Our client TescaraHats (name changed for commercial reasons), a European market leader in manufacturing customized hats, decided to expand its market reach with an e-commerce site where its potential customers could choose, customize and order hats online. TescaraHats used APM tool to fix a number of e-commerce application performance and usability issues we presented in this mini-series. But there was still a missing link that would ensure TescaraHats increased sales.
I guess you thought we forgot about the conversion rate? Nope, here is how an application performance management (APM) tool can tell you why it is so low.
Why Is Your Conversion Rate So Low?
Provided your marketing strategies were successful, users should be swarming into your e-commerce site. But that does not ensure higher sales. According to aforementioned E-Commerce Usability report almost 2/3 of users who found an e-commerce site abandoned it and proceeded to do business elsewhere.
TescaraHats had a similar problem. Its expectation was that there would be a direct correlation between increased site visits and increased orders. The report in Figure 1, prepared with the APM tool, compares number of users each day between secure and non-secure session; the secure session on TescaraHats e-commerce site is used only to complete the purchase. According to this estimates their conversion rate was at the 6-7% level even though according to the ConversionXL report customers like to customize their purchases because it increases the feeling of ownership.
Figure 1. Only to complete the purchase users need to go over HTTPS: CR is at the 6-7% level ↩
Why did so few customers complete their orders?
Nielsen suggests that e-commerce sites should not threaten customer loyalty with an aggressive push for registration requirements. Apart from forcing users to sign in or register for a newsletter in 81% of evaluated e-shops, the study by Christian Holst also lists other usability violations in the checkout process, such as: asking the same information twice (in case of every second evaluated e-commerce site) or lack or improper address validators.
When TescaraHats analyzed the KPIs of operations of the final purchase process in Figure 2 the team discovered that, apart from performance problems with the shopping cart, users were reluctant to fill in personal details in the registration form. When presented with a long registration form, 25% of the customers aborted the purchase by closing the registration page even while it was still loading.
Figure 2. Two reasons why users do not complete the purchase process: slow shopping cart and personal details form ↩
No wonder Nielsen insists to pay more attention to shopping carts, checkouts, and registration. He claims that unless the customer can quickly complete the purchase through a streamlined checkout process, the e-commerce site may lose the sale and turn the customer away from future visits.
Improving sales through the e-commerce site is not only about getting ranked higher in search results. It is also largely dependent on the user experience and the application performance.
When TescaraHats started managing its e-commerce performance with Compuware dynaTrace Data Center Real User Monitoring (DCRUM) it figured out that its sales did not increase when site visits increased because:
- More than 40% of buyers used the Safari web browser but their experience was not top of the range.
- The hats configuration wizard was very slow: it took 12s to load. The major problem was with unnecessary 240 HTTP calls to the server and many HTTP 404 and 500 errors on the backend.
- Listing available hat types was slow due to the bad performance of stored procedure at the database.
- 25% of potential buyers dropped out at the checkout due to elaborated registration form.
Annoyances, whether caused by bad design or low application performance, can be very disruptive to the online business. In case of TescaraHats the conversion rate was as low as 6% at the time it started managing the application performance of its e-commerce site.
(This series is based on materials contributed by Pieter Jan Switten, Pieter Van Heck, and Paweł Brzoska based on original customer data. Some screens presented are customized while delivering the same value as out of the box reports.)