How to do site audits the old fashioned way

Posted by Andrew Tobert
Learn SEO
How to carry out a site audit the old fashioned way at Wordtracker, the leading keyword research tool

The key to success in SEO, as in life, is to watch what you’re doing. If it’s working, keep doing it and if it’s not, stop. It’s that simple, but with SEO, getting the right data to be able to make that decision is less easy. That’s why we’re here.

We'll show you step-by-step how to get relevant data about your site's performance, and what you can do with that data once you've got it.

A site audit is simply the process of looking at your site and assessing its performance - a term that can be defined in many ways - but let’s keep things simple. We’re going to focus on the SEO side of things by getting our hands dirty and opening up the hood, so you can keep your site at the top of its game.

Your home page

We'll start at the home page.

Your home page will, almost certainly, get most of your links, traffic, everything that’s good. Here are a few things to look out for.

Is the content good?

A simple enough question. Google loves content, and what’s on your home page should be top notch and useful. But quality is subjective surely? Well, yes and no.

You should be able to answer ‘yes’ to the following questions. Can you?

  • Is there (at least) a paragraph of well-written text explaining what your site’s about?
  • Is your home page free of pop-ups and pay walls?
  • Is it Flash-free? (Flash is a coding language used mostly for animation. Google can't crawl it, and it’s not supported on iPhones and iPads. Basically the less Flash there is on the home page, and the site generally, the better).
  • Is there more content than advertising? (Google accepts some advertising, but too much on one site (especially 'above the fold') and it looks spammy.)
  • If I, a total stranger, went on your home page, would I be able to tell you what your site was about in four seconds?

What does Google think your site does?

Google your brand term plus one of your products. Eg, “Macy’s jewelry”. What comes up? Is it the page you’d expect? If not, you’ll need to improve the content on that page and make sure there are links to it in relevant places.

If the only thing you can see is your URL, eg, without a meta description (ie, without any text underneath), it’s likely that you’ve blocked Google from accessing your site. Read on to double-check if you have and find out what to do about it.

Now Google your brand. What do you see? Is it there at all? (It should be top of Google's results). If it’s quite a new site, you might just need to start link building so that your brand gets the ranking you want. This will be the ‘quickest win’ of all your SEO activities.

If your site is well established, there should be site links too. (I’ve highlighted ours in the red box below).

Wordtracker site links

Where do yours go? If they’re not going to important pages, it’s because Google disagrees with you about what the key pages on your website are. Help convince it by making the key pages the best they can be, with links from the home page. And you can gently nudge Google in Webmaster tools. It’s only a nudge mind. They can, and do, ignore you.

Wordtracker site links

If you see the URL of your site, but no site links, it’s possible that you’ve been penalized by Google. If you think this might have happened, review your link building activity to make sure it’s all above board. We have lots of articles that can help you do this.

Is there one URL?

There are several possible URLs (web addresses) for your home page. You could have,,, or just Each one of these variations should redirect back to a single URL. So regardless of which variation your users type into the browser, they should be 301 redirected back to (eg,) We’ll learn more about 301 redirects in a moment.

If, for the record, you don’t do this, Google will potentially think you have duplicate content (this is bad) and some users will link to one page and some will link to others, seriously handicapping the power of your brand. So check this now.

Are the URLs simple?

Your URLs should be human-centric and logical. So instead of ...

use ...

The search engines prefer it, and so do humans. (They remember it for one thing).

Where are the golden pages?

All websites have pages with really great content, and pages that are nothing special. A bit meh. Your navigation should be clear and intuitive, but make sure it makes your best content easy to find. This means putting in links from the home page (where almost all your customers go), and from other key pages of your site.

If you’re not doing this very well at the moment, there are a few shortcuts you could try. Like having a ‘bestsellers’ section on the home page, that links to your most popular products. Or if you’re a content portal, you could have ‘most read articles’. Incorporating this sort of thing into your website will give your key landing pages a much needed boost.

So let’s talk a little bit more about these pages. Specifically ...

Your site structure

You should make sure your site is structured around your niches, so all similar content is in a similar place. You can learn about site structure from my esteemed colleague, Mal. There are really two aspects to this:

  • Make sure each of your key content pages correspond to one of your keyword niches. (Eg, all your content on the Indonesian frogfish is on the ‘Indonesian frogfish’ page. And that’s the only thing on that page).
  • Make the navigation as simple and as logical as possible. Most sites follow the structure of home page>category page >product page. It’s logical, and clear for users, so try and relate this back to your site as much as possible.

The reason why we recommend building a site this way is simply that it’s easier for you to rank for the keywords that you want. Google can see all your content, and because it’s well structured, Google can see which pages are relevant for which keywords. So they'll give you a better rank (and more traffic) for those keywords (with any luck).

Landing pages

To check whether not your site is well structured, look at your landing pages in Google Analytics. (See the image):

Client Attractors book cover

So look at your pages. A well structured site will get distinct traffic to distinct pages. So someone who searches for Keyword A will arrive at a different page than someone who searches for Keyword B. Look at your site and make sure the content on each page is sufficiently distinct and that there are relevant keywords on the page. The better, and more distinct each page is, the better your site will perform.

Internal linking

Also, think about your internal linking.

One of the (many) reasons why Wikipedia always appears so much in the search engines is because they always inter-link all their internal pages. All their linkjuice (which in Wikipedia’s case, is a lot) can flow around the site. All pages can benefit, not just the ones that attract the most external links. If there are pages in your site with only one link from the rest of the site, they’re unlikely to rank very well.


If we talk about words we must also talk about images. After all, they’re worth a thousand.

Google LOVES images, and so do users. They break up text and, to a human, give an instant impression about what the site is about. But the problem is people don’t tag them correctly. Google isn’t a human, so if you don’t code up your images correctly, the images are as good as useless.

So how do you know whether or not your images are tagged correctly? Click on your images. Let’s pick this image of Apple CEO Timothy Cook on the Huffington Post.

Huffington Post

Right click it, and select ‘Inspect element’. Depending on which browser you use, a screen will pop up, either at the bottom of the screen or in a new window, with that area highlighted. It’ll look a bit like this:

Alt tag

Scan along to where it says alt="Tim Cook". Now do the same to your images, what does it say? Does your alt tag even exist? You want your alt tags to be as relevant and descriptive as possible.

So, if it’s a picture of Tim Cook, call it “Tim Cook”. If it’s a photo of the Ritz hotel in Paris, call it “Ritz hotel, Paris”. And, where appropriate, include your page's target keyword. It’s easy but you’d be amazed how often it’s overlooked. Get in the habit of using the alt tag every time you upload an image and you’ll be as right as rain. If you’ve not been doing it, well, it’s a big job to find and change all the alt tags on all your images. Sorry.

We’ve started getting stuck into the code now, so let’s keep going.

Titles and descriptions

One of the worst things you can do, which as fate would have it, is also supremely common, is to not pay attention to your title tags (explained here). These should be unique, and include a keyword or two. They should encourage users who see them in the search results to click through, so make sure they’re written with humans in mind too.

They are, along with your meta-descriptions, your one shot at persuading your customers to click though. So make them punchy and informative. (Descriptions by the way, don’t affect your ranking and can be a little longer, up to 156 characters.). Again, there's help on your titles and descriptions available in our Academy.

And then there’s the ‘header’ tags.

Header tag

Header tags appear in the code as <h1>, <h2> etc (it goes all the way to <h6>). You can technically have as many <h2> <h3> etc as you like, but make sure your content is well-structured, readable, brief and on-point. Oh, and only use one <h1>.

Header tags, remember, are a great way of splitting up your content, but they're also good at telling Google what each section is about, in order of priority. So <h1> is the key term, <h2> might be a secondary keyword and so on.

In reality, a site selling cell phones, might have cell phones as their <h1>, then cell phones deals as the <h2>, assuming they had that sort of content on the page. What you’re looking for is that each header tag is relevant to the content on the page and is descriptive. If you don’t, you’re seriously missing a trick.

The language(s) of (website) love

On the subject of code, make sure your web pages are predominantly HTML. Your CSS can go on a separate style sheet and where possible avoid Flash. (Google hates it because they can’t crawl it. And Apple won’t display it on their phones or iPads. But everyone (with newer browsers, anyway) loves HTML5. Do that instead). If what I’ve just written makes absolutely no sense, relax.

These topics require quite a lot of explanation, so we’ll be covering them in more detail later on. If what you've just read made perfect sense and you’re wishing I’d said more, check this out But as a very easy check to see the state of your code, download the Web Developer toolbar for Firefox. This allows you to turn off CSS, Javascript and Cookies. If your site becomes almost unrecognizable (or worse, unusable) you’ll probably need to re-configure your code.


So we’ve looked at on-page elements but of course, that’s only one part of it. Another part is surely how well your site is performing.

Rankings were once the corner stone of any SEO strategy. After all, how better to measure how well-optimized your site was than looking at how well you’re ranking?

Time has moved on and, whilst ranking is important, bear in mind how often it changes for individuals. There’s personalization, localization, social amplification. There are many variances, so take any ranking data that you see with a pinch of salt.

When checking your rankings (and don’t worry, you can do this easily with Wordtracker's tools: there’s a video coming shortly) again try not to focus on individual keywords. Instead, think about your keyword niches. What changes could you make to your pages to ensure you can rank for long tail keywords?

Ok, so we’ve done content, code and ranking, so I guess the only thing left is to talk about crawling. And maybe plug our products a little.


Crawling is how the search engines deal with your site. They go through the pages and click on the links. There are things you can do to help them, but there are also things that can go wrong that you should look out for. We’ll go through those now.

Robots.txt file and robot meta files are tools which let you prevent the search engines looking at certain content.

Please use robots.txt with extreme caution. We get so many service calls where it transpired that the webmaster has inadvertently blocked the search engines from their entire site. So don’t do this.

To check that you haven't blocked either part or all of your site, simply type your home page (eg, and add /robots.txt to the end. If you don’t see anything, it means that you don’t have a robots.txt so Google has access to everything. But if a section is blocked, you’ll see disallow: /site-section, where any page of the form will be blocked from Google.

As you can see here ...

Assetz robot text

... this site, Assetz, has decided to block Google from their ajax and data-entry pages (amongst others). What does yours say? Here’s more information on robots.txt from Google.

Note that you can block different search engine bots from different sections with the user agent: line. Use * if you want to block (or allow) everything.

Site maps

The Romeo to robots.txt’s Juliet. Rather than blocking (or allowing) whole sections of the site, a sitemap guides the search engines through your site, allowing them to crawl all your best content quickly.

Sitemaps can be HTML or XML. If your site is smaller, and/or you’re not a techie, a HTML site is probably easier, as it’s just a list of HTML links on a given web page. Here’s Google's head of web spam, Matt Cutts if you don’t believe me

But what if things go wrong? To identify problems with your site, like when a search engine (or user) clicks on a broken link, you need to check your server responses.

Checking your server responses

To check your server responses, you could use Wordtracker's tools but this isn’t a sales pitch, yet. So log into Webmaster Tools

In Webmaster Tools, once you're at the dashboard for your particular website, start on the left-hand navigation bar, and click the ‘health’ section. Here you can see how Google crawls your site and any problems that it’s having. Google then helpfully gives you data you can export, or an easy to read graph.

Something, in fact, that looks an awful lot like this:

Error line graph

We don’t have any errors (well done us), but if you did, you’d see them underneath, in all their glory.

That solution is OK if you know your way around Webmaster Tools and you’ve got time to spare, but if not, here’s how you get that information in Wordtracker’s Keywords tool (Then I’ll stop the sales pitch). Honest.

Log in to the Keywords tool, and click ‘new site report’. We’ll then run a site report for you, in the background (so if you close the browser it doesn’t matter.)

Then as if by magic, you get all that information laid out for you. Easy.

Tool site audit

Then you’ll get all that scrummy data.

Wordtracker site audit

And look what’s this? Page titles, Descriptions, Headings, why, it’s every thing we’ve been talking about! And all I had to do was click one button!

The tool is pretty cool. As you can see doing a site audit manually is a lot of heartache, so we’ve really worked at creating something that helps you. So why not explore the site yourself? Or hang around, as we’ll be posting another article on how to do a site audit the Wordtracker way in the coming days. But in the mean time, happy auditing!

Get a free 7-day trial

A subscription to Wordtracker's premium Keywords tool will help you to:

  • Generate thousands of relevant keywords to improve your organic and PPC search campaigns.
  • Optimize your website content by using the most popular keywords for your product and services.
  • Research online markets, find niche opportunities and exploit them before your competitors.

Take a free 7-day trial of Wordtracker’s Keywords tool 

Join the discussion

You can keep in contact with us on Google+, find us on Twitter. Facebook and LinkedIn