations, experience and ambition than on numbers. Perhaps they always do, and numbers are only put down to distract busybodies.

Finally, why do you believe that the data you have can predict loyalty?

Let’s say we define loyalty as ‘probability to dine again’ for a restaurant. We’d know the courses ordered the number of dinner companions, the time and date, the waiters, the special instructions, and what not.

But would we know the tip? What if it’s paid in cash (tips are taxable income) and the restaurant doesn’t have a system of recording and splitting tips? ‘Tip’ is supposed to be an abbreviation for 'to insure promptness'. When would the diner want this promptness? On a future dinner, obviously. So a good tip not only says, “Thank you,” it also says, “Keep it up, because I’ll need you again.” Isn’t it possible that the unrecorded tip would prove to be a better predictor than everything that is so carefully measured?

We should stop measuring then? Or wait till we have discovered that universal applicable and universally understandable clear measure? That’ll be like stopping breathing till we realise the meaning of life, wouldn’t it?

All we are suggesting is that debate and discussion be not always equated with intellectual masturbation. By that yardstick, let's-do-it measuring is numerical rape.

(In The Ultimate Question Fred Reichheld announces that the Net Promoters is this measure. Please see www.ultimatequestion.com for an introduction. And do let us know what you think. Mr Reichheld is the greatest guru on loyalty. Yet, it’d perhaps not be a sin to debate about him.)