How To Optimise Your Product Mix - 4 Step Merchandising Made Easy

Posted by James Dunford Wood 28 Jan 14

The vast majority of online retailers started life, and still trade, offline. Most are still small businesses, with online revenues representing less than 20% of their total turnover. But as online shopping begins to take an ever larger share of retail spend, retailers need to take a close look at how they merchandise online - because for many, the online channel is still treated as an avenue to shift stock rather than an opportunity to sell a different mix of products.


It’s as if you had a store in two different parts of town, servicing two different demographics. The visitors you attract online will be in many cases completely different than the ones you sell to in store - so your product mix should reflect this if you want to take advantage of the powerful distribution and reach that the web has to offer. Even if nothing else changes, retailers who merchandise properly online can trade substantially better than those who don’t, which will be reflected in their bottom line.

Here are four not-so-simple steps you should take if you want to start doing it properly:

Step one: get to know your neighbourhood

When you are running a bricks and mortar store, you have many signals and indicators to go on, including well-defined demographic stats for the neighbourhood. For online trading, there is no neighbourhood and there are multiple signals coming from multiple directions. How to make sense of them? Start by running an online survey, and polling your existing and new customers on their needs, buying habits and preferences.

Step two: segment your products

A second, and very important, way to understand your market is to look more specifically at the types of products your customers are buying. Begin by segmenting your products into types - they may already be organised into categories like shirts, trousers etc, but try grouping them under ‘designer’, ‘practical’, ‘gifts’ etc - all groupings that might attract a particular customer or demographic segment, and which may well cross several categories (for example a shirt and a pair of trousers may be in the ‘luxury’ group.) Once you have done this, you can start to segment your customers into groups. Do this in two ways - first, by these new categories - ie customers who buy ‘luxury’, ‘designer’, ‘gifts’, ‘small homewares’ etc, and second, by price point. Ignore average order values, and concentrate on comparative pricing - for example, compared with other items in a similar range, was this product expensive, median or cheap? (You may have to create new product groupings to analyse this properly). 

You should then be able to start to see patterns. Which are your largest customer segments, and what do they like to shop for? Try re-running this exercise over seasonal dates to see how things change. The goal is to end up with several core customer groupings to focus on.

Step three: analyse product performance

The next exercise you need to undertake is to start to analyse what products sold well and why, and which didn’t. You may have 100 or 1000 products in your online store, but regardless of the number, you will find that, typically, a small proportion of them got most of the traffic. This could be down to a number of reasons, often more to do with the search popularity or certain items and how your website is optimised, and what other competition is out there. So hone in on your most high trafficked products, and then analyse by conversion rate. Which performed below average? Can you see why? Was it imagery, poor description? If you are satisfied that these were not at fault, it’s most likely either price or poor market fit. Put them on a list to test with different pricing.

Then look at the good performers, and analyse where the traffic came from, and the customers who bought them. Which customer groups do these belong to? Do they match with one of your core customer groups? If so, you have a hero product on your hands, and you need to see whether you can do more to stimulate traffic from the sources that drove people in to buy them.

Step four: work profit into the equation

So now we have identified core customer groups, products that need working on, and hero products that need pushing. The last piece in the jigsaw is profit. Most online marketers are incentivised by revenues, and this can be a mistake. If you spend enough, and discount enough, anyone can get to a million dollars in revenue. It’s margin you need to look at. So re-work your revenue figures into profit, and see how this changes the mix. Of those products that weren’t converting, how much additional profit (not revenue) could you make if you upped their conversion to the site average? This will immediately tell you which you should prioritise for price testing - not only because you will have more margin to play with than other, less profitable products, but it will also show you how much you will make from your efforts. Even effort has an ROI!

Repeat with your hero customers, and your core customer groups - you will soon see, amongst your core customer groups, which are the most profitable, and at the same time which of your hero products are doing most to lift your bottom line.


I’m not suggesting this analysis is straightforward - but it is very rewarding, literally. If you can crack your core customer groups, your hero products, and the revenue you are leaving on the table through poorly performing products, you are 80% of the way towards perfecting your product mix to match the customers you want to have. Happy trading!



Artboard 10 B-1.png