The vast volumes of data on consumer behaviour available today mean that marketers now have more powerful tools at their disposal than at any time in the short history of the industry. It is baffling, then, that instead of segmenting its target markets down to the finest nuances of preference, demand and activity, the marketing industry insists on creating the bluntest instruments possible.The worst examples are to be found in generational theory, which originated with the so-called baby boom generation born after World War 2. Next came Generation X, supposedly those born in the late '70s and early '80s, and Generation Y, which hogged the rest of the century.The fundamental problem with baby boomers as a marketing category is that it is a purely Western construct, inspired by the increased birth rate in North America and Europe from 1946 to 1964.story_article_left1The boom was irrelevant to Africa. As a result, we've been spared most of the excesses of the baby boomer marketing boom.However, we are now being subjected to something even worse: the millennials. These, apparently, are people born from 1980 to 2000. Once again, it is a US construct. But, since it is not linked to a specific population trend, it is readily appropriated wherever marketers go to market.And this is where marketing strategy is truly baffling. The age curve of millennials runs from 16 to 36. Yet they are all being lumped into one category, with one set of behavioural characteristics. They all, the 16- and 36-year-olds alike, have the same requirements of the work environment, smartphones and social causes.Even a cursory glance at data on narrower age groups in this segment reveals massive differences. Demand or usage for specific products and services rises and falls in direct proportion to age - and that refers to every single year, not every double-decade.For example, with any given new social media network or service, the take-up curve is indirectly proportional to age. Any online transactional activity, on the other hand, rises up the curve, along with disposal income, into the 30s, and then descends steadily beyond retirement age. The fact that the curve rises from 16 to 36 means there is a vast difference between the transactional activity of the 16-year-old and the 36-year-old.story_article_right2It also means one cannot take a 10- or 20-year range as being representative of a group, or providing insight into how to market to different age groups. In many cases, closer analysis will show that the best level of accuracy one can achieve is within a two- to three-year cohort.This means that one would need to compare, say, the groups 20-22, 23-25, 26-28, and 28-30. Even within those groups, one would need to segment by education, income and even opportunity levels.It is astonishing, then, that the fiction of the millennial still exists. How a marketer imagines that a 16-year-old exhibits the same spending or digital usage characteristics as a 36-year-old is as baffling as it is revealing - revealing of how important labels and stereotypes are in the world of traditional marketing.It is little wonder that many traditional marketers cannot get to grips with the nuances of social media, where a five-year age gap can make all the difference between who uses Snapchat and who doesn't, and who uses HouseParty and who hasn't even heard of it.Goldstuck is the founder of World Wide Worx and editor-in-chief of Gadget.co.za. Follow him on Twitter and Instagram @art2gee..