First, what we might call the ‘good news’: as social media marketers have been telling us for the last five years, more useful, fun, non-pushy content is more likely to engage people than straightforward advertising.

The reports says the sponsored content was a St Patrick’s Day page containing a video and UGC. The video mentioned the brand sponsoring the content. ‘Give’ widgets let you create a present – like a badge or a greeting card – for friends. ‘Get’ widgets let you create similar things for your own profile page.

But there is a significant caveat here. Yes, people said they’d click links on sponsored content and give/get widgets to a greater extent than on banners, newsletters and non-interactive brand pages. But not by as much as you’d think. The ‘likely to click’ score for sponsored content is 3.3; for old fashioned banners it’s 2.8. That’s a 10% difference in impact. I suppose you might argue that it all depends on how good the content is, and how relevant it is to the site’s users: I can’t really comment on that since I haven’t seen the media used.

Sponsored content is a good thing, then, as far as getting people to click through is concerned. It’s good for awareness and improving sentiment. Unfortunately, there’s absolutely no correlation between this and persuading people to buy things.

Sponsored content was found to be the least successful in terms of converting exposure into sales. I’d imagine that this is probably down to context and the lack of any particular call to action. If you load up a flash game – some sort of Farmville knock-off – and it happens to be sponsored by Corona beer, I’d wager you might think: ‘oh yes, Corona beer – that exists’. If it’s a good game, then you might think: ‘Fair play to you, Corona. Nice one.’ What you probably don’t think is ‘Right, I’m off down to the shop to buy some Corona’. [Actually, that might have been a bad example. Mmm… beer.]

Better targeted, relevant sponsored content would presumably work better: a good example is the free recipe cards that supermarkets give away. If I worked for a supermarket, I’d be all over allrecipes.com with my free recipes, but not with a car-racing game.

The big winner for brands is having a profile page with fans. The ‘with fans’ difference is that fan pages give users a badge that shows on their own profile. Straight brand pages without fans are just there to look at, and are not so successful. The commitment – however slight it may seem – of publically saying that you like a brand turns out to be a fairly strong motivator to buy things from them. The report’s authors suggest that this is because people hate being seen as inconsistent, or displaying cognitive dissonance, science fans. If you’ve joined the Marmite fan page and then buy Vegemite, then that’s odd and wrong, even to you:

…once people purchase products from a brand, they report liking the brand more. But the reverse is also true – when people declare publicly that they like a brand (by putting a logo on their profile for all of their friends to see) they are more likely to buy from it.

This also explains why ‘Give’ widgets work better than ‘Get’ widgets. If you send your friend a virtual pot of Marmite, then that’s a much more public display of affinity than making one for yourself, so you’re more likely to stick to your professed tastes.

The full report is here [PDF]. via. MediaPost

picture credit: CaptPiper

On most news organisations’ websites, you’ll find a widget called ‘most read’, ‘most shared’ or ‘most commented’, possibly all three. The Guardian’s Zeitgeist experiment suggests an interesting alternative.

Typically, the content found in the most-X sections provides a salutary – if depressing – reminder of humanity’s baseness and stupidity. What tends to get flagged is not ‘Picasso retrospective opens at the ICA’ or ‘Proposed Amendments to Digital Economy Bill’: it’s ‘footballer shags team-mate’s wife’. If you’re seeking the Wisdom of Crowds, look away now.

Here’s the latest from the BBC:

Even worse is the equivalent list from the Telegraph:

Not to mention the Daily Mail:

Oh dear, oh dear. Showbiz, trivia, sport, sex and weirdness. And these aren’t tabloid publications. The Telegraph, in particular, paints itself as a serious business and politics paper with a concern for moral values. Its readers, on the other hand, appear to prefer sex scandals and weird animals. I can’t imagine its editors are especially proud of these results but ultimately have to shrug and be grateful for the extra page-views.

The Guardian has a similar widget, which isn’t as lowlife as the examples above, but again favours the funny and the odd.

Newspapers and news organisations are in a strange position with regard to these most-popular lists. The short-term value is that they flag up the items that new visitors are most likely to click on and enjoy. They get more page views out of their visitors and thus more advertising inventory to sell. They help the organisation bolster their claims to advertisers that their sites are busy and popular. Readers get what they want quickly and leave happily.

On the other hand. There’s a long term devaluation coming out of this for serious papers. When they sell to advertisers, they aren’t just selling so-many million eyeballs much of the time. They’re selling a certain quality of readership and particular brand values. For readers, there’s a similar brand attachment. They go to a serious news site because they trust the brand and want serious coverage. If they then end up then clicking on the story about a funny-looking gorilla, then that’s their own affair. Maybe, rationally, they should have gone to weirdanimalpix.com, but they don’t see themselves as the sort of person who does that.

What’s more. Papers don’t really have an ad-inventory problem. They generate thousands of new pages and hundreds of thousands of impressions a day and rarely sell more than 20% of what they have to offer. The only real reason for driving page views is the arms-war between the Nationals over who is the most popular. And being the most popular isn’t a great argument to advertisers if you are simultaneously claiming that your readership represents an elite, as is likely for any serious news site.

So maybe it’s a good idea to find a middle-ground; a way for serious news organisations’ websites to highlight popular items that doesn’t make them look like a zoo for morons: for readers or advertisers. The Guardian’s Zeitgeist – launched today – is one attempt to find that middle ground.

The idea is that it blends populism and curation. The most popular stories will appear on the grid, as you’d expect, BUT:

• The different sections of the site – news, features, opinion, sport, etc. – remain balanced in the proportions conceived by the editors. So if 90% of its visitors are looking at Sports stories, it still only occupies 2-3 slots on the grid.
• Like is compared with like. For example, Charlie Brooker’s satirical swipes at popular media are perennially popular on the site, but will only hit the grid if a particular column is more popular than the norm.

Guardian communities editor Meg Pickard explains:

…we’re analysing and combining all sorts of things; where people come from, where they go to next, how long they stay on a particular page, if the page is getting passed round twitter and other social websites, number (and rate) of comments and so on.

We’re taking a range of these variables – enough that a single datapoint doesn’t skew the results – and mushing (that’s the technical term) them all together to get a value of “Zeitgeistiness” (another technical term) for each content object.

But – and this is the important bit – each content object only gets compared to other items in the same section, which in real terms means that Football articles only get compared to other Football articles, Technology blogposts against other Technology blogposts and so on. In fact, we go one step further, and take the type of article and day of week into consideration: an Environment gallery on a Monday only gets compared to others of the same type/section also published on Mondays. Because we’ve been storing and analysing this data overnight for a while now, we’ve got a good baseline to work from.

It’s early days for the Zeitgeist experiment, and I’m afraid it’s rather buried away from most visitors to the site, so it will be hard for them to see how popular the idea plays out compared to the regular ‘most-read/commented/shared’ widget. Nonetheless, it’s an interesting project that shows how news organisations might protect their brand at the same time as playing to the cheap seats.

picture credit: Joi

I think it really helps readers to get the most out of your writing if they can diddle about with an IP address lookup tool as they digest your latest outpouring. Nothing says ’sticky’ to me like a page that includes a local weather radar.

There are even plans for a Widget Conference. Apparently (and this is a complete lie), it is to be held in the car park of the San Francisco Public Library…