Friday 28 March, 2008

Should loyalty programmes pay interest?

In all the loyalty programmes I know of points have a constant value until they lapse, when they lose all value. But if one of the avowed aims of loyalty programmes is to keep customers loyal, shouldn't it be somewhat different? Shouldn't points value go up wit time? After all, it's the customers' deferred discount lying with the company.

Ok, that's ridiculous.

However, a time-linked valuation may be an idea whose time has come. It may go like this: Points have 150% value if redeemed within 30 days of their accrual; 125% if redeemed between 1 and 3 months; 105% if used between a quarter and a year; 100% thereafter, till they lapse. (It works like a bank account standing on its head.) There may be hikes in value around customers' birthdays and anniversaries, etc.

Communicating these value changes and deadlines will give marketers excuses to talk about other things too. The whole intention is to make recency work for you.

The technology is surely there by now. Is anyone using this system or variable valuation?

Zero

“Didn't pull anything at all.”

“Hardly any conversions... in single figures. The whole lead generation exercise was a waste.”

“Zero. Ziltch.”

You don't hear these too often, but you hear them often enough to worry. Because word spreads, and soon enough the whole industry is being tarred with the same brush: Direct marketing doesn't work. Lead generation is BS. Cold calling is the only way.

It's a mystery if something, anything, doesn't work at all. Probability loathes unmitigated disasters.

Let's think of an financial services company that agrees to pay, say, Rs 100 per lead. No sales manager would agree to such an amount unless he's quite sure that he will be able to convert a substantial number of these leads. We don't have to stretch credibility to envisage a binomial distribution with p = 10%, that is, there is a 10% probability that a lead randomly picked from the bought list would convert.

With such a probability what should one expect from a list of 2,500 leads, where a lead is defined as someone who explicitly expresses interest in a particular product of a particular brand by filling a form, and asking the company to get in touch with him?

The number of trials and mean are large enough to apply the normal approximation. And this says that there's a 99% chance that one should make between 288 and 211 converts.

The probability of making less than a hundred conversions is... zero, negligible.

“10% is too high,” you'd say. Let's try 5% for argument's sake (remember, it's Rs 100 a lead).

Even after halving the probability of success, we retain a 95% chance of making 103 to 146 sales. The probability of converts staying within double figures is 1.09%. Again, negligible.

So what do we tell the sales manger when he complains that leads were all duds? Logically, you should tell him that his lead management system doesn't exist: It's a wonder that his company does.

In real life, you bow your head and watch him renegotiate the rate, reducing it by 99%. Because theory be damned, he's god.

Numerology

I'm reading E-mail Marketing by Tim Beadle. On page 27 it says, “Research has shown that the optimum length for copy is around 200 words or less. Beyond 200 words, response rates for the SAME offer decline. That does NOT mean this is right for you – test it, try 100, 200 and 300 and see which pulls best.”

Even with the qualifier, I abhor this type of data. The more I think about it, the more harm such 'research findings' seem to have done to direct marketing. No summery is provided, and no source is sited. We don't know how many tests were carried out, across how many brands, in how many product categories. The author is silent about the range of response rates. He doesn't, of course, tell us how much the responses fell by.

One wonders why he quoted the figure at all, except to give an illiterate client to impose a counterproductive and baseless restriction on work. Or enable an equally illiterate agency person to fill the auditory vacuum during a meeting with numerical - numerological? - basalt.

Thursday 13 March, 2008

Which explains?

From the article A child and divorced by Rohit Parihar in India Today (March 7, 2008): “Around 1.6 crore child marriages continue to take place in parts of India... By far the largest numbers are from a caste-ridden Rajasthan where children as young as four, five or even less are married off or traded like so many cattle.

According to the 2001 census, of the 1.6 crore children in India who got married before they reach official marriageable age, 18.3 lakh came from Rajasthan alone.

Which also explains why of the estimated 1.7 lakh divorces that take place before the children reach marriageable age, the largest number, around 6,200, are from Rajasthan. (my emphasis)”

Does the preceding data explain why 6,200 of the divorces came from Rajasthan? I'm afraid it doesn't quite. The state's share of child marriages (11.44%) is more than three times its share of child divorces (3.65%).

And the writer does not give any reason why we should expect a similarity in ranks, that is, why Rajasthan's being No 1 in marriages should lead it being No 1 in divorces.

As a matter of fact, the difference in shares (of marriages and divorces) points to an even worse story lurking elsewhere.

More follows a few paragraphs later: “A survey undertaken for India Today by NGO Prayas ... one of the 3.7 lakh children in her state and 20 lakh across the country who got married before the age of 14 — too young to even remember the ceremony, if there was one.

And like 50,000 others in the country and 1,200 in the state under the age of 15, she got separated from her so-called husband as swiftly as she had entered into wedlock with him.”
The first figure pertains to children below 14; the second to children below 15. In that stage of life, one year may make a great difference. Shouldn't both figures have had the same age cutoff?

Assuming we ignore the one year difference, what do we get? This time, Rajasthan's share of (below 15) marriages (18.50%) is more than 7½ times its share of (below 14) divorces (2.40%)! To be fair, the author doesn't claim any causal link between the marriage and divorce figures here. Yet the numbers neighbour each other, so one can't be blamed for suspecting a connection was hinted at.

I am guessing, without reason, that the later set of figures (marriages below 15 & divorces below 14) came from the survey. If so, one cannot but notice the differences between them and the census figures. For example, the share of marriages goes up (from census to survey) by about 7% (perhaps indicating that Rajasthan's child couples are, on an average, younger than other states' – please see below).

The census figures have no age-cutoffs, which should partially explain the differences. Besides, the census came out in 2001, while the survey was, presumably, more recent. Nonetheless, one wishes the magazine had published the survey, or, at least, excerpts from it on the Web.

My main problem with the article though is that it describes a problem without mentioning any efforts at its solution. For the children's sake, I hope that is the author's worst oversight.

Tuesday 4 March, 2008

Immigrants and locals

A commentator observed that immigration is a non-issue in Bombay because the fraction of immigrants in Bombay's population has come down from (about) two-third to two-fifth over the last four decades. It didn't sound right. I mean the non-issue bit. So I put in the figures in Solver, and asked it what sort of birth, death, new immigration and immigration from Bombay figures can co-exsist with this drop.

It gave a quite realistic set. People had lots of children. Many new immigrants came. Some of the original locals and immigrants died or moved out of Bombay. Yet the fraction of immigrants went down. How? Simple, I counted the children of immigrants as locals.

While the figures were certainly realistic, they were picked out of the air. So I haven't reproduced them.

But I suppose my point is a valid one. One shouldn't ask people to draw any conclusion from a statistic unless one explains how the figure was reached. Immigration is an issue. To argue otherwise would be dangerous.

Rich and famous

There are many books and shows on money, far fewer on fame. Why? More people want money than want fame? Or because money is a more basic need?

Or is there an opportunity staring at us? For how-to material on becoming famous (I don't mean PR related matter here, of which there must be plenty). Or a history of fame. Or both.

Which leads to the (somewhat vague) observation that there are more famous people in the West than over here. That's not just about mere numbers, it's also about density.

A monetary analogy would explain the point. Not only have the West more rich people, they also have higher per capita incomes and wealth. More people per 1,000 are well-off, and so on.

So with fame. There is less to spread around here. Films, cricket, and politics hog the limelight, and here too only a handful of names take the lions' shares. The same names keep popping up all the time.

One hears of the FORTUNE 500, but here only five or six names - Tata, Birla, Ambani, Mittal, IT - predominate the public's minds. The ad world boils down to Piyush Pande and Prashoon Joshi. Yoga means Baba Ramdev. All of science and engineering are equated with Dr Kalaam. In every category of product, we have far fewer brands to choose from than has the West.

Does this have to do with lack of opportunity here, which allows only a handful to stand out (assuming you agree with my basic observation that there are fewer famous people around here)? Do we try less hard? Or does it have more to do with the underdevelopment of niche and specialised media in India? Or do Indians have less time to spare, and less time to make fellow Indians famous.

Perchance, one can define a 'celebrity capital' and study its distribution (we appear to invest ours in fewer people, from fewer areas). I'm sure someone must have worked out the equations, or, at least, tried to. Who?