Ecommerce Metrics - Overall Conversion Rate Trap

Top-level KPIs and useful metrics: Purchase conversion rate is an example of a top-level KPI for a business, but too much contributes to a conversion rate for it to give clarity. Why is it reported as a KPI for websites? This is due to misunderstanding or a lack of realisation of what affects a business's e-commerce.

Having just reviewed another business website, I can say that the purchase conversion rate only loosely correlates to revenue (0.45%) or margin. This observation goes against what many people have been told they refuse to believe, but numbers don't lie. The highest revenue months often don't have the best conversion rate, session and user surges can lead to more sales, but not at the same conversion rate as your regular website users. I've seen negative correlations between the overall conversion rate and sales. This myth of a "good overall conversion rate" needs to be stopped—it's a trap!

But don't deliberately make your website worse. Use better metrics and KPIs to show your business's e-commerce true successes and areas for improvement. It's a good metric but beware of the many elements that affect its changes.

Crucial Funnels - Superior E-commerce (On)Website Metrics

Metrics, and conversions that correlate highly with revenue are critical; stats from engaged users will provide stronger indications of healthy e-commerce stores. Views on product pages are key. Sales funnels from item_views, category pages and search results are key, with funnels around your accounts and checkout success.

Break it down further, new, returning and account users you are looking for account users to convert better than standard returning, and new is likely the lowest. Account users, behaviour is another set of metrics that can be tracked, more on that later.

Add value to these reports and identify where your users are going after they leave the funnel, especially the ones that never return, session and user conversion rates are also important to distinguish.

Reports should also be able to indicate customer lifetime value, paired with an annual customer spend which could be a rolling figure used as a share of wallet indicator.

Engaged User Ratio - Share with your marketing team!

The engaged users/session ratio is an interesting behavioural metric but it's highly dependent on many elements. The page might be good, but the users coming in might not be interested. Viral social media posts reach huge impressions and click-through rates, but the user has minimal intention to purchase, so including them in a session conversion rate is not helpful.

Break large metrics down to traffic sources and user groups making them more specific to those responsible for that channel or section of the website. Is the right audience being reached? Do you need a specialist landing page? Was it the right product/service? Does the website have errors appearing in certain browsers or devices?

Important - NO ONE-DIMENSIONAL METRICS!

The theme that I'm trying to drive home is that top-level metrics like overall conversion rate & engagement rates (or bounce rates) are useful they only become useful in conjunction with other underlying metrics. Presenting the Conversion Rate as a KPI to the C-suite if they assume all those factors feeding into are under your control when they are not, can lead to misunderstandings.

You can apply the same idea to AOV variation caused between categories, which could be a seasonal combination of both that can distort that month's AOV. UPT which could skewed with how bundles are sold or the use of add-on sales/free items. Name any metric at the top level that might look good on presentations. At best you can't use them to inform meaningful updates, at worst, they cause trouble when the updates don't seem to be working even if they are.

Apply Context

Business situational context, during the COVID pandemic I saw spectacular conversion rate increases. This was due to people's behaviour change not going out meant "home entertainment" became much more popular. However, not all funnels and user groups kept up the positive change only looking at the overall conversion rate would have led to not taking advantage of this increased customer base and the insights this was giving us. With the insight gained we were able to ascertain holes in customer retention, repeat purchases not occurring as often, and customer surveys assisted in finding issues in the accounts section which is now far improved.

Any e-commerce manager should have access to key business changes. Not only sales & marketing events and any website-specific changes or challenges. But anything that might mean affects the ability to carry out a sale. Have them appear on your reports with dates, saving time looking for them. For YOY reports why is a conversion rate down? Flash sales, a radio ad went out. In a dynamic business, lots can affect reporting numbers, so be sure you're not chasing a red herring or taking false praise.

Essential KPI's

User conversion rate is your snapshot for the business, and session conversion rate to expose specific website issues, for each of these key funnels, from category pages, search results and product pages. User and session conversion rate

Customer lifetime & annual value, repeat customers will be a good indicator of successful products and services. You will have to apply industry-relevant knowledge to gain information from the data. Selling cars online is unlikely to have a lot of repeat purchases, selling food items or FMCG then you would expect sales to start repeating if not then something in your proposition isn't quite right.

You could add a long list of other metrics to your KPI list but breaking down the key funnels will give you more than enough for a one-page report and a starting point for digging deeper to gain the improvements that translate to improved sales.