Web developers are great: without them, we wouldn't have, well… the web! But unfortunately, a lot of developers can have a bit of a blind-spot when it comes to SEO. While some on-site SEO features almost always come as standard now (ability to edit meta tags, image alt properties, etc.), there are some important areas under the hood that often get missed. Let's have a look…Analytics (Google or Otherwise)
A decent analytics package is essential for any web marketing effort, both for measuring results and gaining insight into how your site is being found and used. Plus, as any marketeer knows, access to historical data from a site at the beginning of a campaign is almost priceless. And yet, it's amazing how many sites, big and small, are launched without analytics. Get it on there from the get go and get it set up properly!
Semantic URLs
A semantic URL is essentially an address for a page that is human-readable and conveys useful information. An example for a page on "blue widgets" would be:
- www.example.com/blue-widgets/
However, all too often we see pages with addresses such as this:
- www.example.com/index.php?page_id=46
The reason we see this latter version so often is that it is much easier to implement. The trouble is that it's a disaster for SEO! Although on-page factors are not that important compared to link building, by and large, having keywords in the URL is the single most important part of on-page optimization.
Make sure your developer implements semantic URLs on your site and gives you control over each page's URL – no excuses! Although you don't get penalized for using a non-semantic URL structure, you're missing out on a big opportunity… and changing it after the site has gone live can be a big headache.
XML Sitemaps
Sitemaps don't have a great impact unless your site is on the large side, but they're easy to set up and cost nothing, so are always worth using. It's not a problem to generate these manually (there are several free tools for doing this), but if your site is dynamic or updated often, this can become a real pain pretty quickly.
It's much better for the site's CMS to update the site map automatically whenever the site changes. A lot of packages will do this natively or with the aid of a plugin, but if you're having a custom CMS written, make sure the developer includes this facility.
Controlling Indexation
If you have a large site, Google will almost never index your entire site: they have a percentage cap of the number of your pages that they will keep in the index. Now, discussing how much of the site they decide to index and upon which pages they bestow the honor of indexation is a post for another time, but suffice to say for now that although you definitely can't tell Google which pages to index, you cantell them which pages you definitely don't want indexed. You do this with this meta tag in the head of the page:
- [meta name="robots" content="noindex,follow" /]
Again, if you've got a small or static site, this isn't a problem to set up manually, but for larger sites you'll need to be able to control this through the CMS. You may even want to develop a strategy for noindexing pages automatically – if they're not getting search traffic anyway, for example. Ideally, yes, we'd like never to have to do this kind of thing, but if Google are only going to index a part of your site anyway, you'd better make sure it's the part that counts.
301 Redirects
Honestly, could this be more important? Unfortunately, a lot of developers don't think so.
If you're migrating from an old site, making sure that pages from the old version are redirected to the new is vital (assuming that the page names or URL structure has changed), but it's also important that your new CMS creates 301s automatically if you remove or change the URL of any page – something that you'll inevitably end up doing if you work actively on your site.
Again, some CMSs do this natively or through plugins, but many don't. If you're having something custom written or your developer is using something off the shelf, make sure it handles 301s for changed pages properly.
Canonicalization
Sorting out canonicalization of URLs from the start is another must – you don't want Google to see duplicate content on your site even for an instant, or it'll be reminding you of it through Webmaster Tools for evermore.
The first step is www versus non-www canonicalization. From an SEO perspective, it doesn't matter which you choose, but you have to choose one and stick to it. Implementing it involves just a simple 301 redirect rule and is easy for your developer to do.
The second step is making sure that your CMS, e-commerce package or other platform isn't generating multiple URLs for each page, and isn't adding a lot of extraneous data to URLs. This might sound like a no-brainer, but Magento, for example, makes each page available by three different URLs by default. Checking to see if you have this problem is relatively straightforward – use a tool such as Xenu's Link Sleuth(yes, it's a crazy site) to check the number of pages on your site. If it's way higher than you were expecting, you've got a problem. If you've not fixed this problem before the site goes live, that means a lot of 301 redirects to set up as you rationalise the URL structure.
Lastly, implement the canonical tag itself. At it's most basic level, this tag tells Google what the definitive URL of a page should be. If your CMS definitely isn't generating multiple URLs per page, it is still worth implementing, as it will help prevent potential problems caused by incoming links with extra URL data in them (e.g., tracking tags from mailing list software, etc.). The canonical tag is dead easy to implement and should appear on every page. It looks like this:
- [link rel="canonical" href="http://www.example.com/page.html" /]
Google is expanding the remit of this humble tag over time as well, so it's worth keeping an eye on what you can and should be doing with it – it all helps with Google's indexation of your site.
Site Speed
Hardcore coders are often obsessed with speed. This is a good thing, as site speed is now a part of Google's ranking algorithm (although perhaps not a large part… yet). The problem is that the main issues to do with a site's performance are not to do with the code itself (at least not for most smaller sites), but rather to do with things such as HTTP request optimization, combining and compressing external files, loading JavaScript asynchronously, using cookie-less domains, etc.
If all that sounds pretty technical… well, it is. Luckily, you don't need to understand how to do it – you just need to ask your developer to look after it for you. If they're not already on the ball with site speed, a number of free tools will audit a site's performance and make recommendations for improvement, such as the Google's Page Speed suite. You may also want to ask your developer about using a content delivery network such as CloudFlare. Going to town on your site's speed really can make a surprisingly big difference!
Summing Up
Making sure that your developer gets these basics sorted right from the beginning will get your on-site SEO running like clockwork, leaving you free to concentrate on building links and great content. None of them are optional!
No comments:
Post a Comment