Real Data Mining

I don’t know where Marissa Mayer read that, apparently…

credit card companies know with 98% accuracy two years before that you’re going to get divorced.

But I see a business model here. We should put real data mining to action. What if you and your boyfriend or girlfriend could submit your credit card purchases before you got married and could get an educated guess from them about the chances – how much more or how much less than the standard 50%? – of your future marriage falling apart?

The Cluetrain Manifesto in 2010

A few years after it was written, it seemed like the Cluetrain Manifesto really had it right. Back in, say, 2001, advertising on the Internet was broken. Banner ads didn’t work, tracking was little more than guessing, and buying ads was a time consuming thing which involved negotiating with different partners, dealing with different creativities and different ad-servers etc. It really looked like the web was a different ballgame, and companies better start doing something different.

Then Google came to the rescue. Pretty soon, you could buy ads next to search results, and then contextual ads, and then graphical ads, all from one single interface, one single check, no negotiations and for a price which seemed to make sense. For those who didn’t get it, or didn’t want to get their hands too dirty, search ad agencies were there to take care of their budgets. And just like nobody had ever been fired for doing Tv ads, the same exact thing was going to be true for ads on Google.

For most companies, “marketing” was once again something which had little to do with their product. Or with their market. But with the success of YouTube, Facebook, and Twitter, came a host of consultants that told companies that “in social media” they had to be cool, hip and down to earth. But there was no need to panic, or to change their ways, for they could do it for them, and provide them with stats about how things were going, and reassure them that there is nothing to worry about…

What Twitter Really Is

Ever ask yourself what Twitter really is? Twitter is a distributed link-sharing system – much like reddit or Digg are centralised link-sharing systems. Sure, sometimes people use those 140 characters to talk about something without linking to anything, but for the most part it’s links that keep Twitter alive. In this regard, I would go as far as to say that Twitter is somehow more similar to trackbacks (links) than to blogs (and content) themselves. Delicious, while we’re at it, is both distributed (my page) and centralised (the homepage). But neither of these two aspects was really used to share links, which might explain why Delicious slowly faded away as Twitter started to soar…

Your company’s urge to tweet

is wrong.

Have you not noticed? Every time something new comes up – be it newsgroups, the commercial web, portals, blogs, facebook, twitter, whatever – your company wants to do ever more talking. And all it has to say is “we’re the market leaders” and, in the end, “buy our product”. How interesting.

So, more likely than not, you spammed newsgroups, created uninteresting brochure-ware websites when the commercial web came around, then bought banner ads like there were no tomorrow, perhaps even created a lame corporate-blog, as if you were not producing enough stupid pr.

All this is so wrong and so stupid. It seems to misunderstand the fundamental change we have witnessed: that on the web – not just on so-called “social media” – everybody can talk. All you need to do is be interesting, and then get out of the way, and your customers will do the talking for you.

So here come Facebook and Twitter, and now you want to have a fan page, and friends, and followers. You would never dare say you need to change that lame corporate site of yours, but on “web2.0” you want to be perceived as hip and friendly and down to earth. I know, it’s a lot more fun.

And so you push hard until you are allowed to launch “a twitter initiative” and “a strategy for facebook”. You bring in funny-dressed “web2.0” types as consultants, then you set up “a plan” and “goals”, and carefully determine who should be twittering and what he/she should and should not say.

Sometimes, you even totally outsource your “social media” activities to some “web2.0 agency” that will do the talking for you and try to make you come out as web-savvy and friendly. And when you reach a couple thousand friends or followers, you declare it a success. Mission accomplished. Yeah.

A couple thousand people willing to receive – not necessarily read – the 140 characters messages coming out from Big Corp. to its millions of customers is considered “a success”. Wow. Gotta love this “web2.0” thing. No need to change anything inside the company, and it’s impossible to fail.

And you get a lot of press, too, and you can go on and say in your CV that you took care of Big Corp.’s “presence on twitter”. How different is it from what online marketing managers were doing with Second Life a couple of years ago? Does it really add any value to the company? I doubt it.

In the meantime, beneath the friendly “web2.0” pose, life goes on as usual at Big Corp. Pr flacks push hard to sell their story to newspapers, and perhaps put ad spending on hold if the paper does not comply; negative blog posts are more often than not dealt with by lawyers; and nobody inside the company gets to do any talking except those who have long forgotten how to speak in a human voice. But you have a tagcloud on your website, and a thousand fans on Facebook, and followers on Twitter. Life is good, and there’s no need to change. No need to change Big Corp’s company culture; no need to confront the market; no need to be open and down to earth and accessible; no need to fearlessly show on your website what is being said about you online on blogs and Twitter and no need to let your employees take part in the conversation; no need to be interesting and friendly so that thousands of people will happily spread the good news about you to thousands of their real friends and followers. With a nice “web2.0” mask on you can fool yourself that there’s no need (for now) to face up to the fact that Obama, not Eisenhower, is now President and we’re not in 1950 anymore.