11/12/06

Digg Canongate

Something interesting happened over the weekend: a fragment of a campaign I worked on 2.5 years ago got dugg (i.e. featured on Digg).

With (currently) 1552 diggs (individual votes for this as a cool thing online), “Names Of Things You Never Knew Had Names” has been one of the hottest things on the web for all of 24 hours. It’s really interesting.

First of all some backstory. Canongate (for whom I used to be sort-of retained to do online work) was bringing out a book called ‘The Book of Lists‘ which did exactly what it said on the tin. Full of insane trivia, stupid things and famous people - and very, very funny - I thought that it was perfect Google-fodder. My pitch was to build a lightweight site that highlighted some of this content (which was bound to get ripped off and put online anyway), and that the site would be built using search engine optimisation to get well-indexed by Google. Therefore, lists such as ‘6 Positions For Sexual Intercourse - In Order Of Popularity‘, or ‘7 Celebrity Couples Married Three Weeks Or Less‘ (both legitimate entries from the book, which was first written pre-Google) could end up driving a lot of traffic to the Canongate site. The pages would all contain a link to other lists (although the site would only offer up a tiny percentage of the book content - it would still be enough to whet the appetite) and to the book of course. It was a pretty basic strategy based on visibility, price, and try-before-you-buy.

It kind of worked. Well, actually, it did and it didn’t. It did in that Canongate’s Book Of Lists site (which was rolled into the main Canongate site that we were building at the same time) got really well indexed, and drives lots of traffic to the site. It still does.

It didn’t in that the book did well, but didn’t exactly go nuts through sales on the web site. But, to this day, if you google (ahem) ‘erect penis’ (which is a pretty popular search query, if my spam filter is anything to go by) then the book of lists site comes up on page two; for ‘sexual intercourse positions‘ (see above) it comes in ranked #6. Pretty amazing really. Less salacious searches are equally successful, just not so popular.

But this all happened years ago. Then, this weekend, someone dugg the ‘33 Names of Things You Never Knew Had Names’ and it all kicked off. The ‘digging’ has now leaked onto copycat sites, reddit, del.icio.us, fark, stumbleupon, and is right now beginning to bleed into the blogs. Probably next is a few other lists following a similar path.

So, Alex and I (we did the site together and he maintains it and monitors traffic, which is how I know all the above) spent some time in between other things, trying to work out how to (very quickly) convert the extra 100,000 people who were coming to this site, genuinely interested in what it hard to say (even if some of it was about Britney Spears’ marriage collapsing), into people who bought the book. Even being pessimistic you’d hope to convert 1/100,000 into a customer?

We came up with a few hacks this morning. All the action took place on saturday day time (PST) and thoroughly over the weekend GMT. But today we got Canongate to discount the title, to 40%, and we moved some of the navigation around a litte. This was because most of the content had, ahem, got a little dusty, and wasn’t so relevant any more, so we had another look at that. We also tried highlighting the content from other Books of Lists (they have one on London and one on football as well as a mini one). And we look at the logs and saw that whilst visitors were browsing the lists microsite, they weren’t going into the rest of the site.

We also figured out that if these visitors were, as the stats said, all coming from pretty low-brow sites such as fark.com and gorillamask.com (when I say low brow guys, I mean in a literary context: this is books, remember - although thinking about it, stet), and the vast majority were using Firefox - then the likelihood was that our demographic was a bunch of geeks. So we tried profiling Canongate’s most geek-friendly books as well, to try to get them to get into the main body of the site. It’s been live for a few hours, and you can still hear the sound of the stable door bolting. But it has been an interesting example - and something that people would pay pretty huge amounts of money to have come their way.

It’s a pretty amazing phenomenon, high-volume traffic. Some people talk about being SlashDotted - where their sites get swamped by huge amounts of traffic after being featured on slashdot.org. It’s like being visited by a swarm of bees, migrating birds, or school of fish. They come, they go - and in this case, they don’t seem to have been interested enough to part with their cash.

It’s a tough moment as a marketer to realise that the strategy didn’t work. I mean, it did in that it drove traffic to the site, but it didn’t in that they didn’t buy. There’s all sorts of reasons / excuses - like the fact that we only caught the tail end of the curve, and that 80,000 visitors came to the site when the book wasn’t discounted, and the navigation was below the fold, and all the rest of it. But it is a little sad to admit that we didn’t sell any books on the back of that many people coming to the site.

Do I maintain that had the book been discounted when the bulk came, then it would have sold? Absolutely, although not in huge amounts.

I guess that the real lessons to take away from this are:

# Traffic is only valuable traffic if it’s the traffic you want. (This goes back to the days of eyeball-driven ad rates and ‘Britney Spears’ meta tags to try to drive traffic) It’s also why Google, in terms of ads but most importantly search / find is *such* a killer invention.

# Always have your ducks lined up. Had the content been fresh, the links to other titles kept up-to-date, and the navigation reconsidered - and the discounts in place - then the site should have sold more.

# Don’t forget that things like this creep up on you. In the world of the web it’s so easy to think that it’s all about the new. This wasn’t about the new. This was about people finding the old - and digging it.

# It’s all about content. Which is still king.

# Finally, Canongate does a great and very respectable trade in direct sales to customers who know that buying there offers benefits over the high street. It has some loyal customers, and people who take a punt on the titles they publish. The Book of Lists is exactly a part of this kind of relationship, so it was the right thing to do. Had it been a bit of Britney-bashing for fark readers, it would not have been. Know thyself - and thy audience.

Posted by Peter Collingridge in Publishing, Web.