How Measuring Key Performance Indicators Can Improve E-Commerce Strategy – Part One
The problem with most e-commerce marketing strategy today is that companies don’t understand how they use things like web analytics. Most e-commerce directors or web marketers are given a budget and told to stick to it, and good analytics don’t usually come cheap. Without web analytics you can’t even begin to measure key performance indicators (KPI’s), which should be a part of any good e-commerce strategy. We often see that marketers face a problem in that they know they need Web Analytics, they just don’t know why they should pay for it and don’t know what to measure. This three part series of articles will hopefully help clear up some of the things that marketers should measure as key performance indicators concentrating on one KPI per article.
What is a key performance indicator?
In website measurement terms a key performance indicator is a metric which will help your organization define and measure progress toward your websites business objective. Key Performance Indicators are quantifiable web site measurements that reflect whether you are successfully meeting or falling short of your websites business goals.
That’s quite a boring definition of a KPI even if it is important, so in a last ditch attempt to keep you from falling asleep lets talk about Formula 1 (or the Indy 500) and the KPI’s they use.
What has Formula 1 got to do with KPI’s?
There are many minute factors in formula one that constitute being a winner. Everything down to the performance of the fuel, the tires, the speed of the pit stops, the quality of engine parts, the weight of the car, it’s aerodynamic ability, everything is measured and tested, long before the driver even gets into the car. The difference between the winner of a formula one race and second place can be as little as a hundredth of a second.
That extra hundredth of a second could be because the fuel used on that particular race day allowed the driver to get more out of his car than the guy in second place.
How did the race team know which fuel to use?
Because before hand they had tested maybe 50 different types, each one tuned for the demands of different circuits – or even different weather conditions.
They got that extra performance by knowing the key performance metrics of the fuel, so they could say with confidence that ‘fuel a’ was better for their car if ‘condition a’ was satisfied.
Condition ‘a’ might have been the cars weight that day the type of road surface and the weather. When all matched together it meant that the race team had a particular choice to make when selecting the fuel for the car.
The web site KPI I’m about to discuss is the fuel that powers your e commerce sales and lead generation strategies. Both are measured by practically all web analytics systems, but both not commonly measured to their full potential.
Page views per session, the fuel behind your web business objectives
For those of you that know why page views and sessions are important bear with me for a paragraph or two. For those of you that don’t here we go.
Why are page views and sessions important?
Page views are a metric that represents the amount of times your pages are viewed by the people that visit your website. On it’s own it might be an important measurement if you’re a very well trafficked content website looking to sell B2B advertising in the form of some kind of ad (banners for instance).
If you can accurately say to an advertiser that you have 10 million page views per week, it’s very likely that this alone will be one of your KPI’s, simply because if it goes down, your advertisers will most likely not want to pay you as much to advertise with you. It would be important in this case that you keep the page view count at least to the same level every week in order to keep the same level of banner revenue for example.
Sessions represent the amount of users (people) visiting the website over a given period. Again it’s a very important metric to know, the general idea being that more of the right kind of people visiting your pages will eventually mean your bottom line improves.
By combining these two metrics however we get a much more powerful way to use the figures.
Combining the two metrics as one KPI is done by taking a ratio of page views per session as an average. So if 1000 visitors viewed 2,000 pages the mean page views per visit KPI is 2, (2,000 / 1000 = 2).
Why is this combination important?
If your website e-store system required that you need to view 5 pages in order buy a product and your KPI is telling you that your site gets an average of 2 page views per session, then the site is under performing badly.
If it takes 5 pages for your visitors to buy something then your goal should be to get an average KPI of at least 5 page views per session. Otherwise it means that the vast majority of your visitors aren’t going deep enough into the process.
Much more importantly deciding on a KPI like this is giving you a measurable objective to work towards. If you know that the vast majority of people are abandoning your website after only viewing one or two pages there is a problem which you need to work hard to solve.
It means you know that somewhere within your web analytics you will be able to detect the areas of abandonment that are the problem. The simple fact of the matter is, if you have a low page views per session KPI then your analytics system (if it’s any good), WILL be able to show you where the problem lies. Once you have found the problem areas, congratulations, you’re becoming a web analytics expert. Now you know which pages have the problem and you just need to figure out the why.
Figuring out why is the real secret
It may be that you’re driving the wrong kind of visitors, such as people who aren’t interested in your offer. It may be that you don’t have enough compelling content to keep visitors interested. It may be that your shopping cart has a problem with abandonment or your lead generation process is too long or has a scary form to fill in. In all cases your page views per session KPI is the first warning signal and you can monitor it quite easily.
Other warning signals
The other side of the coin is if this KPI is too high. What if you have a 1000 people viewing 20,000 pages? Unless you have incredibly compelling content there is a problem. It probably means that people are very interested in something but can’t find it on your site. So having this KPI be too high is also a warning flag that means you need to analyze your web analytics and see where the problems are. Are people skipping around pages? How long do they stay on your site? (another KPI we’ll come to in the next article of this series), have you got site architectural problems with navigation?
Too high or too low, it’s all useful measurement
The point is to find out how to use the fuel remember? If you start measuring page views per session as a KPI, you will begin to see if you have a problem or not. You can get as deep and as sophisticated as you like, I’m just trying to show you the idea with this article.
For instance you could measure page views per session of visitors only hitting your shopping cart, or lead generation system. Content websites could use page views per session in particular content groups to work out how compelling particular kinds of content are. It all depends on the web site business objective.
Developing KPI’s allow you to measure things on your website which directly effect your business objectives. In the example I’ve demonstrated by finding ways to improve the number of pages people view per session there is more chance that those users will complete your calls to action (buy, register, subscribe, whatever web business objective you may have). This first KPI that I’m suggesting you consider is an early warning signal that something is wrong and it’s very easy to determine how to set a measurement. The next KPI I’ll discuss is time spent on site and why it is also important as well as how you can use this in combination with page views per session.