Using Google Analytics to Lift Sales in E-Commerce Sites
Showing the right product to the right person at the right time is one of the most important factors in a successful e-commerce website. This article demonstrates a simple technique that can be used in Google Analytics to identify opportunities to improve sales, by showing the right product.
If your online store is like most, a large percentage of your visitors arrive from a Google (or other search engine) query. An effective search engine optimization strategy can bring a steady source of new customers to your online store.
Keyword searches can be broken into two major groupings, brand and non-brand. A brand search visitor is one who has specifically included your company or brand name in their search query. A non-brand search visitor has not. It seems reasonable to assume that a visitor who knows your brand is likely to be more aware of your offer and hence is potentially easier to convert than a non-brand search visitor.
In this article I will examine how we can use the keywords from the non-brand search visitor to identify opportunities to improve sales.
I have assumed that you have set up the Google Analytics e-commerce tracking script on the page that the visitor sees when they have completed a purchase. If you haven’t already set this up then follow this instructions.
If you do have the e-commerce data flowing into your Google Analytics account, you can use a custom report to identify which keywords are bringing sales, which are performing well and which need work. If you are not acquainted with custom reporting, below is a quick video released by the Google Analytics team on how to create one.
I have presented the custom report described below to many clients who have previously had no idea that they could extract such rich data on their key opportunities for improvement so easily. What we want to achieve is a report that shows various aspects of the keyword that the visitor used to find your website and provide key information on that keyword in one location.
Creating The Custom Report
The custom report will use 4 tabs which will include the following details:
1. Sales by Keyword and Landing Page
This report will show your sales data, including the entrances, bounces, transactions and revenue. This is the primary report that you will use to locate underperforming keywords.
2. Behaviour by Keyword and Landing Page
This report will include some additional behavioural data including time on page, time on site, searches and transactions. This is useful for checking whether the keyword and landing page is at least attracting interested visitors
3. Keyword by Country
This report is included to check that you are able to sell to these countries and identify possible geographical opportunities.
4. Sales by Number of Visits
This report is included to provide a sanity check to see if a visitor may return at a later stage to complete the purchase. In my experience the data in Google Analytics indicates that most sales from non-brand search visitors occur within 1 to 3 visits on the same day.
You should see a screen like this:
Once you have created the custom report you should see the edit screen. Locate the filters area and change the value for “your brand” to your brand name.
This is set to use a regular expression so that it will match a range of queries. If you want to include a list of brands you can do so using the pattern: brand 1|brand 2|brand 3. The | symbol is treated as an OR in this pattern.
Once you have set up this report you can identify which keywords are bringing traffic to which landing page and whether they are converting into sales.
The report works best if you can pick a long date range, say a few months at minimum. If your site has lots of traffic you may find that the reports show that the data is being sampled. In this case you will see a message similar to this just below the data range.
It is important to understand that in this case the reported results for keywords with small numbers of visits and sales are likely to be inflated.
Analyzing The Custom Report
Look for pages that have a low ratio of entrances to transactions. E.g. if there is 1 sale from every 1000 entrances then there is something wrong with that keyword.
The image below shows a very low ratio of sales for the keyword homewares online and homewares. The bounce rate is reasonable at approx. 26% but few transactions are being conducted. This is a prime opportunity to examine what can be improvements can be made on the /homedecor page.
The behaviour report shows that these visitors are conducting searches, but don’t appear to be finding what they want. Looking at their site search queries using a segment would be a good next step.
Using the Google Analytics Site Search report and drilling down into their search queries can provide you with insights on what products to display when a visitor comes into a category based page.
In this case showing a combination of clocks, floor lamps, mirrors and vases may be a good start. Refining this over time based on the changes in sales will lead to a growing understanding of what does and doesn’t work for this market.
This relatively simple approach provides a wealth of opportunities for growing sales by effectively using the data in Google Analytics.