The Drum Awards Festival - Official Deadline

-d -h -min -sec


Understanding the 'fuzzy data' of performance marketing and e-commerce

By Kathy Heslop

February 22, 2013 | 3 min read

Those who were in America when George W. Bush described Al Gore’s economic figures as “fuzzy math,” will remember how whilst he kept repeating the phrase, he never once discussed Gore’s economic reasoning; a rather disconcerting situation, especially when framed in the context of his claim that as President, he would be a leader of educational reform.

William Thurston a professor of mathematics at the University of California subsequently quipped:

“I gradually came to understand that by “fuzzy math,” Mr. Bush meant, “Math is confusing and fuzzy, so ignore it.”

These days in the world of performance marketing and e-commerce, unfortunately we sometimes see “fuzzy data.” Data represented by stories and anecdotes because they are much easier to understand. It’s a form of 'cooking the books' to overstate incomplete or overly reductionist data, a topic we already touched upon in our February 7th blog: The Data-ing Game

“‘…Keeping it simple stupid,’ reduces the dimensionality, it focuses on the noise, rather than on the signals, which could also encourage people to dangerously shoe-horn rules into data, to support the outcome they want to see.”

However, the term ‘fuzzy data’ does actually exist and it relates to soft computing. This differs from conventional (hard) computing in that it is tolerant of imprecision, uncertainty and approximation.

“Fuzzy Sets play a significant role in Data Mining. It is widely recognized that many real world relations are intrinsically fuzzy. Fuzzy Sets and Fuzzy Logic are also considered as a need to represent the inherent non random uncertainty which lies in most part of information and decision processes,” explains Miguel Delgado, of the Berkeley Initiative in Soft Computing.

Analysing data involves complex and specialist skills, and also comes with implicit responsibility. Research integrity is key in how raw data is chosen, evaluated and interpreted into meaningful and significant conclusions that other researchers and the public can understand and use.

So our advice when confronted with a trade-off between “noise” rather than signals?

You do the math.


More from Data

View all


Industry insights

View all
Add your own content +