Welcome to the Q&A Forum

We run a fairly large online community with a popular blog and Europe's largest online shop for drift-specific motor sport parts and our website has been around since 2004 I believe. Since it was launched, the blog (or previous CMS system) has been at the domain root, the forums have been located at /forum and the shop at /shop (or similar) but we have decided to move things around a bit and would like some comments as to whether we are doing the right thing or if you would make any addition or different changes to us. Currently the entire website gets around 3m page views per month from 500,000 visitors, but this is split roughly 75% to the forums, 10% to the shop and 15% to the blog (but remember the blog is at the root so anyone who visits our homepage "visits" the blog). We plan to move the shop to the domain root (since the shop provides the income for the business - surely it should be the 1st thing visitors see?), the blog from root to /blog and the forums will stay where they are at /forum. We have read Steven Macdonald's post here, and have taken notes to help minimize traffic loss and disruption to our army of users and hopefully avoid too many penalties from Google and plan to: 301 redirect old URLs to new ones where they have changed. Submit new site maps to search engines. Update old links where we have control (such as forums where we are paid traders etc.). Send out a newsletter to our subscribers. Update our forum members. Fix errors via WMT before and after the re-structure. Should we be taking this opportunity to actually set each of the three sections of the site to it's own sub domain? Our thoughts are that if we are disrupting things, it's surely best to have lots of disruption once rather than a little bit of disruption several times over a 3-6 month period? OSE shows us to have roughly 1500 inbound links to /shop, 2100 to /forum and 4800 to the root / - if we proceed with our plan and put 301 redirects in place this seems to be the best plan to retain the value of these links but if we were to switch to sub domains would the 301s lose most of the link values due to them being on "different" domains? Any help, advise or suggestions are very welcome but comments from experience are what we are seeking ideally! Thanks Jay

I sent the email below to help@seomoz.org on the 12th April and haven't heard anything back so thought I'd fire it up here in case someone has a solution for me... "For the past month or so I've had an issue in Firefox where my SEOmoz toolbar appears to not sign me in, or allow me to sign in. All metrics are 0 for all websites/pages and if I click the login button, nothing happens at all. If I click the Quick Links button the drop down menu appears but whichever option from the drop down I choose, nothing happens. I'm funning Firefox 11.0 and have attached a screen grab of the bar as it appears no matter where I browse.
Do you have any suggestions how I can get it working again? It works fine in Chrome but I primarily use Firefox..." UNRMw.jpg

Please bare with me, this might turn into a bit of a waffle, but I'll get to my question... I promise! I've just been looking at our CPC traffic for April and 2 search terms jumped out at me. I recognised them from previous keyword research because they are search terms that I expected to be high traffic (from past experience), but Google Adwords keyword tools showed them to have no potential traffic, and next to no potential traffic (literally 0 local searches and 12 local searches per month). Last month search term A had 46 visits, with 19:25 average time on site and 8.70% bounce rate and search term B had 10 visits with 14:47 average time on site and 0% bounce rate. For very boring reasons we are not currently able to measure conversions on these terms since (they are related to consumer finance and when a customer applies for finance it is all done on our finance providers website) but despite the low volume, these are pretty good figures for on site behaviour and so it got me thinking... Is there a more accurate tool to estimate traffic volume that we should be using rather than the Adwords tools? I appreciate that the estimates are probably made based on historic search behaviour and April's traffic could just be a one off, but these particular terms used to be insanely popular 4-5 years ago when I worked at a competing company.

Our shop site has a 3 tier drop down mega-menu so it's easy to find your way to anything from anywhere. It contains about 150 links and probably 300 words of text. We also have a more context-driven single layer of sub-category navigation as well as breadcrumbs on our category pages. You can get to every product and category page without using the drop down mega-menu. Although the mega-menu is a helpful tool for customers, it means that every single page in our shop has an extra 150 links on it that go to stuff that isn't necessarily related or relevant to the page content. This means that when viewed from the context of a crawler, rather than a nice tree like crawling structure, we've got more of an unstructured mesh where everything is linked to everything else. I'd like to hide the mega-menu links from being picked up by a crawler, but what's the best way to do this? I can add a nofollow to all mega-menu links, but are the links still registered as page content even if they're not followed? It's a lot of text if nothing else. Another possibility we're considering is to set the mega-menu to only populate with links when it's main button is hovered over. So it's not part of the initial page load content at all. Or we could use a crude yet effective system we have used for some other menus we have of base encoding the content inline so it's not readable by a spider. What would you do and why? Thanks, James

Hi, Apologies in advance for the long winded question... we need some guidance with our category/product/options structure in our shop. We primarily sell car parts and lots of our parts have multiple fitments for what is basically the same part. Some ranges can have 1,000s of products. We can't work out what is an appropriate level of information and granularity for our product structure.We recognise the importance of having fitments and specific terms in the product title and URL, but we also know that having loads of almost identical product pages is a definite negative and fragments our SEO potential. But where's the happy medium? For example, let's say we have a specific brand of brake pad (we'll call it Brako) with 4 different product-models (Super1, Super2, Super3, Super4), each fits 100 different cars, which are made by 10 different manufacturers. We have a few different ways of presenting/splitting up these 400 simple products: (ignore the URLs here, this is just to illustrate the browsing structure & likely product page titles) 1 category for the Brake Brand with 400 product pages inside, 1 product page for each specific combination of brake product-model and car-fitment. /Brako/Brako-Super1-brakes_BMW-M3.html 1 category, 400 product pages, 0 choices on each product page. 1 category for the Brake Brand with 40 products inside, 1 product for each specific combination of brake product-model and car-manufacturer. Each product page would then let you choose from a dropdown which of the 10 specific cars you had. /Brako/Brako-Super1-brakes_BMW.html 1 category, 40 product pages, 10 choices on each product page. 1 category for the Brake Brand with 4 sub-categories inside for the brake product-models with 100 products inside each, 1 product for each specific combination of car-fitment. /Brako/Brako-Super1-brakes/Brako-Super1-brakes_BMW-M3.html 1 category, 4 sub-categories, 40 product pages, 10 choices on the product page. 1 category for the Brake Brand with 4 sub-categories inside for the brake product-models, with 10 products inside each.1 product for each specific combination of brake product-model and car-manufacturer. Each product page would then let you choose from a dropdown which of the 10 specific cars you had. /Brako/Brako-Super1-brakes/brakebrand-Super1-brakes_BMW.html 1 category, 4 sub-categories, 40 product pages, 10 choices on each product page. 1 category for the Brake Brand with 4 products inside, 1 product for each brake product-model. Each product page would then let you choose from 2 dropdowns, each with 10 options: one for car manufacturer, the next for car model. /Brako/Brako-Super1-brakes.html 1 category, 4 product pages, 100 (10x10) choices on each product page. 1 product page containing options to choose all 400 Brako products using 3 drop down boxes: Car Manufacturer, Car Model and Product-Model /Brako/Brako-brakes.html 1 category, 1 product page, 100 (10x10) choices on each product page. Or we could mix it up and split the sub-categories by manufacturer: 1 category for the Brake Brand with 10 sub-categories (1 sub-category for each of the car manufacturers with 40 products inside each), 1 product page for each specific variation of car-fitment and product-model. /Brako/Brako-brakes-BMW/Brako-Super1-brakes_BMW-M3.html 1 category, 10 sub-categories, 40 product pages, 0 choices on the product page. 1 category for the Brake Brand with 10 sub-categories (1 sub-category for each of the car manufacturers with 10 products inside each), 1 product page for each specific variation of car-fitment. Drop dowjn box on the product page lets you choose product-model (Super1-4) /Brako/Brako-brakes-BMW/Brako-brakes_BMW-M3.html 1 category, 10 sub-categories, 10 product pages, 4 choices on the product page. 1 category for the Brake Brand with 10 sub-categories (1 sub-category for each of the car manufacturers with  products inside each), 1 product page for each specific variation of product-model. /Brako/Brako-brakes-BMW/Brako-Super1-brakes_BMW.html 1 category, 10 sub-categories, 4 product pages, 10 choices on the product page. Obviously, option 1) is going to be the best search match for someone searching for 'BMW M3 Brako Super1 brakes' but that page will have almost identical content to 100 other pages and very similar content to a further 300 pages, which takes it's quality ranking down a lot. At the other end of the scale of complexity is option 5) which concentrates all search potential for the Brako Super1 down to a single page, which can be well written and have great content, but wouldn't have a match in the title, url or product name for anyone searching for 'BMW M3 Brako Super1 brakes'. 'BMW M3' would be mentioned in the page, but only once in a drop-down along with 100 other cars and possibly once in the content if there's something noteworthy about that application. So which option would you go for and why?

Here's our sitemap: http://www.driftworks.com/shop/sitemap/dw_sitemap.xml In google webmaster tools, I can see the sitemap report and it says: Items:Web    Submitted:2,798    Indexed:2,910 Items:Images    Submitted:3,178    Indexed:0 Do you have any idea why our images are not being indexed according to webmaster tools? I checked a few of the image URLs and they worked nicely. Thanks in advance, J

In our Magento store, products are arranged into categories, subcategories and so on. We typically have 3 or 4 layers of category depth. This makes it nice and easy for customers to find stuff, but it means that the end categories have massive long urls. I'd like to keep our category tree structure in place from a navigation point of view, but I feel the url structure is pushing some important stuff to the back of the shop as it were. We have something like 200 categories in total. So, assuming every individual category has an a unique name, I'd like to rewrite the urls so that: ourshop.com/car-parts/
stays as  
ourshop.com/car-parts/ ourshop.com/car-parts/suspension/
becomes  
ourshop.com/suspension/ ourshop.com/car-parts/suspension/springs
becomes  
ourshop.com/springs/ ourshop.com/car-parts/suspension/springs/thismake-lowering-springs
becomes  
ourshop.com/thismake-lowering-springs/ and so on....
I'll need some custom magento URL rewrite  work done, but that's another story. The real question is whether you guys feel this is worthwhile? Are there any other stores with a deep categorised navigation structure, but a flat url structure? thanks, James