Mark Higginson

How people's attention flows on the web

The web is a social artifact.
Here is a selected collection of related items:


17 April 2014

… we’re seeing a trend of these accounts supporting a cluster of records that other accounts support, and no account activity before or after this immediate activity.

Has SoundCloud Created Their Own Bots to Sway Numbers on Selected Accounts?

The linked post highlights suspicious accounts that follow, like, play and repost content on SoundCloud. Where high numbers are seen as ‘better’ it is inevitable that automated services will be developed to game these figures.

Veritasium pointed out how fake followers were a major issue on Facebook, especially when it seems that paying for reach, as Facebook are effectively forcing Page owners to do, exacerbates the problem.

Because these platforms are businesses, where success is based on the activity of their user-base, you are rapidly led down the rabbit-hole of wondering how complicit these organisations are and how hard they are really working at policing this activity. Or even if they are behind some of it.

Further reading: 61.5% of web traffic is not human

Let me know what you think on Twitter


01 April 2014

Business is political

OkCupid have weighed in on the Brendan Eich gay marriage issue by inserting a splash screen for Firefox users actively dissuading them from using this browser to access their website.

OkCupid Firefox splash screen

You could argue this is a fantastic piece of marketing by OkCupid. Their business is inclusivity and here they get to underline their values in opposition to those of Mozilla’s CEO. That said, I’m interested to know what the drop-off rate is for Firefox users clicking away from OkCupid as a result.

This stance is interesting as pretty much all businesses maintain an apolitical veneer. This is of course completely disingenuous as being in business is a political act, framed by and contributing to the dominant discourse.

Let me know what you think on Twitter


30 November 2013

To spawn the next Facebook or Twitter, first launch a start-up contest. Promise the winner a sizeable equity investment, say £1m, and let firms from all over the world compete.

Start me up

This is the culture that has ruined the social facet of the web. Becoming successful, in terms of attracting a large number of people to actively use your service regularly, is enormously difficult. Further to that, if you’re goal is to have a profitable business then you’re looking at achieving the near impossible. Of course, this isn’t the intent of most ‘start-ups’. The game is to flip the business; an expanding userbase gives the business a notional worth which may attract investment on the basis that yet more investment might be attracted at a later date with the potential for a sale to a larger business, or in the rarest of cases, a flotation.

The former motivation, that of a desire to provide a useful service, is to be lauded. The latter, one of profit, is to be condemned. Profit is an outcome of an excellent service, not an end in itself. The reason is that it isn’t the business per se and the mythlogical ‘founders’ who create the value but the users of the service. And it is the users who are often neglected as the business behind the service tries to attract investment or sell itself. Take the appropriately named ‘Sold’ as an example:

“… we wanted to create something that affected people in a positive way. Something they had an emotional connection with. Something they trusted.”

Something which they can now no longer use.

This is why my own level of trust in new services is very low; it is also why I blog here rather than the attractive looking Medium, whose future path to profit is a mystery but I suspect will somehow be at the expense of all those currently creating the value through writing and attracting readers to their posts.

“It’s not hard to make money on the internet… if what you build is popular, it will make money. The question is will it be popular. They’ll make money once we try, if we get popular — not that we’re necessarily going to wait to get popular to make money. There are a lot of ways to do that.” ### Evan Williams

How reassuring.

This post makes further interesting points: The Problem With Profitless Start-ups

Let me know what you think on Twitter


24 October 2013

Closing your eyes and wanting it to be true won’t make it work

I have been saying this for several years and end up spelling it out in simpler terms each time I say it: content marketing on the web does not work and will not work for ‘brands’. No business can make the expenditure on the quality and quantity of content required to win significant attention pay a decent return on their investment; at least not in terms of getting people to care enough about them to turn this into sales. The argument that it ‘creates awareness’ is intangible, but more to the point is likewise untrue because of how attention flows on the web.

This is typical of the kind of post I frequently see:

Transforming Your Brand into the Next Media Company

The examples given are for Coca-Cola and Nike; something common to this type of post is to use businesses that are perceived as successful to propagate this flawed idea. Let me assure you that these companies are flailing around in the dark just as much as everyone else not getting decent page views.

Attention on the web follows a heavy-tail distribution, an abrupt curve from a handful of pieces of content that have received most of the attention through to the majority that have received next to none. Go and look at your own analytics data. Focus on the pages that feature original content rather than those that are there to help visitors navigate, such as your home page. Plot the top few hundred data points and see what kind of a chart you get. I already know what it will look like.

Now divide all those figures by four. What you will be left with is the approximate number of people who bothered to read the whole page in each case.

Now try plotting the number of people who completed a goal on each of those pages; the thing that you want each person viewing your content to do next in each case. You should have set this for each page. Or why else would you be spending vast resources on publishing hundreds of pieces of content as recommended in the above post?

Problem is, there is no way of predicting ahead of time which are the few posts that will do significantly better than the majority. You have to do the lot which is why this whole approach does not scale, unless of course your entire purpose is to be a content destination. It is telling that Michael Brito uses the New York Times as an example, claiming they publish over 1,500 articles per day. They are still at the mercy of that heavy-tail distribution however; it is just that their worst performing post is an order of magnitude more successful than your best performing post.

If you were to publish content at anything like the scale needed to compete against content destinations for whom that is their entire business model then you’d be a publisher too, not whatever business it is that you are actually in. Given the problems that web publishers are having making this work I’d suggest becoming a ‘media company’ is a sure way of sinking the rest of your business. It’s unlikely that you can create advertorial that is going to generate sales and hard to make money off the little attention you will be generating from editorial so far removed from your product or service that it’s irrelevant.

As ever, Baekdal is good on this, making the point that more page views from ‘content marketing’ activity doesn’t mean an increase in an outcome you are actually looking for, such as sales, because of the lack of intent behind the visit.

Have a read of this:

Content marketing fairy dust

It is the sound of the slow realisation that you can create all the ‘great content’ you like and it will still be ignored. What the author hasn’t grasped is the reason why. Even with larger budgets and more time they’d still be unsuccessful because of the way attention works on the web. These days the SEO industry simply can’t scale what it does in a safe way to influence competitive terms. Individually negotiating the posting of one piece of content at a time on other people’s websites is too time-consuming to meet their needs, but this is what they have been reduced to persuading clients to buy into as ‘content marketing’ of one stripe or another.

What is interesting is the contrast between the two posts. In the former post Michael Brito is still operating under the misapprehension that a ‘brand’ can run a content destination that people will actually want to visit. In the latter post Paul May rightly highlights that it is all about getting the content itself onto the wider web. Problem is that, because of the heavy-tail distribution, only a handful of sites will ever be receiving enough attention to be worth your while pursuing. These hubs likely understand their value all too well. This is compounded by the fact that anything you do manage to get someone else to post is unlikely to see many people following a link back to your own website. This means you stand little chance of getting them to take an action beneficial to your business. Check your referral data if you want confirmation of this.

Problem is there are enough people out there who really, really want to believe in this who are prepared to pay other people to tell them it is true:

It’s Come to This: A Newsroom Devoted to Brands

If marketers are prepared to spend money on display advertising then I guess they’ll sure as hell throw money at this. As soon as they do we’ll start seeing fake traffic from fake sites to inflate total page view figures as this will no doubt be the big number waved around to reassure everyone that ‘success’ is being achieved.

tl;dr If someone is claiming that your business needs to be ‘telling stories all of the time’ ask them to demonstrate this working for someone else across several hundred posts. Ask them for charts showing page views, comments, shares and backlinks as well as a list of the URLs for you to check. If they actually do ‘content marketing’ they must have this data. Look at the trends and draw your own conclusions.

Other relevant posts written by me:

Let me know what you think on Twitter


23 October 2013

… fake traffic is essentially systemic to online advertising — it’s part of how the business works.

The Amount of Questionable Online Traffic Will Blow Your Mind

This is perpetuated by people not wanting to look into how the ‘success’ of what they paying for is actually being measured. So much of digital marketing is affected by people unable or unwilling to ask hard questions.

Other relevant posts written by me:

Let me know what you think on Twitter