First off Google loves HTML 5, they have already dropped Gears and said go HTML 5, go! However this does not necessarily mean they will be sharing that attitude when it comes to their search product.

Google knows that if they start relying on a factor which is easily manipulated and not widely used that could be a recipe for disaster. They have to look at not only could how this be used but how could it be exploited from a Blackhat perspective.

Working with big brands I see some awful bits of web design. That’s not to say that every brand website is bad, but a lot of them are. Especially big B2B sites, which usually are stuck in about 1995 (OK slight exaggeration!). These are slowly being dragged into the 21st century but will these be getting updated with HTML 5 anytime soon? I doubt it.

Google sees brands as the way to keep the search results clear, that’s why things like authority is such an important factor in the rankings. That’s also why Google structures its algorithm so that the most trusted result will be shown over the most relevant. With this in mind I personally don’t think that Google will be relying on HTML 5 tags anytime soon as they know that they can’t rely on the big brands to implement them.

This is just my speculation so I wouldn’t go taking this to the bank as it were, but feel free to add your two cents to the mix.

The solution seems to start at home, or at least in the agency sense! Knowing the full spectrum of services your agency offers and how these integrate into what you do is essential. This means you can spot opportunities for integration from the very first meeting with the client and as you build a relationship with them. This also makes a very strong case for giving you the business for the additional services and not another agency. Additionally this can be used the other way to get SEO business out of clients your agency does other services for.

A core case study to read up on and site to clients as often as possible is the Orange “I AM” campaign. Read up on this and go forth and integrate!

“When you enter a longer query, with more than three words, regular-length snippets may not give you enough information and context. In these situations, we now increase the number of lines in the snippet to provide more information and show more of the words you typed in the context of the page.” See the Google blog for details.

This is to better serve long tails by showing the context of all the keywords used in the long tail search. So what about the description?

Well looking at the example below we can see the Description has clearly been replaced in the SERPs by the more relevent on page content:

The top search result (ignoring the misplaced web definition result) has content taken from on page and not from the description, which does not contain the keyphrase.

So what happens where there is the keyphrase on page and in the description, what will Google pull? Well it appears to be the the on page content which wins again, so forget that nice clean description tag you had with an exact match:

Now Google are increasing the leangth of that snippet where needed to show the user the context of the information within the SERPs. The problem with that is the user no longer needs to visit the page to find the information they are looking for. Google will helpfully find it on page and show it to them in the SERPs. This could be a problem.

I took a range of data directly from two sites, this consists of the Total Unique visitors for the domain, Organic new visits and US only visits. As I don’t know how these tools work taking a range of data would be more likely to yield good results.

The results show a surprise winner in the form of mydnstats.com. Although all these guys do is just pull their data from compete and alexa rank the result is a surprisingly accurate match for US traffic, more accurate than the individual sites they pull the data from.

For total traffic trafficestimate.com seems to give a quick and handy guide to work from. This is a very clean and easy to use site that unlike mydnstats loads your data almost instantly. At the end of the day this is the one that I’d use, unless I was looking for specifically US only traffic levels.

Please feel free to use the comments section to add in your own stats for these sites and if you know any other traffic estimation sites why not throw them into the mix.

So what the hell was going on? Aaron Wall over at SEOBook wrote an excellent well thought out and factual post about what happened. If you want to find out about the ins and outs of what happened I suggest you read it.

The question for me, now the results seem to be holding and the hoo haa has died down is; Why did Google do it? From a public perspective it appears that it is Googles attempt to clear out the SERPs. Eric Schmidt has made his feelings more than clear on this in the past:

“the answer won’t come by keeping brands from the space; instead, the web needs them to help clean up the net.”

What Google is saying is that brands are a safe option. By pumping these to the top of the SERPs they know they can keep out spammy, blackhat and generally not that great pages. The reason this has become so important all of a sudden lies within universal search.

The direction that Google clearly is pushing in is that there will be more types of search result shown on any one page, this will mean that there is less room for text results. Google wants to ensure the clarity of these very precious top spaces that will display on the first page. They believe the way to do this is ensuring brands takes these slots.

As an SEO I’ve already been talking to my clients about the importance of building a brand online. Whilst I believe there may be some manual intervention in the SERPs from Google at the moment it is simply not scalable unless done algorithmically, hence the need to build brand and so trust.

Domain age, link neighborhoods, CTR and mentions are just a few of the aspects I think Google is going to be relying all the more heavily on. On page factors importance will deteriorate and so will the wonderful variation of sites in the SERPs. Google is essentially cutting off the lifelines of the small time retailers, who until now had been given a reasonably level playing field. Now that’s gone so will the variation it allowed. It’s been sad enough watching our high streets clone each other and all the independent retailers disappear, now it looks like online is going that way too.

This means that SEO as progresses into a new richer format the way in which we optimise will change. Text results are already less important, this effect is only going to magnify as Google choses to expand Universal Search. As more and more content is pulled into the SERPs there will literally be less room for text results. These could eventually be treated as a format of contant or media type themselves, so Google may only choose to pull in one per page.

Just look at how rich the SERPs are getting already ( I know the wikipedia results is a bit dubious but it just ranks so damn well I couldn’t count it as a ‘normal’ result:

Google have started using Ajax in search results as well. This says to me that they are looking at ways of making the results even richer. By using Ajax the SERPs could handle more information quicker, allowing content to be pulled in from all over the shop.

What we currently have is a situation on line which is a little like the wild west when it comes to trademarks. This isn’t going to last forever and there seems to be very little discussion of what all this could mean for SEO.

So taking an example of a trademark branded item if I want to write a page all about my favourite camera it’s not a problem. If I want to set up, for example a camera site it’s not a problem, but in the future it really could be.

It’s a page all about the Sony Cybershot. My page title would contain the trademarked terms Sony and Cybershot. Well if a court decides that using someone else’s trademark terms in your URL is actually trying to pass yourself off as the brand themselves. ‘Passing off’ is a major part of trademark law and is where one company too closely resembles another so the consumer may become confused between the two.

Now lets say my page URL is www.sonycamerasonline.com/cybershot (that domain is available if you want it) it does look like a trademark layer would have a pretty strong case for passing off. Especially if you selling the cameras as well or have an affiliate link to another organisation, say Amazon which do.

Now if this did ever become the case how would Google prevent this? Devaluing the terms within the URL seems the only way to go, so no longer will this count towards the page relevancy or maybe even appear in the search results.

There are massive gaps in this theory I know, but it is worth thinking about just how trademark law could affect organic search, it may not just apply to Paid after all.

The case is basically is it query or content driven;

Google started out severely restricting the queries that triggered the blended search, hence the Darth Vader search that initially circled round the net. This was query driven, in that there are many other searches that have this kind of multimedia content that could be pulled in, but it was only a handful of key phrases that pulled in these blended results.

As time has progressed more and more searches have been pulling in this type of results, a mix of image and video content. Now it can be seen that a range of searches are pulling in this type of result there is a strong case to say that this is content driven. In that it is not based on what you’re typing in to Google, but what Google can find that is relevant to it.

It’s recognised that views and possibly even ratings have a major impact on what videos get chosen, the relevance being taken from their description. I don’t think the URL has much significance here seeing as the majority of videos come from YouTube which has generated URLs which have no relevance to the content.

I think there is also a case for inbound links being important here as well, with anchor text and relevancy being taken into account for the page the video is on.

This is all pointing to Google using very high levels of quality control for universal search as they move from a query to content driven model. Only exceptionally relevant and popular videos and images make it into the search results at the moment, otherwise the SERPs would be riddled with them for every search.

I managed to wangle Tuesday off work to attend SES London, it was an interesting day on the whole so here is my experience. This isn’t a run down of the sessions though, as you can get that off the website.

Kicking off with a talk from Matt Mason based around his new book ‘The Pirates Dilemma’ which was a great energetic start to the day. I heartily recommend a trip to his website where you can download the ebook for free (although a small donation is always polite!). I have done already and after reading a few pages I’m about ready for a trip to Amazon for a hardcopy.

The sessions are based on a track system; Red, Green and Blue. This seems like a good idea as they can just theme the tracks, so beginner, intermediate, advanced or whatever. Well that would have been the logical choice but there didn’t really seem to be much correlation between the tracks and the sessions within them.

The greatest frustration for me was the lack of distinction between paid and organic. I’m not saying that for me having a knowledge of paid isn’t useful but a talk which hops between the two and so only skims over the point on each rather than going into depth isn’t much use.

It did feel like there was a massive lack of technical focus on some sessions as well. I was really looking forward to the Universal / Blended search talk. Some interesting topics were skirted around but with hints and tips like, if you have images you should optimise them. Yeah, wow, thanks for that one. I wanted some technical nitty gritty and the lack of time at the end for questions killed it for me.

I’m not by any means saying don’t go to SES, because it’s cool and you should. Seeing Rand Fishkin sitting in the audience and giving the Orion Panel some things to think about was awesome.

Also there was a great talk at the end of the day, ‘Why does Search get all the credit’. It was great one to get you thinking about how you can separate out what has driven traffic to the site and where search sits in the marketing mix

So all in all, I’d still recommend attending, but maybe bring a cushion and your laptop, as the seats are rock solid and breaks long and uneventful.