How To Measure Website Performance
Updated: Sep 6, 2020
If a website is a key tool to improve your businesses sales (via online bookings, leads, calls or whatever), then the measurement of its performance is a fundamental requirement of any business.
So what do we measure? Firstly, we need to know what you want your website to do. Every website should have an objective that we want a visitor to achieve. It could be an online sale or booking, filling in a lead form or making a telephone call. These are commonly called conversions.
It is much easier to work with a website if there is just one conversion type but, very often, a website will have more than one conversion type as an objective. E.g. a travel site may offer the ability for a customer to book something online but they may also offer the customer the ability to fill in a contact form for someone to get back to them regarding a more bespoke requirement, say.
Our first and most important measure then is a conversion. This can be measured in terms of units , i.e. 1,2,3 etc, and also may involve a monetary amount. So, for example, a sale of a product is a conversion (amount 1) but it also can have a monetary value tied to it, say £50. Both values are useful obviously. Sometimes though there is no monetary value (yet) tied to a conversion. An example of this would be when a form is filled in on the website by a customer asking for more information or to be contacted.
So where do we see the data for conversions? The conversion action on the website will trigger something that tells you that it has happened. You may get an email confirmation, for example, when a form is filled in on the website. Alternatively, if you sell a product then this information may be entered into a business or accounting system either manually or automatically. So we have a mechanism for tracking the business performance using other systems not directly connected to the website. The tracking of the business in this way is an important source of data but clearly it is potentially very labour intensive as volumes rise.
Google Analytics is where most businesses start on the road to analysing website performance. It is software provided by Google free of charge and it quickly integrates with virtually all websites regardless of platform. We will call this GA from now on!
You can tell GA what your conversions are and how it should look out for them. You can also tell GA if your conversions have a monetary value associated with them. This value can change dynamically depending on what the website customer transacts so that GA knows the exact monetary value of every conversion (if one exists) .
So we understand our conversions and we have a way to measure them. This is a big step forward and it is very easy to see in GA conversions each day, even each hour of the day etc.
Clearly conversions are only one part of website performance measurement though. So what else should we measure? There are literally hundreds of different metrics that we could look at, most of which are relevant in some way but there are actually a small number of fundamental metrics.
Let's talk about traffic. Most people will have heard the term website traffic. In fact, very often it is talked about first, before conversions, but if you think about it that doesn't make sense. However, traffic is very important. Website traffic is a funny term in a way as it does not have one metric it has two, users and sessions. A user of a website is just that a user. It counts the number of people irrespective of how many times they go on the website in a certain period. So if I visit a website 3 times in 10 minutes I am 1 user. A session, therefore, is the number of times a website is accessed and will be a bigger number than users. So in the previous example of 1 user that would be 3 sessions potentially. I say potentially as there are other factors in play but just to give the overall idea. So which metric should we use? In reality, it does not matter. The interplay between sessions and users is important but for other metrics using one, say sessions, makes sense.
So we now have 2 metrics, conversions (of which there could be different types as discussed) and traffic for which we will use sessions.
Conversions and sessions. With these two metrics we can now introduce a third, highly talked about metric called conversion rate. Conversion rate is measured by dividing the number of conversions by the number of sessions;
Conversion rate = number of conversions/ number of sessions. This would be for a given time range, so you could measure conversion rate each day or each week or over a month depending on what makes sense. It is expressed as a % typically, e.g. 3%. Note that conversion rate has nothing to do with the monetary value of a conversion.
So you get 100 sessions on your website resulting in 3 online bookings and 2 online forms being completed. Your conversion rate is 5% being (3+2)/100.
If you measure conversions, traffic and conversion rate over time then you will be able to see how your website is performing and if any changes you make to your offer or website improve these numbers.
These three metrics are definitely the among the most important and for many businesses that can't tie a monetary value to a conversion we would stop there as regards key metrics. However, for businesses that can tie revenue to a conversion, e.g. online sales, then there is one other metric that is very important and which goes with this group.
Per session value is a calculation of conversion value (or revenue) divided by the number of sessions. So, if a website has 200 visitors and from these visitors gets 8 online orders with a combined revenue of £400 then per session value is £2. Simply, £400 revenue/200 sessions. This metric represents the revenue per session, i.e. the amount of revenue for every single session on average. Even though most website visitors don't complete a conversion, when we spread the conversion value over all visitors then it gives us an important idea of what we can spend getting those visitors to the website.
Without going into lots of detail, if our product margin was 50% then in the above example we would be making a per session margin of £1 (from, in effect, revenue of £2 per session). This is an important indicator because if it costs us £1.50 in marketing costs for each session on the website on average then we are in trouble! If, though, it costs 50p then we are making a profit of 50p per session.
Per session value leads us into the area of traffic acquisition which is not the topic of this post but it rounds out the 3 or 4 key metrics as a starting point for all websites.
Conversions (and conversion value if applicable)
Per Session Value
Once you begin to understand how these metrics contribute to your business and how they are affected then you will start the journey of understanding the value of your website's performance.