I’ve been playing around with Trendrr, the online trends tracking service. This has been around for a while, but has just shifted to a ‘Freemium’ model. Think of it as Google Trends on steroids. The sources of trends tend to focus on web chatter (Twitter, Delicious, Flickr etc.), and the featured categories tend to focus on buzz categories like media, music and politics. However, I think it’s most interesting when you use actual transactional data (Amazon, Craigslist and Ebay).

The above example is taken from one of their showcase pages and (gratifyingly) shows response to three articulations of Ogilvy’s campaign for Cisco. You can see that the term “the collaboration effect” generated the most web traffic. What I’ve learned about using these sorts of tools is that they are most interesting when comparing two very disparate sources, something that you can’t do with Google Trends. Try Ebay searches against Tweets of a certain product, for example. It’s not the trend itself that’s interesting (they’re very hard to judge in isolation) but the relationship between two or more trends that happen at the same time.

This morning I listened to a NPR interview with Anthony Wesley, the Australian amateur astronomer who noticed a black spot the size of the Pacific on Jupiter at the weekend (probably made by a comet!). He made the point that the world of professional astronomy is now aided by an “army” of amateurs who are using equipment that only major observatories had twenty years ago. The same is true of trend watching. Amateurs now have the same equipment as professionals. We should look forward to similar discoveries.

In the last decade a lot of research has gone into how fads spread.  Terms like “viral marketing”, “influencers” and “seeding” have been coined to describe different theories on how to marketers can encourage the spread of trends.

Very little, however, has been written about how trends die.

Until now.  A recent paper by Jonah Berger of Wharton and Gael Le Mens of Stanford looks at how quickly trends die.  Ingeniously, they’ve looked at over one hundred years of French and American census data to examine the rise and decline of popular baby names. 

The graph above illustrates this neatly.  Charlene took decades to become popular and took decades to fade.  Tricia climbed quickly and declined quickly

They found that names that become popular quickly also lose popularity quickly.  This is because the very suddenness in popularity is perceived negatively.  No one, it seems, wants a trendy baby name. 

A teacher friend of mine talks about the ‘strata’ of kids names in her school.  Ten to five years ago biblical names, such as  Zachary, Benjamin were the thing.  (My kids, born in this era, conform to this rule, of course.)  More recently, forties era names have been popular:  Jackson, Ruby and so on. 

Most importantly for marketers, Berger and Le Mens proved that, “in contrast to conventional wisdom, identity-relevant cultural products which are adopted quickly tend to be less successful overall (i.e., reduced cumulative adoption)”.  In other words, fads sell less overall. 

This is very important.  It suggests that brand launches shouldn’t necessarily start with a bang.  New products may have to feel that they’re going to be around for a long time to be attractive and your awareness of them may have to build much more slowly than we’d previously thought.