It’s not always about conversion rates!
The question of free delivery is one that every e-commerce store will have asked themselves at some stage.
Do we offer free delivery to increase conversion rates? If we, do what should the threshold at which we offer it be?
Logic suggests that if you increase free delivery threshold, you will increase average order values. Customers will either have to spend a little bit more to get free delivery, or they will have to pay for the delivery themselves.
There will be price to pay for this uplift though. With customers having to pay more to get free delivery, it is likely you will get less orders and your conversion rates will drop as a result.
The trick is to find the balance and that happy medium varies from company to company.
No margin for error
Is it better to have more orders if you are having to absorb the cost of delivery? Is this more profitable than having less orders with a slightly higher order value?
These questions were the topic of conversation with a client recently. The nature of their business model made it really important to get right. With relatively low average order values, there was little margin for error.
Their free delivery point was quite low. We we’re worried that if we raised it too much, it would significantly affect conversion rates. When you’re in the numbers game, this is a high-risk strategy.
We needed to test free to find out what the optimum threshold was. Where did the balance between average order values and conversion rates lie?
Is it better to have more orders if you are having to absorb the cost of delivery? Is this more profitable than having less orders with a slightly higher order value?
How do you measure something with so many variables?
This test isn’t one that we can measure using traditional KPI’s like Conversion Rate or Average Order Value because one is likely to improve at the expense of the other.
Tracking revenue or the number of transactions wouldn’t work either. What if we were to have exceptionally high or low traffic levels? What about seasonality?
We could consider how much revenue each user is generating, but that doesn’t take into consideration that the profitability of the order will be influenced by whether delivery has been paid for by the customer.
The only accurate way to measure to this is with a KPI that many site owners do not track – Profit per User.
Profit per User removes any influence the above variables can have. It accounts for the quality of traffic, the amount of traffic, the conversion rates and gives you the true value of each visitor to your site.
What we did was to test Profit per User at the site’s current free delivery threshold for a given period of time. We then increased the threshold slightly and tested it for the same period of time.
We tested a few variations and compared the results to give us the site’s optimum threshold.
Make sure you look in the right place!
Byfocussing on one KPI – Profit per User – were able to make sure we made the right decision based on what was the most profitable option for my client. We weren’t influenced by headline figures like conversion rates, that may conceal less profitable orders.
By testing different options, in increments of £5 over a period of time we were able to make sure we found the optimum balance and unequivocally answer the question of what the right threshold was.
It’s all about profitability!
Ultimately, it’s a numbers game and these tests are about delivering results. Here’s how we got on…
Unsurprisingly, by increasing the free delivery threshold, we were able to increase average order values.
What was surprising was that conversion rates were not affected. They remained constant, which of course was a nice bonus!
Furthermore, we discovered that there were a large number of customers were happy to just pay for delivery.
The net result was a 41% increase in the Profit per User. Due to the relatively low average order values, high conversion rates and low free delivery threshold, the affect the test had on order profitability was quite impressive.
It also goes to show that there is much more to performance of a website than vanity metrics like conversion rates!
If you would like to find out more about how I can help you improve the performance of your website without spending more on marketing, get in touch today and book your free 20-minute consultation.
