Calum Shepherd     speaking     consultancy     archive     feed

Product Management Blog

Are we prioritising structured data over search intent?

The Semantic Web is an interesting concept, fuelled from a desire to be able to establish URIs for information on the web. The end goal of structuring unstructured data on the web is an ambitious one, which I don’t think will ever totally come to fruition in its current form. Turning the web into the equivalent of a giant database is great, but will need wide spread support within HTML5 and via CMS creators.

“The Semantic Web is a collaborative movement led by the World Wide Web Consortium (W3C) [1] that promotes common formats for data on the World Wide Web. By encouraging the inclusion of semantic content in web pages, the Semantic Web aims at converting the current web of unstructured documents into a “web of data”.” Wikipedia

With the Wall Street Journal highlighting a renewed focus in this area for Google (as reported by Searchengineland, it again raises questions about search engines use of this semantic data. Webmasters and marketers adopt semantic web mark-up to help make the proposition possible, in turn dedicating their own resource and time.  Google has long strived to be able to use this data effectively, pretty much one of their long term goals in fact. This approach could stem as far back as when Google began to provide search results for synonyms, as opposed to results for an exact match of the search query used.

##Search intent considerations The interesting part for me is that searcher intent seems to be a bit of an afterthought when using data from the semantic web. Take football as a prime example. Within the UK football scores are returned regularly for top level queries around recent / current matches. This would see me search for “Celtic v Rangers” and I would be fed back the current score via aggregated sources. This is a mixed blessing, as at this stage we don’t actually know what information the searcher was looking for.

##The great “semantic search” swindle Alongside this, there is the on-going concern that as people are adding structured mark-up on site, Google is taking more and more of this information, whilst providing little in return. The whole experience of search is based around getting the most relevant, insightful content and being able to either narrow your search down, or browse through a number of sites to build a general opinion on the given subject. As any University project will tell you, multiple sources are always better than a single source; with structured data you’re losing the personal aspect of peoples “opinions” on the subject, instead having them grouped together in a “one best answer” approach.

Whilst not applicable in all instances, it is certainly something worth thinking about. The advancement of this area of Google’s game is definitely an exciting one. People will again have to innovate to ensure they see any form of benefit from it.  I would ask;

“Are we prioritising structured data over search intent?”

Increase Like button clicks on informational websites

A colleague of mine Ross highlighted the idea of placing social sharing functionality alongside content on informational sites. A great idea for highlighting specific gems likely to be shared, as button placement currently tends to be commonly top / bottom. Some further discussion and a bit of further research this as being a nice little way of potentially increasing the rate at which people interact with social sharing functionality. Essentially what he described is a route to increasing Like Button clicks on informational sites, in turn expanding your visibility and following on Facebook.

I wanted to find out exactly how this could work, alongside any potential benefits and impact points - this took a little more digging

It is quite common for e-commerce websites to achieve additional Like button clicks through making use of Facebook Open Graph mark-up, declaring products as objects and placing buttons alongside to aid in sharing. This works really well for people thinking about making a purchase, or people about to make a purchase and sharing it to friends and families in advance. ASOS is a great example of this.

This e-commerce method works because your single page is a single object, a product. However, how does this work when you are looking to share a specific area of a page on an informational site, without impacting messaging when sharing the page as a whole? This is the aspect that interested me, as I wanted to know how they were pulling it off and if there were any search implications.

Well, let’s take a look at the example Ross initially provided in the form of Britain Magazine.

Britain Magazine increases Facebook Like button clicks through adding the ability to share quotes on a page out to Facebook. However, the challenge is that you can only make use of one set of Open Graph mark-up per page, meaning that your entire page would have to become about one single quote, even though it will be providing information above and beyond that.

##How do they achieve this? Simple! They use one Like button for the homepage, then on the same page link through to unique URL’s for each quote, with unique, relevant Open Graph mark-up.

This means that every quote has its own unique URL, so you aren’t actually sharing THAT quote, your sharing the same quote but on a different URL (Something the typical user will never realise). You can see the true URL you are sharing within the iFrame.

##The user journey

A great idea with wider implications

  • Google will crawl these URLs, thus for every quote a unique URL may be indexed
  • When shared, visitors will arrive on a page with a quote and navigation, not the original section of the homepage where the quote was shared from
  • A likely increase in like button clicks for your site’s content as a whole

Essentially, it provides a good route to increased like button clicks, visibility and brand interaction. Alongside, it potentially provides minimal search value and some indexing issues unless spider controls are used e.g. noindex. All in all, a great idea!

What does Google know about me?

Google rolled out Search, plus Your World to quite a bit of fanfare, mixed opinion and wide ranging commentary a few months ago now. I was very much of the opinion that it was more Your World, plus Search, as opposed to Search, plus Your World. This opinions was formed off the back of having a heavy weighting on Google+ content, social recommendations from friends who you hardly know and more.

We’ve discussed how Google can associate you with people if you have a Google Account that has Gmail bundled in, with people you regular email, with Google+ accounts, appearing within your results recommending various pieces of content you may be interested in.

However, after digging around my profile today I actually discovered how deep this social graph extends. Check out your Dashboard, Available under Accounts > Products > Sign into Dashboard. Upon doing so, scroll down till you see the social connections and content area. Click on “View Social Connections”.

Although this isn’t new since Search, plus Your World launched, it does give you an idea how deep the social graph extends. People really are the centre of personalised results.

This should give a good indication as to why our media architecture is so focused on social, social links and social platforms when creating strategies in conjunction with the search teams. Public knowledge for some, some “nice to know” for others. Thoughts?

Your World Plus Search (Not Search Plus Your World)

“Search, Plus your World” has begun rolling out on for logged in users, which is big news within the industry. Since the launch, we have seen high volumes of coverage, both within the industry and in mainstream media.

Search Plus Your World, providing rich social oriented results.

This article discusses the more topical aspects of the update, with their being plenty of excellent advancements for us to shout about in addition to these! If your a little behind on the news front, then the LBi blog does a good job of highlighting the main areas that “Search, plus Your World” impacts, alongside why it is so fundamental.

##Best man gets the best job? The changes announced by Google really do represent a fundamental shift in the volume of social oriented content seen on the first page of results. Alongside this, the weighting on the algorithmic process behind them appearing there is also likely to of changed. When personalised search was first introduced many in the industry were up in arms. On going updates then saw closer integration of social content, including profile updates from Twitter. This was designed to give users up to date, topical content on developments as they happened.

It appears that this update really does appear to begin to move Google away from providing results that follow the “best man gets the best job” type analogy. Building upon algorithmic signals that highlight recommended content from friends as being more relevant for you, we see an increased presence of content weighted on what your friends like or discuss. However, there is a clear focus on these being taken from a limited area of the internet in the form of Google+. This a interesting point that has also been raised in a Searchengineland articles

##Do your friends really know best? With the first page of results now dominated by personalised results - I’ve begun to realise that this is great in some scenarios, but not so great in others. When searching for news on a specific subject, I’d like to see the most authoritative source of information, not results heavily weighed by someone Google feels is connected to me (via signals from platforms such as Gmail).

I’d go as far as to say this is “Your World plus Search”, not “Search plus Your World”.

With a heavily personalised first page and a raft of search results hidden from page two onwards, it does paint a strange picture. I get the impression that this begins to move Google Search towards being a window to the Google Universe first and foremost, whilst being a generic search provider second. Google’s main source of revenue is advertising, yet they seem to be impacting results in a big way by pushing a non profit making Google+. In turn, we may well see further paid spend to cover gaps and increased resource devoted towards full bodied Google+ social strategies.

##No matter what, it’s great news for Social SEO The development of forward thinking social SEO strategies are again in the lime light. Working with social teams to achieve strong coverage, whilst focusing on more traditional aspects such as easy sharing of website content is essential.

It will be interesting to see what direction Google continues to take, with recent comments from Twitter also highlighting concerns with the new approach. The LBi blog has a nice run through by Simon Hayes, so feel free to check it out as well!

The occupy Flash campaign

(This post was previously posted in November 2011. I have now duplicated here for greater reach. To view the original post, please visit the LBi blog.

Denouncing Flash and beginning a campaign to swat its usage, Occupy Flash have received strong coverage in mainstream media. The first thought I had was:

“is all this really necessary?”

Shockwave Player launched in 1995, with Flash being made available to download the following year. Launching in a very different digital landscape from the one we see today, under the guise of Macromedia Flash Player. Flash was brought into the limelight with the intention of providing multimedia functionality through web browsers, something that would be sure to impress the surfers of the era.

1996 was not just the year of Flash, it was the year when; Intel launched the the Pentium Processor, where Apple was in the process of going down the pan, and the year the 16MHz Palm Pilot launched - providing computing power on the move. The era now looks a lot like the technology equivalent of the iron age, alas Flash really has lasted the test of time whether people like it or not. I would be inclined to say that Flash was the very beginning of a shift away from traditional software, putting us full steam ahead towards web based deployments mirroring traditional software functionality (think Office 365 – just minus the cloud bit!). Today it is one of the most widely deployed playback technologies ever distributed online.

Flash has provided interesting functionality to a range of different websites. It was always a firm favourite for branded micro sites, providing rich multimedia experiences for visitors. A sole reliance on Flash for web design is in decline, for a number of obvious reasons. However, this has been aided through a well publicised spat between Steve Jobs and Adobe’s pride and joy. This saw a rejuvenated campaign of anti-flash sentiment making itself known amongst the Apple crowd.

Today the desire to adopt HTML5 and CSS3 is front and centre, heralded as a saviour from buggy browsing on the internet. With Adobe announcing the retirement of Flash Lite (mobile), the Occupy Flash campaign and Microsoft’s exclusion from the Windows 8 Metro UI Internet Explorer - the future does indeed look bleak.

It seems industry are putting a focus on Adobe in general, not just Flash. Many articles and commentators have overlooked what Adobe actually provides. As mentioned by Carlous Naxareno within an article on the Thenextweb, Adobe is very much in the business of creating and selling design based tools. With a reaffirmed commitment to develop software that aids the design of HTML5/CSS3 websites, Adobe are on positive track in what could have been a tough time.

Website traffic from mobile devices is constantly on the rise and with a ongoing debate around the absence of Flash on the new Google Phone, it becomes hard to for see continued strong development in this area. However, actively encouraging its downfall seems unnecessary, with a successful outcome only limiting the reach of web based multimedia.

Markers need to be savvy and pick and choose their technologies. This is no different. A campaign against technology is never one I would choose to be a part of.