On the surface we’ve been tracking a number of good metrics for years, but at the same time it feels like there are a lot of assumptions being made in our reporting, and that there is miles of room for more data to enable better actual analysis.

I’ve been re-reading through a pile of articles online, all the way back to basic “Analytics 101″ type posts, to refine my thinking on what what we currently collect, identifying what other data we have available and could use, and what data we want but don’t have. These last are the things which then become tasks for our WebTrends guru – because we certainly have the ability to assemble new reports if we know what we need.

Other articles I’m referencing:

• Choosing SEO KPIs and loads of other things on Occam’s Razor
• the recent Analytics for SEO Whiteboard Friday video
• post after post on WebAnalyticsDemystified
• and this massive, headbreaking post about what Google likes.

Part of the challenge is of course understanding why various bits of data are useful, and learning how to interpret it in meaningful ways.

It’s going to be a task, for sure.

6 Things to Expect if Google Decaf Gets a “Caffeine” Boost
A pretty thorough comparative dig into Google vs. Google. When Caffeine rolls out of beta and then migrates to the UK, what can we expect to see?

Also did a whole bunch of reading up on HTML 5.
http://www.w3.org/TR/html5/
http://dev.w3.org/html5/html-author/
http://www.seomoz.org/ugc/7-html-5-elements-that-will-make-seo-more-enjoyable
http://www.searchengineoptimizationjournal.com/2009/06/22/html-5-seo/
http://www.ibm.com/developerworks/library/x-html5/
http://www.sitepoint.com/blogs/2009/07/24/google-html5-and-standards/
http://net.tutsplus.com/tutorials/html-css-techniques/html-5-and-css-3-the-techniques-youll-soon-be-using/
http://www.alistapart.com/articles/previewofhtml5

…the last one being a particularly useful writeup, explaining in pretty good detail but in a nicely readable way.

All this reading was prompted by a sort of ambiguous requirement included in our ongoing CMS migration – restructure code so that SEO-valuable content is at the top .

On the surface it makes sense – get the central content box as high up in the code as possible, move all the excess crap to the bottom, and position it all onscreen with CSS. A strict interpretation of this, though, is a nightmare when using shared resources across multiple sites, and causes all kinds of headaches for accessibility. A full code revamp is happening anyway to better linearize the code for accessibility, and the question came up whether coding to HTML5 standards would be sufficient. If a search engine discounts boilerplate code, can we assume that the HTML <nav> or <footer> tags will be recognized and equally discounted, regardless of where they are in the code?

Clearly nobody yet knows. What is the actual SEO benefit of HTML 5? There is absolutely logic in thinking that a cleanly organized page which essentially follows a DOM tree will be more easily crawled, and the content within more easily parsed and indexed by a spider. But that’s completely speculation at this point…and as the HTML 5 spec is still very much in draft mode (and likely to remain in development for years), we seem likely to make a stab without really knowing at all.

You see this all the time, people bragging on digitalpoint that they’re #1 for their target term. What they never say is quite what that term is.

In the quest for traffic, you can aim for a slice of a competitive term, or you can look for opportunities with little or no competition and build your own market. Ranking number 1 is easy – the value is in getting people to search for that term.

I’ve had a music blog for a while now, occasionally showcasing some new band I’ve found. The name derives from the title of a song by a band I like. I looked up this morning and found that I am now sitting comfortably at #1 (and #2), overtaking all the sites referencing all the lyrics sites and the band itself. Well, that’s great. But what’s it worth? Less than a visit a day.

So how do I improve on this?

• More regular updates – my wife updates her theatre blog at least a couple times a week. More updates mean more crawls and more visibility for more terms. Individual posts easily land on page 1, and searches related to the shows we’ve seen drive far more traffic than do searches for the name of her site, which is also comfortably number 1. Relevance isn’t just about keyword usage, it’s about being timely and topical as well. I admit, I haven’t posted an update in over a year, so I’m losing out on the fact that fresh content encourages traffic. One of the bands I wrote up 2 years ago is now up for a Mercury award. When I first posted the review I hit page one with minimal effort. Maintaining the momentum in the intervening time could have paid off mightily in terms of traffic.
• More visibility – more content will generate more readers and more links and ripple into more traffic. New bands are hungry for good press, and getting a post referenced on a band website or myspace page can mean thousands of new eyes seeing the name of the site as a viable source for information. There’s also the knock-on effect that more readers mean more people who know the name of the site, and thus more likely to search for it.
• More links – generated through more visibility, and through more active self-promotion. It may be a term nobody searches for, but so was the word ‘google’ at one point. Building the site as a brand means building the awareness of the brand term, which in turn will encourage more searches for the term and more traffic associated to it.

Ranking number 1 for a junk keyterm isn’t actually valuable by itself, but in the process of building a site as a brand, sometimes it’s a good start. Now that I’ve got the top spot, my challenge becomes building the brand and traffic to it, which will be far easier than trying to start off trying to rank for a more competitive term like ‘music blog.’ If the site takes off based on the merit of the content, over the course of time it will begin to register for broader and more competitive terms on its own. I’ll end up with that traffic anyway. I’ve just come in through the back door.

So, clearly, the thing to do is to set up an account for your website, and then set up accounts for all of your other websites, and just have all the accounts watch all the other accounts, creating this vast network of interlinked twitter profiles .

Um. Is this just corporate use of twitter gone horribly wrong? Ignoring the fact that all these websites are creating one big link farm already¹, since almost everything on Twitter is nofollowed, I’m trying to think through the logic of this approach. Is it about visibility, is it about traffic, or is it just about getting the content indexed?

¹since these guys are consistently showing up as #1 in Google for ‘jobs in [location],’ what I see as very dodgy interlinking would seem to actually be doing them some good.

Gotta say, that’s some confidence there.

One of our brands has recently been pursuing a campaign with a similar angle – instead of a web address, we’ll prompt you to simply drop this key term into a search engine, and dimes to dollars (pence to pounds?) you’ll find our site at the top of the results.

This particular campaign is being managed by an external agency, and apart from dropping a link on the main brand site, the current Google rankings are being driven exclusively by their linkbuilding efforts – and the fact that they’ve accomplished the promised #1 and #2 spots for the two landing pages may speak well for them…though you know it helps that the term in question is absolutely unique on the web. It’s kind of like, "why does Flickr rank number 1 when I search on the word flickr?"  Maybe because it’s a totally made up word? That’s what this campaign has done. The real challenge is raising awareness of the term in the first place so people know to search for it.

Certain search marketers will have business cards which simply prompt you to search on their name. "Just google me – you’ll find me." That’s a little easier to control, perhaps, unless you’re named John Smith. Then, I imagine, it’s a bit more of a game. But I could do that, and I haven’t really been trying too hard. I’m just more active online than that guy in Tucson is, now that he’s stopped racing bikes and getting listed in the sports section every two weeks. But as a potential client, you want to be able to look up the person you are paying and know that they can do their job – and if most of the top ten results point to the same person, you get a pretty good feeling that they know how to play the game.

I have to say, though…I was impressed by the cajones involved behind a campaign that was relying on being able to perform based on something as generic as ‘I am.’ Even my jaded self was prompted to go plop down on the laptop and do the search – and this is where it backfires.

See, I use the CustomizeGoogle extension on Firefox, and I block paid ads. What do I see when I search on "I am?" Humorously, i-am-bored.com. Wikipedia. A London-based branding consultancy who you can bet wasn’t part of this campaign. I Am Legend. I Am Kloot. And a bunch of other stuff that has nothing to do with Orange mobile communications. Whoops.

Of course, when I enable ads or search on Yahoo!, there they are, right at the top…but how disappointing.  That’s just a matter of being willing and able to spend more money than anyone else, and once again for something that probably not a lot of people are either searching or bidding on. And I’m let down, just a bit, because somebody had some rocks to sell this advertising idea, but the execution, in my book, is far less impressive than it could be.

Here’s my first email back to her.

‘Online reputation management’ is a big issue these days thanks to situations exactly like this. I thought I’d send over a few articles which talk about it.

http://searchenginewatch.com/showPage.html?page=3628265
This is a good, broad overview.

http://www.stuntdubl.com/2007/07/11/reputation-management/
This is very much centered on *personal * reputation rather than a brand.

http://www.webpronews.com/blogtalk/2007/03/27/online-reputation-management-basics
This is very brand-focused, but it still applies: Your name is your brand.

http://blog.venture-skills.co.uk/2006/11/09/top-5-ways-to-establish-an-internet-identity/
Even if you’re established already, there’s more places you can be, over which you’ll have more control down the road.

http://www.seroundtable.com/tag/reputation%20management Links to a bunch of other articles which may be of interest.

The theme you’ll probably see is that t here are no silver bullets, but there’s definitely things you can do. The fact that those articles/comments are now 2 years old will probably help – feed the search engines some new information that’s more up-to-date, and the older stuff should fall away. You won’t get rid of them entirely, but off the first page of results would be a great start.

Think of it as a bit of an advertising campaign for yourself; like any campaign, there’s some thought and planning that needs to go into it, but it can be done.

I’m intrigued by this as an exercise because of the challenges involved. If she doesn’t own (and use) her-name.com then is there value in starting it up now? Possibly, but that alone isn’t going to take down the very well established site which contains – admittedly – some very valid commentary. So, what to do?

Look for specific ideas from me later.

We’ve got a handful of small niche sites that recently went live. Each site is a thin vertical slice of information aggregated from several broader datasets, essentially mashed up with other new and relevant content. The plan is for us to each take charge of one site, and within the limits of our acceptable SEO practices, go head-to-head to build links, secure SERP placement, and generate traffic.

Each niche is roughly equivalent in terms of specificity and search volume, and the only budget is time – there are no paid campaigs, so it’s seems a pretty even playing field. I’m thinking that we benchmark on several factors:

• overall page impressions + percentage growth
• page impressions from organic search + percentage growth
• Google SERP position on a pre-defined set of 5 to 10 key terms
• inbound links as counted by Yahoo!
• downstream traffic to the parent sites (this is the conversion metric)

Benchmarks will be taken at 2 month intervals for 6 months, with a lunch on the line for the mutually agreed leader at each checkpoint.

Dev resource and access to certain tools (Hitwise) is limited, and that’s a detail we need to work out before this can start, but there will be some ability to make changes to the sites themselves and do competitive research.

It will be an interesting challenge, to say the least.

Representatives from major crawler-based search engines cover how to submit and feed them content, with plenty of Q&A time to cover issues related to ranking well and being indexed.

Yahoo!, Google, and MSN all put themselves on the podium and each ran through an overview of their webmaster tools. Anyone who is already using these tools and/or keeping up on the industry blogs probably didn’t get much new out of it, though I did catch just a couple tidbits:

• Yahoo! will now accept the robots.txt sitemap.xml URL being on a different domain. I think this is pretty new, and useful for anyone who may have issues with getting things hosted on their corporate servers.
• The protocol for supplying a Google news feed is different than the standard sitemap.xml protocol. I hadn’t looked into this much in the past, so that’s probably not new, but it’s good information to have heard.

For once, the Q&A was the heart of the session, and after a bunch of standard “I have this very specific issue with my site” kind of questions, I couldn’t resist the opportunity to try raising a ruckus with a meaty one. I wanted to know how the engines were currently viewing the use of the rel=”nofollow” attribute on internal links to a website’s own pages. Though Matt Cutts has gone on record about it several times, I think there’s still confusion, so I asked. The responses:

• Sean Suchter of Yahoo! said they are not using a nofollowed link for calculating the “link quality distribution” score for a page. He did not specifically say, “don’t use it on internal pages,” but he did specifically say, “I would be wary of using it for sculpting pagerank.”
• Evan Roseman of Google did a marvelous bit of dancing and basically said “go find Matt’s post,” but also said that there were some situations where an internal nofollow might be appropriate. He did not comment on the use of it as a sulpting/siloing device. (The post in question is, I believe, on Matt Cutt’s blog here, but there are other more recent – and not always clearly consistent – quotes on SEOMoz, SEORoundtable, and Dave Naylor’s blog as well. In fact, there’s a lot of opinion out there, if you do as Evan suggested and Google: Matt Cutts nofollow )
• Nathan Buggia from MSN dodged entirely and simply said he thought there were better uses of your time and money than worrying about it. Which, interestingly, is pretty much what Matt said in Dave Naylor’s post, and with which, honestly, I have to completely agree…though he didn’t, technically, answer my question.

I was a bit amused that Evan referred to the question several times as a “pretty advanced topic.” Sorry, man, the SEO 101 session was three days ago, and I didn’t cross the ocean to go to it.

Search result multiplicity is not a new phenomenon, but recent advancements will guarantee the world of search and marketing will be changing forever. Before you attend this week’s optimization and best practices sessions, hear from industry gurus about how search, marketing and information seeking is changing the industry that follows the search. Our ongoing series on universal search will include research data available only at SES.

I almost didn’t go to this, and in fact showed up a bit late, but in the middle of a day of sort of uninspiring sessions, this genuine conversation in panel format ended up making me glad I went.

As I walked in, comScore’s James Lamberti was discussing a very interesting graph they’d built. Their research into universal search results showed a direct correlation between type of search result and clickthrough rate. In their model, if “no universal results” provided a 100% clickthrough rate, including video results showed a slight decrease to (I think) 98%. As more types of results came into play (images, maps, and so on), the clickthrough rate continued dropping, and result sets including “news” or “stock quotes” were showing less than 50% clickthrough.

Predictably, the Google rep-du-jour (Jack Menzel) then got raked over the coals for the rest of the session and spent a lot of time denying that they’d changed their business model. If they were intentionally providing information which did not lead people to click off the page, aren’t they then becoming a portal site? How are they going to monetize this, and how will that affect the downstream sites ability to monetize themselves?

Lamberti commented that the future value in search results will be not in the click, but in what is being displayed in the results. If people are clicking less, then it’s all the more important to be showing them something of value in the window of opportunity you have. The follow on question I have is a practical one: how do you measure this? Right now, universal results are showing for a small percentage of search traffic, but I have no idea if my sites are showing as part of an integrated SERP or not. I can’t get impression data for organic results, now, can I? No, I cannot.

The other Big Issue that put Jack on the hotseat was the fact that Google owns space in many of the channels now listing in the universal search results, and it’s hard to believe that there is no bias. YouTube has the most traffic and the most videos, but does that mean they have the best video for a particular result set? No. But the perception is that YouTube gets preference because Google owns it. Is it true? Jack insisted not.

Of course, with a G-man on the stage, the conversation was bound to focus there, but clearly Yahoo! and Ask and everyone else are taking their result sets in this direction as well, and in the theoretical or “big picture” view, the questions directed at Jack are going to be relevant to all. The final takeaway comments from the panelists were worth summing up, as they really seemed to encapsulate some very key bits of the future of search:

• Lyndsay Menzies, Managing Director of Big Mouth Media thinks it is important to understand how the searchers of today are different people. There is a whole generation growing up with YouTube and Flickr and social networks and they are interacting with the web in new ways, and their expectations are different than Google’s original “ten blue links.”
• Lamberti agreed, opining that this is the way search has to go, because it’s what the consumer wants.
• Jack Menzel simply said Google were not changing their business model at all: they are continuing to try presenting the best, relevant content on the web.Universal search is just reflecting the fact that there are more images, there is more video and images and maps and etc. available.
• Finally, John Battelle of Federated Media commented that we’re at a unique turning point, paralleling it to the shift from DOS to Windows. The difference being, instead of 200 developers in Redmond creating something in a vacuum, Google is engaging their users and advertisers in a conversation, and this is just one step along a continuum of changes leading to an interface and experience we don’t yet know.

All in all, quite a provoking conversation, one which offered no solutions or tips and tricks, but addressed some hard questions and I think left everyone in the room with a lot to consider.

Now, even back in the pre-internet days, the library (remember those?) had a search function. You would start at the card catalog and find the book you wanted. These days the card catalog has been replaced by a terminal screen, but the function is the same. Or, you could wander the stacks – organized by category – and manually find things that looked relevant. Or had good reviews on the back, or had a pretty cover.

Search engines are the card catalog of the web, though each of them has their own version of a Dewey decimal system. Based on a couple words or concepts, they’ll deliver a set of pages which they’ve determined match your needs. Just like some of the books you find in the card catalog won’t really be what you want, some web pages in the SERPs are going to be more useful to you than others. At the same time, search results can be the equivalent of the categorized stacks in the library, letting you browse through a variety of pages around a related topic.

What a physical card catalog doesn’t do, though, is offer suggestions of things I might like. Search engines are aiming to do this with personalized search results, as are sites like Amazon (with the “Amazon suggests” feature), Last.fm, or Stumbleupon. These sites want to be the friend who passes me a book and says, “I think you’ll like this.”

The thing is, even if someone knows me really well, they’re not always right…because frankly, personal preferences aren’t rational. Even within a genre I like, some things will work for me and some won’t. I can’t really tell you why I like James Bond books but not Mike Hammer books. And if my Last.fm playlist leans towards melodic singer/songwriters like Cat Power or Neko Case, their algorithm is never going to even consider offering me any Norwegian death metal or west coast hip-hop. But I like both of those things, too.

So to the original question: no, personalization won’t replace search. We’ll always have a need to find new things, and there will always be personal, irrational filtering that a software program will never be able to consider. And now and then I’m still going to want to just browse the stacks and pick up something with a pretty cover.

It’s interesting that SO MUCH of what counts as SEO these days seems geared either towards mom and pop sites (start a blog, make widgets and funny videos, get eight million links), or to bloggers wanting to make a buck out of AdSense (use digg, use stumbleupon, whatever, get eight million links). I’m not saying that common linkbuilding strategies WON’T work for a big corporate site, but I think the game is different. For one thing, there’s going to be a whole lot more at stake any time somebody in marketing gets an idea to do a silly video or a widget that might be a valid bit of linkbait, and it’s going to take a whole lot more buy-in to get it done.

But there’s loads of considerations in a simple linking campaign as well, and what I’m particularly considering at the moment are free directories.

Directory links are recommended, and a dime a dozen. Or less, since there’s so many free sites out there. So, given a target of x number of links in x weeks, go for the low-hanging fruit, right?

I’m pretty skeptical.

First off, some of our sites are pretty powerful already. Is there really going to be a return on getting a link in that free directory? For our smaller brands, there may be value to free directory listing simply in building visibility, but our focus, first and foremost, must absolutely be to protect the brand, and thus we need to be extra diligent in how we source those links.

So, I’ve been driving the team to do really do their research on a directory before submitting for a link.

• How relevant is the directory? Is it really just a random collection of links, or is it a site which is actually useful to a user?
• More specifically (for us), how relevant is it to the UK, and to our brand site?
• What kind of link will it provide? Direct? Nofollow? Redirect?
• What kind of traffic does the directory have? Any?
• Does anybody actually link to it?
• Does it allow links to adult/hacker/poker sites? (All potentially bad neighbors)
• Does it live on a shared server with adult/hacker/poker sites?

Really, these are all valid questions that anyone doing link development for any site should be considering. Are free directories worth it?

The relevance question is a huge one. We can be way better about finding opportunities than a listing of free directories (and in fact we are…). Anybody can find a list of free directories and claim 20 new links in a month. What else can we do?

To that end, I’ve also been doing a lot of reading to get my head back in the flow to start really pushing them to get creative, and here’s a few linkbuilding articles that I’ve found particularly worth the read in the last week or so.

• With excellent timing, Rae Hoffman posted her interview with 11 SEO experts on link development this week. An excellent companion to last year’s group interview on building backlinks.
• SearchEngineWatch.com’s articles Link Value: Top Rankings, Secrets and Lies – Part 1 and Part 2
• Also at SEW, 7 Tips for Training Link Developers
• Text Link Quality and Valuation Guide at Stuntdubl.

There are of course loads more out fantastic articles out there, and I’ll probably add to this list as I keep going.

What else can we do? Well…there’s loads, but that’s another post.

Week 2 of 2008:

Wikia launches:
The much-touted launch of Jimmy Wales’ stab at user-generated search results. Is it really an SEO free-for-all? Only time will tell.

Graywolf on Wordpress SEO
Michael takes a great look at how to maximize the keyword benefit of your post titles, post slugs, and page names.

Self Made Chick
Self Made Chick talks about blogging in first person. A particularly useful post as I ‘find my voice’ here, but the rest of her blog is great as well. It probably resonated more with me this week than it might normally as I’d just been talking with a friend about ways she could get a little extra income for herself, and SMC has some great first-person experience doing just that.

AdSense changes the rules
Google announced changes to the AdSense referral program, and there’s lots of interesting commentary going ’round about it. Problogger calls it flat out stupid, while Andy Beard rationalizes and then looks at exploiting it, and Bruce Clay offers a very reasonable opinion: it is bad business to alienate your best customers. I couldn’t agree more.

I found a basic, but functional, plugin called Twitpress, which does exactly what I wanted. Except…I’m also using the All In One SEO Pack, which rewrites page URLs into an SEO-friendly format. (Really, a must-have plugin.) Twitpress by default will tweet the stock version of a post URL:

http://RelevantText.com?p=24

instead of the format I want to show:

http://RelevantText.com/making-the-most-of-server-errors-20080111/

Now, I know that a)Twitter links are nofollowed, so this doesn’t really matter for the spiders, and b)Twitter also automatically turns long links into tinyurls, but it still bothered me (more on why in a minute). So, I set about to fix the plugin.

After reading through what the plugin code was doing, I surfed through the WP database tables a little bit, and discovered that I needed to change one line in Twitpress. Hooray!

In the twitpress.php code, replace line 85:

$proto = str_replace("[link]", get_option('home')."?p=".$postID, $proto);

with

$proto = str_replace("[link]", $post->guid, $proto);

‘guid’ is a field in the wp_posts table, if you care.

Bingo. I’m very pleased with myself.

So why, you may ask, do I care about how the links look in Twitter if they aren’t spiderable? Because I’ve also installed the Twitter App on Facebook, so any time I update Twitter, my Facebook status updates as well…which means the link is then being pushed out along the newsfeeds of all my contacts there. The link is still not spiderable, but it is potentially much more likely to get seen, followed, and possibly linked to. Through the tinyurl redirect, it now goes to the right version of the URL, and when people subsequently link to the post, I want them using the right one. This, I think, will help that along.

Jan 14 Update: After my initial excitement, I’ve discovered that this is still slightly buggy – notifications occasionally appear on twitter with the p= URL, and sometimes with no URL at all.  This seems to only happen when a post is first published, and not when later edited, but I’m not clear why, as the ‘guid’ field is populated with the first publish of a post. So, this is cool when it works, but I’m still looking at it. 

I was reminded today that people often don’t deal with 500 server errors at all, but on a large dynamic site these errors are just as bound to happen as 404s, and they’re far more troublesome. They are unpredictable, untrackable (unless you want to trawl through server logs, which I for one don’t), and harbingers of doom for your site because more often than not, they are indicators of something very bad going on behind the scenes…and you can bet your AdSense check that if a user sees them, a search spider does, too. When a spider hits a server error, it’s usually dead in the water, and that spells disaster for your rankings.

The good news is, it’s actually not too hard to deal with them properly.

One of my large corporate sites was having some massive issues with server response time last year, and as a result we were seeing a significant uptick in the number of 500 errors being reported in Google WMT’s crawl stats.

For the most part, it seemed like simply backing up and reloading the page usually got past the error, but GoogleBot isn’t going to do that. We really had no way of knowing just how pervasive the problem was, but we knew it was affecting the user experience, and clearly killing GoogleBot on a regular basis. While the technology group worked on the backend issues, we stemmed the problem from the front end by creating a custom error page to display any time a 500 error occurred.

The criteria were minimal:

1. Improve the user experience when an error occurs
2. Provide search spiders a way to continue through the site, and
3. Be able to solidly track the number of server errors being delivered as part of our overall statistics

Fortunately, both .NET and Apache make it very easy to define a custom page to display when a server error happens.

In Apache, it’s dead simple – add a line to your .htaccess file like this:

ErrorDocument 500 /friendly500.html

(the nice thing here is that you don’t need to tweak the server config file, which you probably can’t do if you don’t manage your own servers…)

Microsoft servers are a little more involved. For a friendly error page in .NET, IIS tells yout to edit the web config file to include this code:

<customErrors mode="On" defaultRedirect="errors/friendly500.html">
</customErrors>

As noted here on Techrepublic, you may define different pages for different errors:

<customErrors mode="RemoteOnly" defaultRedirect="errors/ErrorPage.aspx">
<error statusCode="400" redirect="errors/friendly400.html" />
<error statusCode="401" redirect="errors/friendly401.html" />
<error statusCode="403" redirect="errors/friendly403.html" />
<error statusCode="404" redirect="errors/friendly404.html" />
<error statusCode="408" redirect="errors/friendly408.html" />
<error statusCode="500" redirect="errors/friendly500.html" />
<error statusCode="503" redirect="errors/friendly503.html" />
</customErrors>

The page can have either an .aspx or .html extension, but keep in mind that if the server is having problems there’s no sense in trying to deliver another dynamic page. Keep it static.

One caveat : IE will try to display a friendly error message of its own, unless the error page is over 512k, so put some text on it.

As our existing 404 page is essentially a sitemap, we quickly realized that we could simply duplicate it as ‘error.html’ and with a few text changes, use that. Users now get a friendly “Oops!” message, and spiders and users alike have a variety of useful links enabling them to continue navigating the site instead of going elsewhere.

Results?

A snapshot report from Google WMT in July showed 218 server errors that happened during their crawls in the previous two weeks. Today, there are none listed at all. (To be fair, the tech guys have been doing loads of work to make things run better as well, and credit where credit is due.) But we can also now see in our statistics that regardless of what GoogleBot is seeing, the error page has actually loaded…um…let’s just say “rather a lot” this month so far, and we can now start assembling solid numbers of how much the server issues are affecting the user experience and to argue for even more improvement work on the backend.

The argument is that a brand-new page won’t have a bunch of backlinks pointing to it, so there’s no reason it should appear near the top of the SERPs directly after being indexed…unless Google is giving greater importance – at least temporarily – to newer content.

Maybe it’s only true for whatever they currently identify as ‘hot topics,’ but it’ll be worth watching. I may try a little experiment to document later today or next week; I’ve certainly seen Google give a nice big boost to a newly indexed page before it fell off into a more stable position. If, by interpreting ‘more recent’ content as ‘more relevant’ content, Google has slipped up here, it’s only a matter of time before the blogspammers start capitalizing on it by simply posting more crap more frequently to maintain consistent high rankings.

1. No more supplemental index
   Supposedly, this means more relevant results for all searches, all the time. (Interesting, since I thought that was the goal anyway…) But probably a key thing here is more relevance for foreign language queries as well, which may ‘translate’ into Google getting a bigger slice of the bits of foreign search they don’t already have. I think it will also mean less confusion about just how deeply/thoroughly a site is indexed.
2. Sphinn
   Sphinn is no Digg. Only SEO’s are going to see any traffic boost from Sphinn; it’s not something that Bob’s Widgets is going to try to game for traffic and links. But it has quickly become invaluable as a means of connecting the vast network of search marketers out there and bringing attention to important and interesting news or opinion…without having to monitor eight hundred blogs every day.
3. Universal search results
   Of course, with Google’s acquisition of YouTube happening this year as well, it followed that YouTube content would start getting a higher visibility in the SERPs, but Google and Yahoo! both started integrating video, news, and image results into the ‘main’ results page this year, and it all seemed conspicuously timed as a response to Ask’s big facelift. But it’s much more than a presentational change; it’s really completely affected how search marketing works and shifted the focus of what’s important to get noticed and rank well.
4. Facebook?
   Sure, Facebook has been around for a few years, but it was this year that anybody with an email address (i.e., not an academic one) could join, and it blew up into the place to be. And now everybody and their dog’s company thinks they need to build a Facebook app. I think it remains to be seen whether a good Facebook app has real SEO benefit, but it can a big deal for brand recognition, which of course can have a real downstream impact on what people are searching for.
5. I got a job
   Okay, this is a cheese-out, but it’s true. Landing in SEO seems to have really taken all the bits of technology and marketing and general geekery I’ve been cobbling together over the years and focused them all into a very clear path.

Todd Malicoat (aka Stuntdubl) was speaking, and mentioned a bookmarklet he used which would give a listing of the sites associated with an IP block. Right, so, Google: “stuntdubl number of sites on IP”

Number two is a blog entry titled “Tools of the Trade,” which is the seoroundtable liveblog entry from the session. With a reference to the tool I’m after: seolog’s Reverse IP domain tool. Boom.

I know that Google has gotten really good in the last few months with basically “instant indexing,” but this is the first time I’ve really seen it in action. Nice. A little scary, but impressive and powerful as well.

With images on, the page looks like this:

With images off, you get this:

“Basic image-replacement CSS,” you may think, but it’s not even that involved. The text is, indeed, just the alt text for the image.

<h1 id="logo"><img alt="The World Wide Web Consortium (W3C)" height="48" width="315" src="/Icons/w3c_main" /></h1>

OK, big deal, right?

Well, what caught my eye was the <h1> wrapping around the logo image. With images off, the alt text renders as an <h1>-level heading. In this case, it’s completely proper for the site, since it is at the very top of the page. But in my understanding, this is also what a searc h spider would see when crawling the page.

I did loads of digging around to see if anyone was talking about the potential to exploit this. Surely, if the header logo of your site could be replaced by a big fat relevant <h1> keyword or two at the top of every page, this would be something everybody knew about, right?

Apparently not.

So, we’re trying it out. One of our sites uses an image for the header tagline, and the alt text needed attention anyway. We’ve now wrapped the logo and tagline images in an <h1> tag, so with images off we have a keyword-rich, branded header appearing sitewide. I’m not convinced that this will be any kind of SEO silver bullet, but it will be interesting to see what – if anything – happens with traffic for the keyword. I don’t think it will hurt, although there is some question whether the header logo link will come off as any more spammy than normal. If nothing else, though, it’s a big step for usability.