Search engines love text. Unfortunately they’re not so hot on images, so when you have to optimise a site that has thin content and rich imagery you’ve got your work cut out.

Navigation

1. McAWilliams.com
2. Some Initial Suggestions
3. A quick note about usability
4. 1. Site Architecture
5. 2. On-Page Optimisation
6. 3. Off-Site Optimisation
7. Conclusion

McAWilliams.com

John Williams’ site was nominated in the best Photo Blog category of Irish Blog Awards and contains the regularly updated stock of John’s photos (well worth a look).

The site appears to be running on the Moveable Type platform, and you can see many issues when you check Google’s index:

1. ~ 578 pages indexed;
2. ~ 438 pages in supplemental.

Straight away that rings alarm bells for me. I checked the site with a crawler and there are a number of issues with the linking patterns observed:

• comment pages seem to be in the supplemental bin;
• search pages are being indexed;
• high number of broken links.

How to tackle these issues

There are three areas I’m going to have a look at:

1. Site architecture;
2. On-page optimisation;
3. Off-page optimisation;

A quick note about usability

But before I go any further I want to mention the most obvious issue I had with John’s site – the navigation. I just couldn’t figure out what was a link and what was plain text. As time goes by I’m becoming more and more inclined to stick with convention, even if it affects aesthetics. Blue underlined text for links cant go wrong. The links within the text area are especially confusing – it took me an age to find the comments link (more on comments later).

1. Site Architecture

The supplemental index is Google’s secondary index where it places all content that does not meet the primary index grade algorithmically. This is not necessarily a reflection on your pages – the issue is primarily caused by a lack of PageRank. Generally the solution to supplemental listings is to increase the internal Pagerank of your site.

Pagerank flows through links, both off-site links pointing at your site, and internal links pointing at internal pages. One of the easiest ways to elevate a page to the primary index is to add external links pointing at that page. Another technique is to increase the overall in-bound links pointing at your site. A third, and often neglected tactic is to reduce the number of internal pages that share overall site Pagerank.

If John published the comments for the image pages on the image pages themselves he could instantly improve the SEO of his site. Most of the comment pages are in the supplemental index. The overall size of McAWilliams.com could be reduced by a figure equal to the number of image posts instantly, and the remaining pages would receive more Pagerank as a result.

I would also block all bots from the search pages (not sure how they got crawled in the first place):

User-agent: *
Disallow: /movabletype32/mt-search.cgi


If John moves the comments into the main pages I would simply block the entire /movabletype32/ folder:

User-agent: *
Disallow: /movabletype32


(I’m not familiar with MovableType so you should check carefully that you wouldn’t be blocking something that you might want crawled.)

John could also NOFOLLOW all internal links to non-relevant pages such as search and comments (should he integrate them into the main pages). The basic goal here is to reduce the spread of Pagerank and filter it toward the most important pages on the site.

The one thing that has me flummoxed is that John’s about page is supplemental. It doesn’t make any sense, and my best guess is that this is down to on-page elements.

Possible Duplicate content issue
I noticed that the homepage displays the latest post. Within the post description there is also a link to the post, but at the archived address. While I know that Google will figure this out when a new page is published to the blog, I think that the archived link should only be published once the page has been archived and is no longer available on the homepage.

Technically this link is pointing at a duplicate page hosted on a different URL. This means that the archived page may not be indexed for a short while after a new post is published (you can see that the current post is not cached).

Permalinks
If John can change the permalink structure and combine year and month into one variable (/year/month/ to /year-month/) it may also help with indexing issues.

Canonical URL
Almost missed this. Normally the first thing I check is whether a site is accessible via both www. and non-www. URLs. If so then some link love could be split as Google basically sees these as separate sites. John should fix this with a mod_rewrite rule in his root .htaccess file.

2. On-Page Optimisation

1. Page Title & Meta data

   The page titles really need more content. I would also swap the site name with the page name:

   <title>McAWilliams.com photography | All about John</title>

   to something like:

   <title>About John McWilliams - Photographer and Photo Blogger | McAWilliams.com photography and photo blog</title>

   That way you get more keywords in there. You should wash, rinse and repeat for all pages.

   For the image pages:

   <title>McAWilliams.com photoblog : Locked</title>

   to something like:

   <title>Locked - an image of a Dublin canal lock photographed while bird watching | McAWilliams.com photograhy and photo blog</title>

   Now they aren’t necessarily the best I could come up with, but hopefully they reinforce where I’m trying to go. I’ve managed to get some relevant keywords in there – ‘image’ and ‘photographed’.

   The page descriptions also need some TLC. I wrote previously about the optimisation of META Descriptions – it’s quite relevant for John’s site. I would make sure that you give good descriptions of the images in the META Description and include your main keywords again. And make the descriptions unique for very page. This may help get some of the pages out of supplemental.

2. Navigation

   The main navigation in the header includes links that should really be in the footer. I would relegate the Links and Validation links, and consider moving the Sitemap also.

   I saw some comments from others who also found the navigation difficult. I would act on these comments as users are more important than search engines in my view.

3. HTML Mark-up

   Here’s an extract of the HTML for the main image:

   <a href="http://www.mcawilliams.com/[... ]duck.html"><img alt="copyright John Williams 2004-2007" title="this image : Locked © 2004-2007 by John Williams / click to view the previous image" src="http://www.mcawilliams.com/locked.jpg" class="imageborder" border="0" /></a>


   John should really be using the image ALT and TITLE attributes better:

   <a href="http://www.mcawilliams.com/[... ]duck.html"><img alt="Locked - an image of a Dublin canal Lock | copyright John Williams 2004-2007" title="Locked - an image of a Dublin canal Lock | copyright John Williams 2004-2007 | click to view the previous image" src="http://www.mcawilliams.com/locked.jpg" class="imageborder" border="0" /></a>


   Semantic mark-up will really help these pages:

   <div class="blogdata">
11.03.07: <a href="http://www.mcawilliams.com[... ]html"><span style="text-transform: uppercase;">Locked</span></a><br/>


   John should really try to use Header elements here, and as much as possible include some keywords:

   <div class="blogdata">
11.03.07: <h1 style="text-transform: uppercase;"><a href="http://www.mcawilliams.com/[... ]html">Locked</a></h1>


   Better still, John could take that styling attribute and stick it into an external style sheet. If it didn’t upset the apple cart too much I would try to repeat the title in that h1 element:

   <div class="blogdata">
11.03.07: <h1><a href="http://www.mcawilliams.com/[... ]html">Locked <span class="color: #666">a Dublin canal Lock</span></a></h1>


   Basically it can pay off to use header tags <h> 1-6. Google will give you some benefit if you do.

4. Technorati links

   Because John has so many pages in he supplemental index I might try sticking NOFOLLOWs on those Technorati links. After all, it’s not like they need the link love. This should retain a little more of the Pagerank in-site rather than leaking it out.

5. Image Search

   I already mentioned about image ALT and TITLE attributes. Another important element to use is the image file name. If your image does rank well on an image search the file name will normally be displayed below the image. So it’s a good idea to match the file name with the image:

   dublin-canal-lock.jpg

   If you search for ‘image search optimisation’ you’ll find a whole bunch of resources for this topic.

3. Off-site Optimisation

Without a doubt link building will help get many pages out of the supplemental index. My gut feeling is that the site architecture isn’t helping out though, and I was surprised to see that the site didn’t rank well for related terms and phrases. I expected to find it for phrases like ‘photo blog ireland’ (it’s number #1 for ‘photoblog ireland’). Diversifying the keyword mix should help to increase search volume.

If John can get some deeplinks pointing at some of his older images it would help the site considerably. Building quality on-theme links can be one of the most difficult areas of SEO. I know personally that getting noticed can lead to countless links. Try to comment on forums and blogs related to your site – if you ad to conversations people will take notice, and this can often lead to links for your site.

I’m pretty sure that the low-hanging fruit such as directories will also help McAWilliams.com to improve its rankings. Even forum signatures may help promote some of the supplemental pages.

Unfortunately I can’t go into much detail on link building as my time is up on this particular mini project. I will say one thing – lean on bloggers

Conclusion

John’s site is an interesting test case. Two blogs in one, and the primary blog containing thin-content. My gut tells me that the site architecture and on-page issues are really holding this site back. Yahoo! shows over 3,000 back links so I’m cautiously optimistic that John has enough link juice to rank well.

I wish John the best of luck with the site, and hope that some of the advice above is useful.