Greenlight's Search Blog

The UK contributes more pages to the Internet than any other European country

A million new books are published globally each year, a fifth of them in the UK. The UK in fact publishes more tomes than any other country on earth[1]. Is this also the case online? Does the UK contribute more to the Internet’s pages than other countries?

Research undertaken by Greenlight, focussing on countries in Europe, suggests that this is indeed the case. Amongst the 20 largest European countries, the UK has contributed the most pages to the Internet, around 5 times more than France and Germany, the study found. Whilst this doesn’t account for the quality of those pages and who has produced them and for what reasons, it’s a good relative indicator that the UK’s offline publishing leadership is being mirrored online through a combination of corporate and consumer publishing.

Further findings:

•         45% of the pages on the Internet were created in Europe, 8% in the UK
•         There’s a dramatic drop in the number of Internet pages created ‘per head’ as you go from the north of Europe to the south
•         The Ukraine has the most Internet pages per head, almost 8 times more per head than Romania, Greece, or Turkey, who have the fewest
•         The UK has published more pages on the Internet than the 10 least active European countries combined

---

[1] Goldfarb, Jeff. "Bookish Britain overtakes America as top publisher", Reuters Entertainment, May 10, 2006

4% of websites have page load speeds detrimental to their search marketing efforts

Website page load speeds will become an increasingly important factor in 2010 for search advertisers, and demonstrates that page load speed performance cannot be assumed, even for some of the biggest sites in the UK. In a sample study of 100 of the UK’s most popular websites, Greenlight determined that 4% of them had page load speeds slower than the acceptable threshold set by Google, beyond which the advertiser may see increased click costs.

 

In June 2008, Google revealed landing page download time has an impact on a marketer’s Quality Score in Paid Search. This meant that ‘latency’ began contributing directly to a campaign’s performance and ultimately it’s ROI. At the close of 2009 there was also much speculation over whether this would also make its way into its natural search algorithms too. Matt Cutts hinted quite heavily in that direction in a video interview for Web Pro News, where he said:

 

“Historically, we haven’t had to use it in our [natural] search rankings, but a lot of people within Google think that the web should be fast. It should be a good experience, and so it’s sort of fair to say that if you’re a fast site, maybe you should get a little bit of a bonus. If you really have an awfully slow site, then maybe users don’t want that as much.

 

I think a lot of people in 2010 are going to be thinking more about ‘how do I make my site be fast,’ how do I have it be rich without writing a bunch of custom JavaScript?’”

 

Greenlight took a sample survey of 100 of the UK’s most popular websites across four industries – consumer electronics, clothing retail, travel and finance. These it determined from its independent sector reports which reveal the most visible websites in Google natural and paid search across various industry sectors, per quarter.

 

 Greenlight tested their download speeds at the same time of the day, outside of seasonal peaks in server load and made multiple requests to get an average. Greenlight’s survey found load times ranged from the exceptional (Argos.co.uk at 0.29 seconds was the standout) to the painfully high (a top high street electronics retailer at over 15 seconds), and everything in between.

 

To define what would constitute an unacceptable average load time, Greenlight devised a methodology that mirrored Google’s method of determining an acceptable maximum. The threshold Greenlight determined was 4.97 seconds (i.e. 3 seconds above the national average). Anything above this would almost certainly fall foul of Google’s Quality Score download time guidelines, as outlined here.

 

Greenlight’s results revealed that:

 

·         4% of the sites analysed had average page load speeds far in excess of the 4.97 seconds determined as the threshold and therefore run a risk of seeing increased costs per click

 

·         3% of the sites had average page load speeds in excess of 8 seconds, which research points to as being the point at which users are likely to abandon a site

 

·         The best performing sites, in descending order were Argos, River Island, Holiday.co.uk, Fool.co.uk, and Comet, all of which were exceptional within the group

 

·         There was no industry pattern – all sectors had a broad spread of high and low page load speeds and the size of the company made no difference either

 

Google, incidentally, if it were part of this study would have performed best of all the sites as it had an average page load speed of 0.11 seconds - definitely leading by example.

 

Approximately 4% of the UK’s most successful websites have page load speeds that are to the detriment of their Paid Search Quality Scores. This affects their Paid Search performance and will be compounded further if Google decides to use latency in its natural search algorithms too. Ironically, poor download speed is actually very easily fixed. Businesses need to look at things like content distribution, cache control, and even simply reducing the number of HTTP requests their pages make.

 

As a follow-up to this research Greenlight is preparing a guide to reducing page load speeds that will be released in Q2 2010.

 

How search engine market shares look around the world featuring Bing, Yahoo and Baidu and others

How search engine market shares look around the world featuring Bing, Yahoo and Baidu and others.

Google’s quitting China spells loss of advertising dollars and greater dependency on foreign operators

 Google has indicated it may cease to operate in China following a cyber attack aimed at gathering information on human rights activists.  

If Google quits China, then it leaves MS yahoo and Baidu as the primary sources for accessing the Chinese searching audience. Without Google, assuming Yahoo and Bing remain, that 30% share will re-distribute. Most likely to Baidu, which will further cement its leadership position, and the remainder may not split equally among Bing and Yahoo users, which would further upset the balance of power for one or the other, and means growing in the Chinese market is likely to make it very difficult for at least one of them.   For advertisers looking to target the Chinese market, it means a greater dependency on foreign operators who are less familiar to them, and less integrated with their ad operations.  

For the US and global advertising industry it means ad dollars are going to go overseas into Chinese pockets as opposed to strengthening the US coffers too.

France’s “Google tax” – Sounds like protectionism

A report has been made public proposing that France begins taxing the likes of Google to subsidise its own creative industry. The authors of the report suggest that this new taxation could raise up to the equivalent of $28 million, which in the grand scheme of things isn’t much money at all. The suggestion that the tax will help France subsidise music artists and book publishers doesn’t therefore sound like the sole objective here, given that this amount of money won’t make much of a difference if it’s spread so thinly across France’s creative community. On this basis it feels like protectionism of the worst sort – instead of collaborating with successful, innovative companies, or creating an environment that promotes innovation domestically, France appears to want to give its own industries an unfair commercial advantage by taking money from non-French firms. If the proposals are made into law, this then obviously sets a precedent which would allow France to raise this taxation as it saw fit and usher in a period of conflict between the French government and some of the world’s most innovative online firms.

 

It is of course understandable why France would be concerned, particularly given that Google has approximately 98% of the search engine market in that country, but instead of treating successful companies as the enemy it should look to the music and publishing industry to reinvent themselves and offer the consumer an alternative to Google et al. Google and other ad-driven websites exist because they offer something of value to the consumer. If the creative community have left a vacuum in terms of how they serve the needs of their online-loving consumers, and then this vacuum is filled by the likes of MySpace and Google, then the blame shouldn’t be put at their door. Punishing innovators for innovating is a very short sighted stance to take.

The history and evolution of SEO

Ever wondered when links became such an influential component of SEO? Or when Google released the infamous "Florida" update and what effect it had on search engine optimisation? Perhaps you're interested in the origins of the unsolicited link exchange, or want to know when (and why) nofollow was created.

All these questions and many more are answered in Greenlights SEO History of the Internet, a graphical two page "cheat sheet" plotting the evolution of SEO and key events that have happened during the fourteen year history of the industry.

Here at Greenlight the history of SEO is something we find quite fascinating.  But while there are many disparate sources of information on the topic, I was unable to find a single source that packaged all of the key events that have happened in our industry in a single, graphical timeline.  Thus, I give you the History of SEO (click for full res images):

 

The images feature a chart showing how the importance of various factors in search engine algorithms (and therefore SEO) has fluctuated over time, from the early days of search engine submission and on-page SEO, to the introduction and dominance of PageRank and other link based metrics, to today's much more complex and nuanced picture. The factors plotted on the chart are as follows:

Search Engine Submission - Manually or automatically submitting a site to a search engine, necessary in the early days due to the fact that search engines had limited (or non existent) crawling capabilities, and still yielding limited results years later.

On Page SEO - The practice of ensuring a site is both accessible to search engines and relevant to the terms you are targeting by the placement of keywords on page, on page SEO includes considerations of website architecture, build and content writing.

PageRank (or similar) - The use of links by search engines to judge the importance of a page flipped SEO on it's head and became the dominant factor in the early years of the decade. In this chart, PageRank essentially refers to the raw credibility conferred by links to a given page.

Anchor Text - The use of link text to add context to the raw credibility provided by PageRank et al...

Domain Authority - Link factors, including PageRank and links from trusted sources, assessed at a domain level rather than at a page level; the idea being that all other things being equal a page launched on a site with higher domain authority will rank better than a page launched on a site with low domain authority.

Link Context - Link context refers to the placement of links on a page and, more importantly, the context of the page and the site the link is on. Link context brings an additional level of complexity and detail to the analysis and acquisition of links beyond merely looking at the raw credibility of a link and its anchor text.

User Signals - The use of user behaviour, such as click through rates, bounce rates and search patterns, by search engines to assess the level of trust search engines have in a site. User signals were not felt as a serious part of search engine algorithms until this year when Google implemented the Vince algorithm update.

Alongside this chart we've described key events and turning points that have influenced or evidenced the evolution of search engine optimisation - whether that be the launch of a new search engine, the first SEO conference or an algorithm change.

For newcomers to SEO this is an invaluable primer on the industry and a fascinating journey into the past, while we hope that it will bring a (sometimes rueful) smile to the face of you old timers out there.

References

By necessity the task of assembling an accurate history of SEO consists in large part of compiling and comparing dates and information from a wide variety of other places. During the course of my research I came across too many resources to mention them all individually, but the key sources of inspiration and assistance to which I'm indebted were (in no particular order):

 - Brett Tabke's brief history of SEO thread on Webmaster World.

 - Mark Knowles and his site the history of SEO.

 - Rand Fishkin's how google's ranking algorithm has changed over time.

 - Google's milestones corporate information page.

 - Aaron Wall's search engine history.

 - Last but by no means least some invaluble input from Danny Sullivan of search engine land.

Google takes its first bite of the Apple

So, less than a month after filing the trademark application for the ‘Nexus One’, Google has launched its first hardware retail product direct to consumers and starts a direct assault on Apple’s dominance of the mobile sector. This is a great example of ‘speed to market’ from the most over-exposed brand on the planet launching a product in the most over-exposed global consumer market. In fact, the only other brand that has shown that it can operate as effectively as this is, er... Apple.

Google is breaking new ground by becoming a direct retailer of this handset. It suggests that they will be braver than Microsoft in entering the PC hardware retail space. Can we therefore expect this year a Google laptop running their Chrome PC operating system to rival the much rumoured Apple tablet (due possibly in Q1 2010)?

The question we should be asking ourselves as marketers is: what kind of promotional opportunities and threats are likely to arise from Google managing not just the search space but the operating system, software and hardware as well as, it seems, the retail consumer relationship for these products?

The short-term opportunities are abundant.  Tracking could include non-web-based activity such as emailing, gaming, even potentially working, giving markets the chance to deploy behavioural targeting in a far more comprehensive way than previously imagined by web-only systems such as Phorm. It will encourage brands to follow the path shown by the iPhone application development revolution by launching Apps that demonstrate the usefulness of their brand rather than being limited to shouting the loudest about the attractiveness of their products.

However, in the long-term, the impact on the marketing landscape may prove extremely disruptive. Brands will be able to get closer to users than ever before, leaping from the web page into the software and operating system. This has obvious advantages for brands with products that are directly relevant to this environment such as software and hardware providers but for others it adds to the increasing pressure to make their products relevant to the digital user environment. How does a soft-drinks manufacturer or fashion retailer squeeze their message into the closed environment of day-to-day PC user activity as potentially owned by Google (or Apple or perhaps even Microsoft!)?

Just as relevance is a key ingredient in achieving success in search marketing today, brands are increasingly going to have to contrive ways to be relevant to users’ everyday digital activity that remind them of their products’ key attributes and brand values. Relying on users finding your brand on a website or search results page may no longer be enough as Google wraps itself around users giving them every reason not to look beyond the data and toolset it can provide.

The importance, then, of brands will be paramount but it may be a type of branding we are not used to. To get close to and stay close to consumers, brands will have to be useful not just attractive. They may have to find ways to partner with consumers to help them achieve things that are not relevant to the core product set of the brand. For example, a beer brand may help you get the album or concert ticket you were looking for by giving you prioritised access or even subsidising the download cost. A high street bank may provide user access to sophisticated budget planning or tax applications (or subsidising access to third party applications).

The problem here is that users will only tolerate this degree of ‘closeness’ from a very small number of chosen brands. Google may well be one of these and they may help to package up brand relationships to provide ranges of data and functionality to suit different user groups. However, it is highly likely that far fewer brands will ever achieve this level of intimacy and permission from users which inevitably suggests either massive brand consolidation or a fairly brutal hierarchy of partner brands versus the rest.

How will your brand position itself to users behind the Google walled garden?

Hold the front page....Man has turkey sandwich!

Google released real-time search this week, incorporating Twitter and kin. So we at Greenlight took it for a road test. We evaluated what the pros and cons are for users and for marketers.  Amidst all the hype, I can’t help feeling the usefulness of the results seems to range from the inconsequential, to the surreally disappointing.

Admittedly these are early days, but in our first road tests, which happened to be on searches for ‘Turkey’ i.e. the country not the game fare, the impressive list of relevant results, which included Lonely Planet, The State Department, The CIA, The Turkish Tourism Ministry, had somehow been permeated by the resounding news, that ‘thoughtbeast’ (presumably an alias), was enjoying a sliced turkey sandwich for lunch. With milk and Doritos, no less!

Ok, fair enough. Seasonally relevant perhaps, but way off the mark. Having been in search marketing for 10 years, and seen all the Q&A that goes into search results to ensure they aren’t ravaged by nonsense, I am forced to ask just how far we have come, if ‘thoughtbeast’ and his late lunch can pip the CIA for a page 1 spot on a one word search term of complete parallel. Already sceptical, we are now immediately unimpressed.

Other searches produced similarly mundane and useless results, this time in the brand space of certain large household names.

A search for ‘Tesco’ and ‘Asda’, (for me at least) revealed little else, other than how useless and trivial the social sphere can be sometimes, and just how undeserving some of this stuff is, of inclusion in search results. So initially for me, it’s just not groundbreaking. 

The majority of these inclusions are either just trivial pieces of inconsequential chat, or they are citations of the sort of information that is already available on the Google news results window that has been there for some time now.

Take the very moment of ‘Tiger woods’ (see below). The news results do their job. They are well placed at the top of the page given the current relevancy of the topic. Then more news content, plus a little bit of useless social chit chat.

At which point we’re forced to ask just what value it actually adds. Why not make the Google News insertion real-time if that’s what you want to do?   

Fools/under pressure public companies, rush in

Given Google’s efforts to organise the worlds information, this live deluge of yet more news, interspersed with random transient chit chat feels very un-Google. It is more like a knee jerk first attempt to get something live to compete with Bing who for once, (enjoy it while it lasts folks), has got the jump on Google, in getting real-time search in there first. Arguably, Bing’s done it in a more relevant context too.

There is still much debate as to whether social comments like this are worthy content or just references to them. Bing’s approach to listing the most socially cited links seems far more useful when compared to Google’s. Google’s  feels more like those times when you would pick up the phone and suddenly end up with a crossed line, and find yourself sitting there listening to two randoms having a natter about something.

If you really want to present an overview of the information available in the social sphere, there are better ways to do it than this. A ‘tag cloud’ might do a better job.  Consider Greenlight’s mock up (below), of the continuing ‘Tiger woods’ story which we have created based on keyword analysis of Twitter results. A ‘tag cloud’ could organise the myriad of chat and disparate nuggets of information telling  person searching what the key trending topics are in the social sphere, and allowing them to navigate to an aggregated and specific view of the collective tweets relevant to the tag.

In my view, search has become too obsessed with presenting everything in one multimedia interface. Whilst this is a fairly noble pursuit, users are not overtly concerned with having to make their way to information as long as they get what they want. A more useful approach is definitely required.  For the non-tech savvy, ‘Tag clouds’ are not perfect, but even a simple list of popular sub-topics would make this more useful.

No doubt this isn’t it for real-time search. It’ll be improved on.  However, the bigger issue for real-time search is what search engines do with it. Tweets on the whole are generally either citations of other content, messages, or publicly aired views, so real-time data isn’t always useful to present, although there’s no doubt it’s a very effective barometer for what’s of the moment.

Search engines need to better work out when they need to use real-time data to create better results, and when real-time data, actually is better results. It isn’t all one and the same.

Implications for marketers and search marketers

The validity or quality of real-time content aside, what this is doing is adding yet another window of 3rd party content into an already crowded search engine results pages(SERP), pushing down some natural results, and creating another opportunity for visibility and/or detraction by third parties.

The big issues are:

1. Where is this likely to appear in keyword terms?
2. How does it affect the control we have, or want to have over those spaces?

In our daily rank checks at Greenlight, we monitor hundreds of thousands of keywords. We identified real-time results in less than 0.5% of them, and most of these were brand oriented keywords.

This leads us to conclude that for the time being, and as with most ‘universal search’ elements, this is more likely to pop up in your brand space, rather than in generic terms.

In other words, a brand like Tesco is more likely to see this popping up in searches for ‘Tesco’, than it is in a search for ‘Samsung 32 inch lcd’. This means that the impact on most generic and product-led search activity will be minimal, unless they suddenly gain significant and concentrated social attention. 

The bigger issue is reputational, since there’s now a real-time window in your branded search space for people chatting about you however glowingly, controversially or mundanely that might be. This creates further reputational issues in your branded search strategy.

Opportunities

The right fostering of the social community in things you do in real-time, can make their way into search far faster than they would have done historically. For example, you have a special offer, running a sale or promotion. These can now work their way into search results. As such, there is definitely a positive angle for this integration if it makes its way into your brand space.

What is interesting is whether this may actually have an impact on branded terms that are part of your paid search strategy.  Pay per click (PPC) was previously the traditional route for rapid response or timely tactical promos. If the social sphere gets wind of your latest promotion, then it’s quite possible that natural real-time results could end up winning more brand traffic share, which in turn could lighten the pressure on branded spend, and act as a more compelling click area than the many competitors trying to hijack traffic in your branded search space. Worth noting is that if this happens, it will make full click path conversion tracking absolutely crucial, as it may appear that your branded clicks and conversions from paid search are declining.

Threats

The bigger issue is that there is now yet another window for less than favourable conversation and discussion, to make its way in to your branded search space. Of course this isn’t a new thing. Universal Search has been active for some time now. Images, video, news, even blog posts, as well as traditional listings, are all well known loose cannons that need to be continually watched and tied down.

What this does is add one more, and underscores a long lamented need to develop a strategy to tie down your branded search space, and stay vigilant for any incursions into it. Some brands already do this quite well and will likely find themselves fairly well insulated. If you consider Vodafone below, there is good control of the branded space already, and so there is less of an opportunity for uncontrolled universal elements like real-time search, to make its way in.

If you consider a brand like KFC below, the brand page is littered with detractors in all forms and formats. For a brand with this kind of representation, real-time search is yet another can of worms to add to this, as an already open brand space filled with detraction can only get worse.

 

What else could Google do with real-time integration?

What’s clear is that real-time isn’t going anywhere. But the integrations thus far, are not really doing it justice.

As a final point, we thought we might as well go blue sky with a few ideas that might make good real-time integrations.

Key listing integration: If you have a twitter account that is associated with your company, you could opt to have its feed included in your key brand listing. This would be an enormous boon for brands who could then use the brands coverage in natural search as a real-time tactical marketing tool

1. Twitter listing integration: If you’re going to list a twitter page in search results, then a feed of the data in the result would do its listing more justice, than simply the biog line. Again, if a brand’s Twitter account can show in its branded search space, brands have a real-time voice in to natural search results, enabling them to deliver real-time messaging to their customers, and help combat any detractive comment.  
   
2. Real-time categories: Real-time data is disparate, and a 6 line scrolling feed of random ‘as it comes’ chat, is far from ideal as an interface. Categorising the noise into determinable topics of possible interest alerts individuals to the real-time aspect of what they are looking for, allowing them to further investigate the item that’s most relevant to them.

Perhaps the ideas expressed above are already in the pipeline. If not,  they would most certainly add more value to both users and marketers alike, and provide for far more credible and useful real-time search capabilities compared to what is currently available.
                                                                                                                            
Again it’s still early days. As such, the mantra for now is still the same as it ever was for marketers, defend the hill, and keep watch. If you don’t have a media monitoring strategy and action plan in place, it’s time to get one, because social media integrations in search can now give quite a bite.

The News Industry and Search Engines – Two camps emerge

Google has unveiled ‘The Living Stories’ prototype, a project in conjunction with The New York Times and The Washington Post.  It explores a new way to get news to the masses online, which Google hopes other publishers will want to exploit with them too. The central concept is that particular news topics could be centralised under a single URL, updating in real time or near real time, to reflect what’s happening out there and how the respective subject, issue or story is evolving. Whilst these pages currently live on the Google domain, they will be migrated over to The New York Times and The Washington Post early next year. Pages include one about Health Care, Global Warming and other major topics. 
 
Google has essentially found a way to try and bring publishers on-side by offering them free search engine optimisation (SEO), and what in his view, would appear to be a win-win situation for all involved.
 
This marks a significant collaboration between Google and some major heavyweights in the publishing industry. It is immediately very different in substance and sentiment from the Murdoch media empire’s cynicism with regards to Google and how it is an unwelcome entrant into its business, going as far as expressing the desire to move to a paid model, partnering with Bing and exiting its content from Google’s index entirely.
 
Of course only time will tell of which approach will gain the most traction.

Google’s ‘Web History’ - Tailored to fit or exclude, is the question

Google has launched ‘Web History’, the new name for its personalised search feature, by default across its user base. Essentially, by monitoring what you click on in their results, Google can learn what sites you like and give them a ranking boost in your search results. This adds a further dimension to how Google ranks sites and pages, which had historically focussed largely on analysing on-page relevancy and third party links pointing into a site. Some of the advantages and disadvantages this will bring. For example in the case of the enterprise, new entrants online will find it more difficult to break into Google page 1 results if the sites a user has visited before have been given a preferential boost. On the other hand, for those firms that customers like, Google’s ‘Web History’ will enhance their ability to cross-sell new products far more effectively. It also calls into question just how informed a decision consumers might make given that it could be argued Google’s ‘Web History’ might inadvertently restrict the breadth of sites delivered to them.

 

Advantages

• Google’s results will contain more of what you like. It will be tailored to fit in with how you typically interact online. As such a searcher should find what they are looking for faster and be more satisfied with the results. The ordinary searcher will theoretically come to think of Google as delivering better results

• This opens up the future possibility for sites to receive a boost in a searcher’s results if their friends like a site. Whilst this isn't happening right now, this would be entirely possible and make search ‘social’ for the first time. Imagine searching for ‘restaurants’ in Google and being presented with restaurants that your friends really like and therefore essentially recommend

• Allows companies which consumers really like to cross-sell their new products far more effectively as they will appear higher up for searches the consumer makes for things that they weren’t aware the firm offered

Disadvantages

• This might restrict the breadth of sites that are delivered to the user leading searchers to only see sources of information that they typically agree with, which deals with cognitive dissonance in a detrimental way for society at large. Consider how this might affect search results for Barack Obama if he searched for ‘health care reform’ or Al Gore if he searched for ‘climate change’. One of the great benefits of the Internet was that it would allow for a multitude of sites to be presented to a user, broadening his/her horizons and forcing him/her to question their positions and beliefs. Search, philosophically at least, should accentuate that and not diminish it.

• Small businesses that aren’t as well known as the bigger brands won’t be clicked on as much and won’t then get the opportunity to appear in results for future searches

• New entrants into a market would find it more difficult to break into your Page 1 results if sites the user has visited before have been given a preferential boost

• Many will have privacy concerns if Google is collecting and using this much browsing information. This is an opt-out however, which should go at least some way to alleviating some concern

Much of the above will depend on how aggressively the Google’s Web History functionality is implemented. For example, many of the above concerns would be alleviated and perhaps entirely dispelled if only 1 listing out of 10 in a search results page was boosted based on one’s web history. Conversely, too few listings boosted in rankings may well negate the value of the innovation entirely. To determine this, we at Greenlight will be analysing how Web History behaves going forward.