In the early 1950s, American coffee roasters faced a dilemma. The price of coffee beans had risen dramatically, and they were faced with two distasteful choices: either absorb the increased price, partially or totally, hurting profitability; or pass the cost on to their customers and risk losing market share or having customers switch to other beverages.Isn't it great! It reminds me of the Fast Food Fallacy from Jerry Weinberg's The Secrets of Computing:
They came up with an innovative alternative. The coffee roasters' business consists of buying, ageing, roasting, and blending coffee beans to achieve the desired state and smell. Coffee beans, like all agricultural commodities, are highly variable. Two beans can be quite distinct. Even beans picked at the same time from the same tree can be different. A bean picked from the top of the tree, which receives more sunshine, tastes different than a bean from the bottom of the tree.
Blending is a critical part of the process. The leading roaster of the time tried experimenting with different formulations. It found that gradually changing the formulation, substituting lower quality beans, was unnoticeable to the customer. It begins to slowly change the blend. Every two weeks a few more of the less expensive beans were substituted for the heartier, more expensive ones. Most consumers couldn't notice the difference in coffees bought two weeks apart. But if they had tasted, side by side, two batches made six weeks apart, they would have noticed a slight difference.
In effect, the roaster started training customers to accept an inferior blend of coffee. The other roasters noticed what was going on and responded in kind to avoid losing market share. In a few years the American consumer's standards for a decent cup of coffee were radically altered. The managers did their jobs and enjoyed their bonuses. At the time this response was viewed as a triumph of ingenuity. But a funny thing began to happen. Per capita consumption of coffee began a slow but steady decline. The business stopped growing. The roaster that started it all began to experience profitability problems. It is now part of a huge conglomerate and is still experiencing problems in its coffee business. Consumers have discovered gourmet coffees. But ironically, the coffee drunk by Americans during the 1940s was on a par with what we now call gourmet coffees.
No difference plus no difference plus no difference plus ... eventually equals a clear differenceI emailed the story to Jerry and he replied:
Great story. Even though I've never had a cup of coffee in my life, I can appreciate the dynamic. Indeed, the same dynamic has occurred over and over. I've written about white bread's deterioration - and now we have "gourmet" breads.
And, I suspect, the same thing is now happening in software - gradually lowering the quality of apps, as expectations lower. Then we will have software selling on the basis of freedom from those annoying bugs. Same thing in security, I think.