Click by Bill Tancer

Review by Donald Clark

Most books on the internet look at it from the outside, but Trancer is an expert in inside knowledge – “search data”. He harvests data on what people search for to predict trends. He uses searches on porn, pills, politics, dresses, bands, real estate and anything that takes his fancy. He’s the search king and we’re his subjects.

Porn

Sex sells, but our porn habits on the web are sometimes surprising. It’s a fragmented landscape with 500 sites accounting for 56% of the traffic, but did you know that we spend 6.5 minutes on average on a porn site, with Friday night being the peak viewing time. A surprising 27.4% of visitors are women, especially pornographic fiction sites. You may complain about Viagra spam, but it’s the third most searched for drug on the web, and searches peak in January/February, presumably when we need a little uplift in our lives.

Seasonal searches for diets in January are matched by ‘false hope syndrome’ and a rapid fall off in interest as we clearly fail to reach our goals. Our fears also become obvious with the top searches being intimacy, rejection, people, crowds, failure, sex, commitment, public speaking, being alone, love, girls, abandonment and a broken heart, all exposing our human frailties.

How to…

A big riser on internet stats is ‘how to’ searches, with ‘how to tie a tie’ at the top of the US charts, but not in the UK, where it’s only 51 (we wear ties as part of our school uniform). The top ‘how to’ searches are task-specific, namely, how to tie a tie, do a Rubik’s cube, make a movie etc. Second comes sex, and you really don’t want to know what the full list of 173 sex searches are. Then there’s the aspirational stuff, losing weight, making money etc. It reveals what we really want to know, as opposed to what education thinks we want to know. Strangely enough, the top ‘why’ question, has been ‘Why is the sky blue?’ and has been for the last two years (search me).

Politicians are using search data to identify ‘hot’ political topics. In the US, abortion was a clear winner in political search but is now being replaced by national debt. It was this type of data analysis that was used by all camps in the recent US election, leading to the first ever e-President.

Participation inequality (1:9:90)

The chapter on web 2.0 searches reveals that the data shown a 1:9:90 rule. 1% of visitors to a web 2.0 site actually up-load or create content, 9% interact by editing or commenting and 90% are lurkers. This is not surprising but neither should we be much concerned. Lurkers are learners. With Wikipedia it is largely the old teaching the young, with an older demographic editing and a younger demographic reading.

I click therefore I am

Tancer uses data to provide trends analysis to financial investors and companies as it’s often far quicker and more reliable than traditional marketing surveys. But one has to be careful. The most obvious data set is often misleading. After some success in predicting the winners of reality TV shows, he made a mistake with a female wrestler. Turns out most of the searches were for nude pictures of her on porn sites, nothing to do with the voting demographic, so her main rival won. He lost the bet on that one. His predictions on house market prices were also flawed until he finessed the data. In the end, however, this is a powerful analysis tool, used increasingly by forecasters.

Lessons learnt Google Analytics and Google trends are typical of the powerful and free tools one can use to do a spot of your own trend spotting, or to identify progress, or problems, with your own brand, product or company. The data is pretty direct, as people search for what they want. That’s not to say it’s wholly reliable, as advertising can drive searches. However, this book is different from most speculative web books such as Leadbetter or Shirky, in being data-driven. It looks at the web from the user’s point of view. This makes it a welcome addition to sometimes predictable books about the web. One problem is that it is very US-centric, but not always, and Tancer is at pains to point out dramatic national differences, but the core examples of prom dresses and Obama are a bit too ‘yankee doodle dandy’ for me. I found the analysis of ‘prom dresses’ rather depressing and dreary. I’d have preferred him to have chosen more international, and less culturally-specific examples. But don’t let this put you off, reading this book will deepen your understanding of the web.