So you have decided on a design makeover for the website. Or perhaps its the boss who thinks it needs refreshing and a bit of a lift.

Carrying out an emetrics audit before a redesign, or any major change to a site, is essential if you are going to assess the benefits of your investment. It will also enable you to make sure that everything you change is for the better and that the things that are really working well dont get swept away.

So what do you look for?

Current overall visitor levels this might seem obvious but if you dont know how many visitors are coming to your site before the change how can you assess its impact?

Returning visitors how many of your current visitors have visited the site before. Your returning visitors usually represent your best prospects, after all they are interested enough to bookmark your site and come back. Identifying returning visitors requires the presence of cookies. If you are using a hosted emetrics service (one using JavaScript) to monitor traffic this will be done automatically. If you are analysing log files then your site must set a persistent cookie in order to recognise returning visitors.

Conversion rates how many purchases, enquiry forms, subscriptions or whatever your site purpose is, are you getting both in absolute terms and as a percentage of total traffic?

Search engine referrals and rankings the more sites that are sending traffic to you the better. This will depend upon your rankings and overall visibility, which in turn requires that the spiders are crawling your site assuming you are doing your seo job properly. It is essential you prepare a rankings report on your major keywords before you undertake any major site changes. It is equally important to know how well your site is being spidered before the changes are implemented. This will show either that the current architecture is spider friendly or that the design changes need to include better spiderablity if that is a problem.

Site stickiness how many of your visitors leave the site immediately, usually defined as being less than 5 seconds, without navigating beyond the entry page. The figure of short visits needs to be kept as low as possible, and how low is acceptable will depend on how well you can, and want, to define your target market. If you have a well established market, such as digital cameras, then anyone not looking for digital cameras is of no use to you. You may feel differently however if, on the other hand, you are working in a new market, photobooks for example. Here not everyone who is a potential customer is familiar with the product and therefore looking for it, so you may decide to try and capture anyone looking for anything on photography. This is going to reduce your site stickiness but may be a worthwhile price to pay in order to increase awareness of a new product or service.

Lots of factors affect site stickiness, for example the graphic design of the site and the way in which the navigation is structured in particular will affect how effective a site is in retaining its visitors. Making sure you have a bench mark before any site changes are made is vital if you are to assess whether any change is a success and to what extent.

Entry pages which pages are visitors entering your site by? Web designers often feel most comfortable with everyone entering through the front door and following the path set for them. However, if this is what is happening then your site is missing a huge opportunity. Every content rich page on your site is potential entry point, and has the potential to be optimised for specific keywords. Further, conversion rates are likely to be higher if a visitor enters the site on a page that offers him the particular information he is looking for.

Navigation paths the idea that there is one unique route from home page to order page is a fallacy. The number of ways of getting from the entry page to the order page soon reaches infinity for any site with more than a very small number of pages. But you can look at how long it takes a customer to reach the order page and this is a very useful metric. The sooner the visitor places his order with the least number of clicks the better.

So how do we use all this information? There are two things it can do for you.

Firstly it will show you what is, and what is not, working before any changes are undertaken and this provides an invaluable starting point for the design team. For example, if you have a high number of returning visitors, look to see why they are coming back, and make sure whatever it is stays there after the makeover. Conversely, if there are very few returners, ensure that the redesign addresses this problem; perhaps a tip of the month on the home page is the answer, or a special offer will bring them back. It may simply be that the home page needs to make it clearer what the reason to keep coming back is!

Secondly, it provides a benchmark so that you can measure the effectiveness of the changes you introduce. If the very worst happens and the new site performs less well than the original, your website audit will be invaluable in providing with you the ability to identify what went wrong. If the traffic increases but more people leave, you will probably look first at graphic design issues. If search engine referrals suddenly drop, then the first place to look will probably be the spiderability of the architecture.

Web analytics and emetrics give the webmaster abilities to understand their customers and potential customers to a degree unimaginable in the offline world. Failing to capitalise on it is like trying to find the bulls eye with your eyes shut.