The advantage that comes with trading online is that you have a surfeit of data at your fingertips on who is shopping, when and why. However, many online retailers struggle to find quick, painless and effective methods of converting the data into something meaningful for their business. Hosein Moghaddas, VP & MD International of GSI Commerce, offers top tips on how online retailers can take the pain out of analysing what their customers are telling them, leaving the road clear for intelligent, actionable business decisions.
Asking the right question
The evidence is continually mounting that engagement-hungry customers are keen to do more than just browse on static websites; they want to do everything from recommending products, to participating in web or publicity design. The question is, are companies really getting what they need from customers in web design?
A common pain point for many e-commerce managers is not in gathering their data, but rather in organising it in a way that makes it easy, quick and painless to scan and action.
You might hear a merchandiser detailing the success of a promotional campaign and be impressed that a specific spot pushing a free shipping promotion produced £10,000 in sales for the 2 weeks it ran. However, the key question to ask is: ‘is that good?’ Or even more specifically, is that good for a free shipping promotion advertised on a pivotal point on the site for a duration of 2 weeks, that was clicked through 5000 times? Put simply, is your data arranged in a way to ensure your customers are getting the message?
What are hurdle rates?
In a nutshell, hurdle rates set success criteria for a given data point. It’s setting a number, based on past performance of your site, which allows you to quickly answer the question: ‘is this good or bad?’ In practice, they let you spend more time optimizing problem areas and less time identifying them.
Here’s a simple example. An e-commerce manager for a fashion retailer notices that in the past month, the phrase ‘tea dress’ was searched for 1,000 times and converted at one per cent. Previously, that same analyst had established a set of hurdle rates for onsite search and knows that over the past two years, terms searched for between 2,500 and 4,000 times in a month convert, on average, at 10 per cent product traction. That manager has, in a manner of seconds, targeted an item for investigation and potential optimisation.
So, how do I use them?
So, you’ve simplified the data on your page and are ready to define your hurdles. In doing this, the onus is on you, the retailer, when defining to be careful; doing so with the understanding that simple averages are rarely effective. For example, creating a hurdle that identifies your average conversion rate for search keywords and ranks all searches relative to that average might be a good starting point, but isn’t an optimal focus.
Instead, look to flesh out those terms in a variety of ways that are meaningful to the business. For a luxury fashion mass merchant, categorising high-performance branded terms independently, such as ‘Jimmy Choo’ or ‘Chanel’ might be one tactic. Similarly, for a site with typically high shipping rates, a free shipping promotion probably shouldn’t be compared directly to a buy-one-get-one free promotion. Defining the proper segments for your hurdles is nearly as important as establishing them in the first place.
Where to use them?
Hurdle rates can (and should) be applied to all types of metrics in order to understand exactly what constitutes good and bad performance. Think about how long your customers are visiting your site for, how many times they re-visit per month, and how much they are linking to your site through blogs, forums and other social media platforms.
“
