Instant demographics. Just add volume.
Marketers do not care much about population studies, unlike academics. After all, the target audience for most marketing activity is usually a segment, rather than the entire population. Provided data can be found that picks out the right group, all is well.
But when the population is undergoing dramatic and rapid change, how can marketers be certain that the segment they want to target exists, is viable and has not been superceded by another, potentially more profitable group? Backward looking segmentations are getting out of line with current demographics.
Even the Government is having trouble working out the real nature of the population. It has just announced via ONS that unemployment has reached 2.5 million. Yet claimant numbers for benefits have not been climbing as rapidly, especially among young adults. This gap shows that direct indicators and research-based samples are often out of line with each other.
Banner Ad
Marketing is in precisely the same situation. New generation data sources have been giving the impression that it is possible to keep track of demographic changes. Online surveying in particular, combined with lead generation, make it seem as if target groups are being clearly defined and reached.
The key question is how you know whether you are penetrating that group deeply enough if you have no idea of its true size. Universal data segmentations, such as Experian's Mosaic, are recognising just how fast-moving the population currently is by rebuilding every six months.
Out of this complex data situation, a new insight practice might be emerging. Its starting point is to assume that no universal data set will offer either 100 per cent coverage or be 100 per cent accurate. Instead, confidence scores will be assigned that reflect how widely and deeply the data owner has been able to spread its net.
Marketers will need to get used to working with relative, rather than absolute knowledge of their markets. From that starting point, however, confidence can be built by running online surveys. These indicate whether assumptions about the interests, scale and value of any group are worth expending marketing effort on.
The last decade has seen a tremendous increase in the ability of analysts to model variables or classifications from small group data sets onto much larger ones with a high degree of accuracy. Provided this approach is carried out in a unified way, without building in assumptions from the start, it can be highly effective.
The result can be an instant fix on the market that is applicable in a volume manner, without being locked in to static population profiles. What is notable is that the Government will be adopting something similar in the wake of the 2011 Census. Knowledge transfer between the commercial and public sectors could prove to be of tremendous benefit to both.
| Bookmark with: |