The basics of web analytics for an online retailer
What to write about first? Well one of the strongest sectors for Logan Tod is online retail, our current clients include ASOS, Debenhams, Focus DIY, HMV, to name a few, so lets start there. A key thing we have found over the years is that the first area we look at for an online retail website is always the same. This is relevant for all online retailers, whether they are in the top 10 in the UK or a small project being run by someone trying to see if they can quit their day job.
Whilst each business/website is unique, they do share common characteristics in the stages of the purchase process and these should be the starting point when looking at their web analytics data. The stages of the purchase process for an online retail website are:
- Arriving on the Website
- Engaging with the Website
- Finding a Product
- Creating a Basket
- Commencing the Checkout Process
- Placing Order
There are many different reasons why a website may have a poor conversion rate and by breaking down the performance of the website into these areas, you can understand where the biggest problems are and thus where you should focus your attention. Trying to fix everything at once is a recipe for uncoordinated disaster, so by knowing which area to work on first, you can maximise initial gains from web analytics (and cover yourself in glory!). Some brief points on each area are noted below and hopefully each will be covered in more detail sometime during the next year.
1. Arriving on the Website
So you have traffic. People have found you. How? Where did they come from? Did you have to pay for them or were they “free”? The key point to look at during this stage is whether there are any opportunities for more traffic that you hadn’t thought of.
2. Engaging with the Website
Or more accurately (in reverse), not immediately leaving the website. This gives a view on whether what the visitor saw when they landed on your website matched what they were anticipating. If not, there is a fair chance they will leave immediately meaning you have no chance to sell anything to them. The key here is to drill into the details, to look at the level of engagement (or once inverted, the Bounce Rate) by traffic source and by landing page, to look for weak points.
3. Finding a Product
This stage is calculated as the proportion of visitors who view at least one product page. It is a requirement for purchase that the visitor gets to this stage because they can’t purchase what they haven’t seen. This stage is a test of navigation and of the range of products on the website. Do you have what the visitor wants and crucially, can they find it?
4. Creating a basket
This stage is reached the first time the visitor is interested enough to add a product to their basket (or cart or bag or …). It is the first real signal of interest; the customer has found something they are interested in purchasing. Key areas that can detract from this are how well you promote the product (image, copy on page), whether the product is available, the price point of product – all sounds fairly similar to the old 4 Ps of marketing really.
5. Commencing the Checkout Process
The visitor has found products they are interested in, in fact they have a basket full of them. What can you do now to push them over the edge and get them to start the final journey towards giving you money? The basket page is possibly the single key page on an online retailer’s website with everything about it needing to motivate the visitor to click that big red flashing “Commence Checkout” button (or personal variation on it).
6. Placing Order
The final stage covers the checkout process, whichever details you need from the customer before you take their money. They are so close, make it as simple as possible. Quick tips, don’t encourage them to navigate elsewhere on the site and most definitely do not require the visitor to register before they transact (instead collect necessary details for the purchase and ask them if they would like to give you a password after they have made their purchase).
Next Steps
The big question now is likely to be how can you tell if you are performing well or poorly within each stage. To identify this, you do need to compare against industry benchmarks or targets. Again more on this in future posts although you can get a sneak peak with the tool we commonly use for this type of diagnosis, the Logan Tod Performance Gap Model.

