Previous month:
July 2012
Next month:
September 2012

SEO Isn't Dead, It's Just Changing


The SEO industry is in an interesting place right now. Last year's Panda update to the Google algorithm used logic acquired from human quality testers as the backdrop to a new machine-learning algorithm that could assess the 'quality' of website content in a more sophisticated way. Panda downgraded sites that delivered poor user experience and when it was updated earlier this year, an 'over-optimisation' penalty was deployed. The subsequent 'Penguin' update went further into penalising sites that used so-called 'black-hat' optimisation techniques (keyword-stuffing, duplicated content, 'cloaking', link farms/schemes and so on).

The mission of any search engine is to get the content that is most relevant and useful to your search to the top of the results. The better it can do that the better the user experience. The point about updates such as Panda and Penguin are that they have incorporated new ranking factors that have enabled an assessment of content quality based on a highly scalable application of very human contexts. This is leading some to suggest that SEO as we know it will be 'dead' within two years. The methods deployed to try and 'trick' the algorithm into ranking your content higher are becoming more and more irrelevant.

Perhaps. But there's a whole bunch of legitimate practices around creating well-structured, well-linked-to content with a great user experience that still matter. There can be little doubt that the future of search is going to incorporate increasingly sophisticated re-application of data not just within individual applications but derived from and deployed across many different services in order to provide better contexts to results. Google is already integrating data from Google+ and many other sources (location, device-specific, knowledge graph, Gmail and so on) to create new search experiences (like Search Plus Your World which incorporates content shared by people you are connected to on the web into the results) that are more personalised and more relevant.

So whilst SEO may not be dying, it is certainly changing and changing fast. As the link between so-called 'earned media' and search engine results gets stronger and as the data sources that form results become more numerous and more complex, the requirement is increasingly for SEO to be integrated into everything that the brand does. That means not only the quality of your content, but how you produce it and what you do with it. The skills of a good SEO person have always been about increasing visibility of content and getting it in front of people who are prospective buyers/customers/consumers. And that skill will surely be in more demand than ever.

Image courtesy

How I Write

Apple keyboard 1
One of the regular readers of this blog Olivier Legris was kind enough to give me some good feedback on a recent post and as part of that happened to ask me about my process for writing. I'd never really thought it through before, but doing so made me realise that I definitely have a way of doing things. Whilst I'm not suggesting that my way is the right way (everyone should and no doubt does have their own), or that I know more about writing than anyone else (for tips from people that really do know what they're talking about, take a look at some of Maria Popova's posts on the subject), I thought it might be useful to share some of those thoughts here, so here they are:

Connecting the dots

When I started thinking about how I wrote, I found I couldn't get very far without thinking about why I write. Andrew Sullivan once wrote in The Atlantic: 'Blogging is to writing what extreme sports are to athletics: more free-form, more accident prone, less formal, more alive. It is, in many ways, writing out loud.' I guess for me, blogging is thinking aloud. It helps me figure out what I think about stuff. To order my thoughts and connect lots of disparate thoughts that seem to swim around in my head, often provoked by something I've seen or read, until I can write about them. Posts often combine over time to into strands of thought that I revisit or which inform thinking about other related things. It's a fantastically useful thing for me. And the feedback I get in the comments serve to make it even more so. I also write a regular column for New Media Age, which is an interesting discipline in that (unlike blogging) there's a (quite strict) word limit and I can't lazy-link out to explain concepts or ideas. I link out a lot. It's useful not only for the readers, but for me as a way of pulling disparate things together and making sense of it all. 

The importance of reading

This may be a personal thing, but I've found that the quantity and quality of input directly affects the quantity and quality of output. So seeing and reading thought provoking opinions, articles, talks (and lots of them) is really important. And I'm not talking about all those list posts that lots of people seem to like to share ('10 best this, 20 worst that, 7 ways to blah, blah, blah). I once wrote about why I thought companies should be encouraged to employ people who blog. Writing about the industry that you're a part of enables you to start fires and make connections that wouldn't otherwise exist. It shows that you have an opinion and, frankly, that you are bothered. I think this is still true. 

Make use of the tools

I use a surfeit of bookmarking/reading tools to help: Pocket, Evernote for some things, Twitter favourites, Delicious (I keep meaning to migrate my 6 years of bookmarks over to Pinboard as the Delicious UX is getting no better but somehow haven't got round to it yet), RSS (these days usually via Flipboard), Zite. They're all good at different things but finding your own composite of different tools to use is important I think. 

The only writing app I use (despite there being many good ones) is a text file. I find it useful to list ideas/thoughts/quotes/snippets that I've found interesting there. It often helps me remember them, but they also often combine to make something new. They say you make your own luck. To a certain extent I think you make your own serendipity too.

These days I tend to write posts directly into Typepad, but I think John Willshire's Artefact cards can be a useful supplement to help get disparate thoughts into some kind of order. The important thing here is to find what works for you.

It's a Commitment

Writing takes a lot of time. At least it does for me. So it's got to be worth it. I'm all for people leaping in and trying it out, but to write continuously for years means that it has to become part of the fabric of what you do and the way you work, in order to not get squashed by the latest client deadline or episode of Masterchef. This may change but for now I can't imagine a time when I didn't blog about something. It's an itch I have to scratch. 

Finding your own style and voice

Most of the blogs I go back to again and again have a strong and unique tone of voice. Finding that tone of voice can be difficult but the best advice is to write for yourself. As the audience grows on some blogs, it can feel as though the author is starting to write for what they think their audience might like rather than what they find interesting or obsess about. I can see how this might work but in my experience those are never the most interesting blogs. Another thought on this comes from Stephen King, whose On Writing is still one of the best books I've read on the subject. In that, he talks about writing for your 'ideal reader' - one person who can represent your readership, give you feedback and tell you when you've written a load of old hogwash. There's no right and wrong with what you choose to write about as long as it matters to you, and blogs should rightly evolve. Somehow, for example, I seem to be doing less of the 'saw this and thought it was interesting' type of quick post, and more of the 'here's what I think about that' type of exposition. There's clearly room for both, but I do wonder if the former hasn't shifted more permanently to the likes of Twitter and Tumblr. 

Anyhow, as I said at the start, this is a personal view and simply what works for me. But I hope it's useful in some way.

Image courtesy

Are We Forgetting Things Faster?

"The bubble bursts faster and faster with each passing year. We are losing interest in the past more rapidly." Erez Lieberman Aiden

An interesting snippet picked up in the latest issue of Wired talking about Google's NGram viewer which enables anyone to mine the data held in Google's database of over 5 million digitised books to reveal cultural changes over time. Jean-Baptiste Michel and Erez Lieberman Aiden of Harvard's Cultural Observatory, who worked with Peter Torvig at Google to create the viewer, say that we are forgetting things faster. Where it typically took 30 years for mentions of a particular topic to halve in 1880, this had compressed to just ten in 1970. Makes me wonder what it is now.

Google SearchFirestarters 2: The Mobile Future

Google Searchfirestarters 2
Last week saw the great and the good of the Search Marketing world (and a number of interested generalists) come together for the second in our spin off series of SearchFirestarters events, this one on the subject of mobile. As I said in the run up mobile is such a rich subject right now so we chose to focus on not only what marketers need to think about today, but on what the future could look like, and we had a great line-up of speakers from client, agency and media owner side to help us.

The context for this is that whilst smartphone penetration has risen rapidly (now at 51% in the UK), it is of-course what people are doing on their phones which is making all the difference (one study indicating that 60% of time spent on smartphones is new activity). Some notable milestones have been reached over the past few weeks including the fact that 60% of access to LOCOG’s digital assets was from mobile devices and upto half of U.S. and UK Olympics video streams served by NBC and the BBC were to smartphones and tablets. Search activity on mobiles is up 400% since 2010, and yet according to Econsultancy only 25% of companies are using mobile search advertising. In her latest Internet Trends deck, Mary Meeker highlighted the gap between consumer time devoted to mobile and the proportion of budgets being allocated to it, and how much lower eCPMs are on mobile in comparison to desktop internet (the FT speculated a couple of weeks ago that the emergence of the mobile internet could be a replay of the Dotcom era where a similar thing happened). Meeker also emphasised not only the breadth of potential disruption across many different areas (the "re-imagination of almost everything"), but also the speed of disruption, noting that "this cycle of tech disruption is materially faster and broader than prior cycles".

Gareth Jones, who runs Digital marketing for Carphone Warehouse spoke on the journey they had undergone from a situation where in many ways they were playing catch up to consumers to a position where they now think of mobile not longitudinally but as a behaviour. What he means by that is that mobile can and often does bind a lot of their retail activity together. The journey has transitioned their mobile presence from a mobile version of the desktop internet site, to one that was optimised for mobile, high on functionality and utility, and promoted through mobile optimised search marketing. This has led to a pretty huge differential in performance between desktop and mobile (+11%/+225% year-on-year). Interestingly, they are seeing a high elasticity between spend on mobile search and traffic - unto 60% of traffic to the mobile optimised site comes from paid mobile search, but the conversion rate is notably lower than desktop. This suggests that people are using mobile for both research and purchase, but it also suggests poor quality conversion. Counter-intuitively, the reverse is true. Conversion to higher ticket items is better on mobile (perhaps driven by early m-commerce adopters) meaning that it has significant value to them - mobile is now bigger than a flagship store.

Gareth Davies of Vodafone also talked about how, for them, mobile is the 'glue' between channels. They are seeing an exponential increase in data consumption on the network. During the Olympics, average data traffic was up 10% (25% on busy days), on the day of Bradley Wiggin's gold medal win it eclipsed both the royal wedding and New Years Eve, and on 'Super-Saturday' the network served up enough data to stream video to one screen for 40 years. Indicative of the pace of change required in this environment they've been through four iterations of their mobile site in just over a year focusing on creating a mobile optimised look, feel and functionality (like mobile checkout) and smart ways of joining up the back end with the desktop site. They've also been thinking smarter about marketing (mobile search is 16% of total search spend, and since 30% of email is opened on a mobile device optimising e-mail marketing for mobile) and about how they can join up with offline activity as it's increasingly common for them to see customers turn to their mobiles when prompted by offline. Data will play an ever bigger part for them in the future - in more sophisticated targeting for their mobile advertising, to join up critical areas like telesales, online and retail, and in attribution to better understand the value of individual touchpoints and pieces of activity.

Mobile specialist and veteran Simon Andrews, founder of Addictive, focused initially on how many clients are (still) spending (or rather wasting) money on search marketing that is directing traffic through to non-mobile optimised content. Search engines are increasingly delineating between optimised and non-optimised content, both visually in their mobile results and in the ranking. Simon talked a bit about perceptual filtering - the idea that our brains are pretty good at filtering out stuff which is less interesting and useful to us and how this is a pretty good model for the future of search. Using the kind of data that mobile is uniquely good at, more information will increasingly be available on the results pages themselves to help users in their search (e.g. results that use your location to show you stock levels and prices for the product you're after in nearby shops), giving much better contexts to search results. Because of this, permission will become more important, and attribution starts to become interesting since more parts of the journey are conducted using one device.

Husayin Savas, a product manager for mobile ads at Google, then talked about a vision for the future of mobile search. Picking up where Simon left off (and referencing some aspects which can be seen emerging in Google Now) he talked about the importance of improved contexts and 'answering with knowledge', search beyond type, and search that works seamlessly across all screens. For the former, he talked about the recent launch and future integration of the knowledge graph and how search can use data and different contexts relevant to us to give us more helpful results (e.g. searching for a flight will show if the flight is delayed, search history shared seamlessly across different platforms). There were some quite amazing demonstrations of voice and visual search, and using Google Drive to easily share stuff from mobile to the cloud (e.g. taking a photo of a recipe on a mobile, have the Google engine translate it into text in a Google doc and enable a friend to contribute/edit it).

Across the four provocations, we moved from present day challenges and opportunities to those that will become increasingly important in the future, and the questions from the audience at the end reflected this mix as well. The next SearchFirestarters event will be in a few months time. We have some thoughts on what would make a great theme but as always are keen to encourage suggestions from the industry so if you'd like to put something forward for debate, please let me know. In the meantime, you can see the talks illustrated in all their glory by Scriberia here, and view a Storify of the event here. Thanks to the speakers, and to all those who came and made it such a fascinating event. 

On The Decline Of Magazines

People love a good a good funeral (and as David Hepworth put it "in the digital age they don't even have to dress for it"). In only the past few days I've read two articles that pronounce the 'death' of marketing (only to then go on to explain how marketing is still very much alive, albeit changing. *Sigh*). So it is not without some disheartenment that I read articles about the troubles of a medium which is close to my heart: magazines.

David Carr's piece in the New York Times focuses on the difficulties at Newsweek but makes a broader point about how magazines, like newspapers, have been in a steady slide that has now brought them to "the edge of a cliff". Last week's ABC release revealed hefty declines in US newstand circulation across just about every major magazine category: "When 10 percent of your retail buyers depart over the course of a year, something fundamental is at work". In the UK, the circulation figures released yesterday by the ABC also revealed a market that remains extremely tough with the total circulations of the Top 100 actively purchased magazines falling by 4% in the first half of the year and only a fifth of them able to put on sales year-on-year.

Publishers tradionally have two major revenue streams: circulation (newstand, subscription) and advertising. Declining circulations are likely to be a double whammy since advertising revenue, though not directly linked, is reliant on the reach provided by circulated copies and readership. Sure enough, Carr quotes figures from the US Publishers Information Bureau showing that advertising is down 8.8% year-on-year year, suggesting that magazines are "in a downward spiral that not even their new digital initiatives can halt".

Publishers have arguably been caught in the classic innovator's dilemma, becoming great at improving products and processes that have an established role in the market, but less good at investing in disruptive technologies and ideas, perhaps hindered by their accountability to their existing customer base (who, as Christensen suggests, can be a substantial barrier to innovation and often ask for better versions of current products), or by ways of thinking informed by their position in the market or precedents.

On the face of it, magazine owners have much of the raw material to generate considerable digital success: an abundance of high quality content and the talent to create more of it; established relationships with commercial partners; engaged audiences focused around tightly defined common interests or outlooks (as I've said before, the best magazines make you feel as though you are part of an uncommon community); established subscription models; a huge offline channel which can be used for promotional purposes. Whilst it is undoubtedly a challenge to re-tool and re-work skills and processes for digital, I know that many publishers have been working hard at just this and there are some great examples of teams doing great things in the digital arena.

My favourite writer on the magazine industry, David Hepworth, noted that the CEO of Hearst Magazines in the US recently told the Economist that magazines already need five or six revenue streams in order to be successful. Whilst that may seem like a case of simply chasing the money, my hunch is that most publishers have a very good idea of what those streams will be and at least some sense of how they're going to get there. The challenge however, is that this will take time and (much like newspapers) despite building large digital audiences, for every online user they currently derive a fraction of the revenue that is made for every reader of a print magazine. Which makes it very difficult to replace print losses with digital gains.

Finding a solution to this dilemma undoubtedly requires different thinking. If I was running a publisher right now I'd be thinking about smart applications of not just destination thinking (something publishers have traditionally been very good at) but also distributed thinking. Something akin to what has been described (I think by Emily Bell, though I couldn't find the original source to attribute so apologies if I'm wrong on that) as "being of the web, not on it". Or weaving content into the "fabric of the internet", a phrase that The Guardian used when they launched their Open Platform (which allowed partners to reuse Guardian content and data for free). Here's an example of what I mean.

Clay Shirky once wrote "Society doesn't need newspapers. What we need is journalism". So perhaps the opportunity lies (at least in part) in something magazines have always been brilliant at: great curation. But done differently. I've said before that I believe that the future of content will be about three pillars of content curation:

1. Algorithmic:- we see stuff because a software, or a technological process interprets, anticipates, or predicts our needs. Examples of this include Google's personalised search, Amazon's recommendation engine, LastFM's scrobbling, Facebook Edgerank, aggregator apps such as Zite, Flipboard and Currents.

2. Professional:- we see stuff because skilled editors and commisioners use their insight and knowledge of audiences to determine what might interest them - magazine, newspaper and website editors, radio DJs and so on

3. Social:- we see stuff because we, our friends, or a wider audience think it's good and/or relevant. Examples of this have been around for a long-time (links shared via social networks, social bookmarking tagging and voting, Twitter lists, most-shared etc) but this is word of mouth writ large and digital.

Over this is a layer of self-curation (the choices we make) which gives an explicit and implicit dimension to each one of these.

Increasingly, I think good content production, distribution and consumption will involve smart combinations of these three elements. They work together to give me the content I want as well as allowing for a healthy dose of serendipitous discovery of new stuff. Using algorithms is an excellent way to not only anticipate needs and personalise digital content (so that, for example, I see more of the kind of content that it knows I like) but also to recommend relevant content from elsewhere (something the BBC have called Perceptive Media). Social curation is also a way of surfacing the best content about the stuff I'm interested in, or that which a community of people with interests similar to mine think is good. Combine appropriate forms of these with brilliant professional curation that remains as critical (arguably more so) as ever and you have something very interesting.

The question of openness is critical to this. One of the reasons I don't really get closed publishing apps is because they are effectively dead ends, and research has shown that people don't really want dead ends, they want to use digital magazines as "exploration springboards" (there's a thought here about technology and the potential of HTML 5 to create brilliant content apps as the Financial Times have shown). If your website or app is to be a true springboard for exploration, this requires publishers to curate and promote not only their own content but high quality and relevant content from elsewhere. This is a change in mind-set as well as approach but something that also requires new skills. As David Hepworth puts it:

"Editors used to be picked for their ability to predict what was about to be interesting to people. In the future, they'll be picked for their ability to note where the interest is and minister to it. The old idea of followers and leaders doesn't apply any more. Many of your readers know more than you do."

As Dave Winer once said, "if you want to make money on the web, send them away". The more you curate great stuff from not only you're own content but from wherever it exists across the web the more useful you become (and once you start to become far more useful, who knows, perhaps you have the basis for a more rounded and attractive digital subscription model).

Which leads us to the question of scale. Approaches like that above can work with small, tightly defined audiences and interests but it's also highly scalable. Scale brings benefits in not just reach and revenue generation, but also because it gives a bigger platform with which to do interesting things (particularly through data and targeting). There are many ways of acheiving scale. You can work hard at it over a number of years like the MailOnline have. Or you can aggregate and partner.

There are some interesting precedents here. The Huffington Post of-course, which is both a news website and a content aggregator, having a core group of regular contributors but also a huge number and range of different bloggers. Glam Media (who, in 2010, were named as FastComany's most innovative media company) have acheived significant scale (280 million unique users) by aggregating over 1,500 lifestyle websites and blogs, curating and bringing together content from other sites in the network making them not just a simple ad network but more of what their CEO Samir Arora describes as a 'distributed media company'. Website owners get access to revenues they would otherwise find it hard to secure, Glam gets good content and eyeballs to sell to advertisers. This is an interesting approach for publishers for a number of reasons: they are taking a leadership/facilitation role in their given market; they get access to an altogether different level of scale and reach (and more data means more targeting which means more revenue); they also have access to a wealth of great content from which to curate the best. An example of this happening is The Guardian Environment Network which brings together 'news and comment from the world's best environment websites'.

Once again, the metaphor of the API is an interesting one. Thinking about content as an API also pulls in aspects of both openness and scale that I think are important. Publishers have finite resource and so finding smart ways to apply that resource in the service of generating scale but also efficiency makes sense. The FT's approach, summarised nicely here in Adam's write up of May's FT Mobile event, is insightful:

"Having entirely seperate workflows and content systems for each digital product is clearly a non-starter. A clear seperation of an expression layer - the web app, the web site, the newspaper - and the content layer, with the gap between the two being bridged by an API allows rapid and efficient development of new products for new platforms, because of good, basic infrastructure hygiene. Essentially, they're thinking of their content as a dataset that can be interrogated by their products through APIs".

Much of this involves a shift toward creating content driven services and experiences and that's a hard cultural, technical and behavioural shift to make when much of it is anathema to what you've known for years. But as David Hepworth has said: "These challenges are going to force us to decide whether we’re people who make stuff or people who do things". I'm not suggesting that any one of these approaches is a golden bullet but as I said at the start, much of this is about smart ways to combine distributed and destination models. Publishers have been exceptionally good at destination thinking. But if I was running a publishing business right now I'd be challenging my teams to find new ways to be the best facilitators, aggregators and curators in their markets. I'd be asking them how our content API could apply good, traditional skills and output in new ways. How we can become more embedded in the fabric of the web. I'd be asking them what's the publishing equivalent of Google Ad Sense, or the YouTube embed functionality, or the ASOS Marketplace, or the Amazon Associates programme?

I'll admit that I wrestled for good while with this post. Partly because these issues are structural, cultural, and deep-seated. But partly because I did not want to appear to be overly negative about the fortunes of a medium and an industry I have a lot of time for. Magazines will be around for years to come. Magazine owners potentially have a very bright future. But it's a future that is dependent on their ability to embrace and lead change. So that's where I'd start. I'd be interesting to know what you think.