It’s festival season and I have been browsing here and there getting all excited about seeing some great bands and having a few beverages in the sun rain.

Anyway, NME is first for music news (apparently) but it also seems their forums have been totally nuked by spam. The main forum boards has a nice toolbar PR of 6.

All forums are subject to a little spam but what gets me is the location based forums have had this problem from at least July (I got bored of going back) so it seems nobody at NME is really paying attention.

Forums are a great for the stickyness of your site - building relationships with visitors and having a captured qualified audience for the commerical side. This forum should really be huge in the current climate.

So why have NME left it to rot exactly?

There is still some debate whether the search engines place any weight on keywords used within page urls and that’s actually what I wanted to quickly test. So I set up a couple of pages from this blog and included unique words within the page url, one separated by hyphens (dashes) and the other by underscores. Obviously the unique ‘words’ were not used onpage or anywhere else including linking to the page. The pages were set up as separate pages rather than posts, to ensure they were not syndicated or linked to from anywhere else.

Although a very rough kind of test, I wanted to see whether using keywords within the url only would be enough to pull back the page by itself under search. Second, I wanted to see whether Google is treating underscores as word separators as suggested they might some time ago.

Hyphens (dashes) In URL - /test-ljkdldlk-hdhdhdjqe/

• Searching for the two words with a space (”ljkdldlk hdhdhdjqe”) brings back the page
• Searching for one of the words individually (”ljkdldlk” or “hdhdhdjqe“) brings back the page
• Searching for the two unique keywords together without a space (”ljkdldlkhdhdhdjqe“) did not bring back the page

Underscores In URL - /test-lhdldkjjgjk_lkhjhnues/

• Searching for two words with a underscore (”lhdldkjjgjk_lkhjhnues”) brings back the page
• Searching for one of the words individually (”lhdldkjjgjk” or “lkhjhnues“) or together with a space did not bring back the page
• Searching for the two unique keywords together without a underscore “lhdldkjjgjklkhjhnues” did not bring back the page.

Conclusion

First of all, it looks as though hyphens and underscores are still treated very differently by Google. Currently underscores are still not being treated as word separators from these tests. The results match Matt Cutts comments from back in August 2005 on how Google would return the pages under search.

By no means do I think this was the best test in the world, but it does suggest that Google do give some weight to keyword use in urls, no matter how minor the weight might be.

Sometimes its good to analyse these things in context aswell - We know already that Google and Live seem to completely ignore the use of meta keywords (Yahoo and ASK give them some weight), while keywords used within a meta description alone are not enough for Google to retrieve the page under search (only with the use of anchor text aswell). So this gives us some idea of the weight keywords in a url may have.

Ok, so I was just reading the official Google blog today about their new 3D virtual experience product called Lively when I noticed some big red lines.

Anyone who uses the SEO for firefox extension knows that red means the ‘nofollow’ attribute is being used on a link.

So what are Google applying the ‘nofollow’ attribute to on their own blog?.

Their own ‘What We’re Reading’ list, which includes respected industry sites such as Phillip Lenssen, Search Engine Land, Search Engine Roundtable & Traffick amongst others.

Can Google please explain how that can make sense?.

Google are actively saying they read these sites, they get value from these sources which they must respect enough to link to in the first place. Yet they won’t allow any pagerank or link juice to be passed unless its to another Google property?

Googles nofollow advice from their help center in summary -

• Untrusted content (comments, where there is a lack of editorial control)
• Paid Links
• Crawl prioritisation (’Sign in’ links etc)

So which of these exactly does their own reading list fall into?.

The Hypocrisy is scary. This is exactly how the nofollow attribute should not be used.

Say One Thing, Do Another

Use ‘nofollow’ for sites you can’t vouch for or necessarily trust is how Matt Cutts has always expressed it.

“In an ideal world, nofollow would only be for untrusted links.“

So why do Google add nofollow in this case on content that is editorally controlled?.

Matts own words on what makes a great link -

“So, what are the links that will stand the test of time? Those links are typically given voluntarily. It is an editorial link by someone, and it’s someone that’s informed. They are not misinformed, they are not tricked; there is no bait and switch involved. It’s because somebody thinks that something is so cool, so useful, or so helpful that they want to make little sign posts so that other people on the web can find that out.“

The funny thing is, Googles own link based algo is what made them the biggest search engine in the world and yet they are contradicting it by creating a FUD campaign that even their own employees have failed to understand.

Google created the directive, they created the rules on how it should be used and yet they don’t adhear to them. Taking away the hyperlink structure of the internet is destroying themselves in the process.

Interesting experiment by the Neutralize Guava guys over on Search Engine War blog where they tested and proved that Google is now indexing links in flash.

Next up for them to prove is whether text in a flash button acts as an anchor on those links…

Pretty big achievement by the Google/Adobe engineers, flash still won’t offer the same kind of information html tags will but it is a start.

Anyway, good news if you were of the mind to have a site that needs to hide the odd 10K keyword rich article with some nice juicy links own a flash site.

Anyone thinking free flash intros, movies/widgets?.

I always wanted the ppcblog.com domain, but unfortunately it was registered before I got the chance.

At the time, the .net & .org were available aswell as the .co.uk and nobody had built a brand around the name ‘PPC blog‘ in the same way there were seo blogs seemingly everywhere.

About 2+ years ago when I developed this site for a little fun the ppcblog.com domain used to redirect over to payperclickblog.com and the site was aptly named after that domain. This was great as I always wanted to work with the shorter url and ‘brand’.

Aaron Wall of SEO Book is however the owner of the domain which his wife Giovanni and he have decided to develop in the past week, seperate and away from the payperclickblog.com domain.

Aarons SEO Book is a fantastic resource, so I have no doubt this new site will be equally full of great regulary updated content. Regular enough to stay in the Toprank search marketing blog list (ping) which I got kicked from for not updating regulary enough…

As I have never really had a goal for this site other than to share random thoughts, have a nice testing platform for ideas or vent frustration it doesn’t bother me that it shares the same name. But it does highlight the importance of getting all the TLD’s you can if you want to establish yourself globally and block out competition.

Geographical filters are pretty strong, unless you are a high authority domain and in my view you still can’t rely completely on Googles Webmaster Central (and others) to set geotargetting for the sub directory/folder route. Getting all relevant TLD’s for each country in my opinion can still be the safest way to go. Which I might expand upon in another post as I am going off subject.

Anyway, so don’t get confused guys with your .coms and .co.uks - or do and I might get more traffic.

Another example of a Wordpress blog being hacked and injected with hidden links to viagra, tramadol and the like. Their blog is on a subdomain of the main site - http://inside.123-reg.co.uk/

If the self acclaimed ‘UK’s largest web host’ can’t protect their own blog or even spot they have a big problem, what hope is there for the average website owner?. I’m guessing this probably isn’t helping their domain rank.

Maybe they will give me a free hosting account for letting them know about the problem…

Maybe not, lol

Btw - Patrick at Blogstorm suggested using Google alerts to keep an eye on hacking.

This question seems to come up a lot still and while I believe sitemaps in general can be great for helping search engine spiders crawl your site and an effective way of spreading out linkjuice (with some nice anchor text) they can also bite you in the arse when not used intelligently. Certainly submitting sitemaps specifically to the search engines in my view can have its drawbacks.

Still Won’t Rank

The key reason for submitting a sitemap is to help Google crawl your site and find those deep pages. Now those pages may well be included in the index because of the sitemap submission, but the problem is they are still never going to rank for anything. There is a reason why they were not in the index in the first place and Google will simply not start ranking those pages.

Hide The Truth

In many ways sitemaps hide problems or issues with your website. Personally I would only want pages included within the index from my site that ’should’ be there naturally. Data you get from using the basic site command in Google is more valuable as you can analyse pages or areas of the site that are having problems being crawled and indexed. This then helps you identify problems with site architecture, internal link structure, content etc and resolve issues properly which will help these pages rank.

Overall Site Quality & Trust

Often pages that are merely included in the index because of a sitemap will or can fall into the ’supplemental index’ results (its still there) which are low pagerank less trusted pages. Now I certainly don’t like the idea of having too many of my sites pages in the supplemental index and showing Google more of your content can mean just that. Instead of Google believing you have 100 pages and only 10 of those are in the supplemental index, your site could now have 200 pages with 100 of those in the supplemental index. So Google is now looking at your overall domain and classing that 50% of your site is not in its view ‘quality content’.

What can that do overall to your domain in terms of authority, trust and rankings?. In this way, you can give too much information to the search engines.

Huge sites with masses of inventory will probably find some of the tools that Webmaster Central can offer extremely valuable, but for the majority of sites there are better things you can spend your time doing than worrying about submitting a sitemap.

It’s been a couple of days since Googles keyword trademarking policy change and it’s interesting to see many advertisers have already been slapped out of the bidding. In a quality score world, its not as simple as bidding against any brand due to quality based minimum bids. Although you might be able to display against any brand you like for a short period, Google will soon kick you into touch if you are deemed not relevant enough.

Patrick over at Blogstorm posted about the effects of the new trademark policy with screenshots of adverts against once trademarked brands such as Tesco, Asda, Lastminute.com and Amazon. But it was just time before many of them simply fell away…

As I mentioned previously in my post regarding preparation of the Google trademark change, quite simply, it will be extremely hard to be able to display against many core branded terms due to low click through rates which will mean you will get slapped with a high minimum CPC.

Lets take a look at those same brand examples today -

First and foremost, there has been a massive reduction in adverts as expected. Some of the examples may have been rather bad ‘advanced matching’ from Googles broad match feature otherwise the adverts left within the bidding for the above examples are either -

1) Maintaining a high enough click through rate (and therefore quality score) to keep a reasonable level of minimum CPC. For example, notice how Lastminute.com are having more trouble protecting their brand as many competitors can get away with using ‘Last Minute Holidays’ within the advert title which will be helping to maintain their CTR and advert alive.

2) Advertisers within high CPC markets will find the high minimum CPC’s less of a problem than those in low CPC markets. Advertisers in insurance, loans, mortgages etc markets already pay a high cost per click (£5-£30 max CPC’s), so a high minimum CPC of say even £10 might be lower than their usual average CPC anyway. So for example, Tesco has just a couple of loan related advertisers who can pay that price.

For the above reasons, brands like Amazon will probably have less concerns with competitors. Some of the main areas effected are highly competitive markets with high relevance between brands and product offerings where consumers actively comparison shop for the very best deal. Brands within ‘car insurance’ for example have been hit hard.

These guys pay high CPC’s and have a relevance/comparison shopping nature -

I just spotted Google is showing onebox results for queries relating to English football clubs. A simple club search query brings back their next fixture. Screenshot below -

Only seems to be Premiership clubs currently. Cool.

Update - Google have now added last game results alongside the next fixture. As below -

Ok so the 5th of May lift on Google keyword trademarking date is nearly here.

Are you prepared? Are your clients?

Are you going all out aggressively to bid against competitors trademarks?.

Are you going to start bidding against your own brand as protection?.

Perhaps a mixture?. Or are you going to wait and see what your competitors do?

Looks like Tesco have taken the ‘moral high ground’ issuing a statement that they will not be bidding against anyone elses branded terms.

Will this moral high ground mean that they get less competition against their own brand?.

Lets see.

The key thing to remember here is ROI. You can go and bid wildy against any brand you like if you are willing to pay through the nose for it, but with a lack of relevance and ultimate conversion its expenditure you can do without.

Lets not forget, Google has quality based minimum bids, so although you may think your brand is relevant, if your click through rate is not high enough your minimum bids will get very high. Trademarks are still in place for adtext and this is a massive factor in click through rates.

Against core branded keywords advertisers will find CTR will not be high enough to allow bidding against brand unless you are willing to pay extortionate CPC’s. This is because core branded keywords contain a high number of navigational queries with lazy searchers who would rather search for the brand than type the url directly into the browser. These searchers can be blind to anything other than the brand they searched for.

I recently set up a competitors adgroup for a client with these types of competitor terms. The competition was directly related selling the same product and all keywords were on phrase match. However, as expected CTR was simply not high enough to keep them active. Unless Google lower CTR thresholds you will see this -

The Hitwise blog highlights the gap in brand traffic lost between the US and UK (where the US have always been able to bid openly against keywords), so there is brand volume to be had from competitors. It just needs to be taken from longer tail queries and more inteligently than simply bidding against core terms.

Another key thing to remember is the relative nature of Googles bidding platform.

If you are bidding against your competitors brand as the only advertiser against the brand owner, then your CTR will be poor in comparison. Google takes into account all advertisers against that keyword. So if you bid in a pack with 8 other advertisers who will also have relatively poor click through rates then the quality threshold might be a lowered a little allowing for lower minimum bids.

So will the threshold naturally decline over time?