Search
Close this search box.

Guest comment: Online product recommendations – five steps to improving profitability

This is an archived article - we have removed images and other assets but have left the text unchanged for your reference



by Darren Vengroff

Online product recommendations have been around for almost as long as e-commerce itself. Most online retailers today offer some type of product recommendation on their sites, but the truth is that many are not making the most of the opportunity. They plunk recommendations down on product pages, do little to monitor performance, cross their fingers and hope that sales will increase a bit. Often there is an initial spike in conversion, but performance tends to fizzle out after a few months. If product recommendations are irrelevant and ill-timed, the retailer risks alienating site visitors so much that they may go elsewhere for their immediate purchase and never return.

When done correctly, relevant product recommendations can deliver a 5 to 30 per cent lift in sales conversions instantly and sustain, or even increase, these levels over the long term. The following top tips will help online retailers to get more out of product recommendations and to provide customers with the best possible shopping experience.

1. Use a variety of recommendation techniques

There are really four main ways to deliver online product recommendations. Segmentation divides users into groups based on characteristics such as age, gender and geographic location. Collaboration, or collaborative filtering, starts with an individual and attempts to locate others like them. Personalisation relies on a user’s prior actions to determine what they are likely to do next. Similarity starts with products, rather than users, and models relationships between them to serve recommendations.

Most recommendations platforms use one or two of these techniques, but in today’s high-stakes market, that is simply not enough. To deliver relevant recommendations, retailers need to use a platform that leverages all four at once. Retailers like Sears and Overstock.com, which use more than one recommendation technique, have boosted online sales by up to 25 percent over the long term and driven significant increases in user loyalty and repeat purchases.

To create the best shopping experience, a recommendation system should continually test the performance of all recommendations, optimising and automating the combination of techniques to deliver the best results.

2. Give merchandisers control

A common mistake is the adoption of a recommendation system that takes merchandisers completely out of the equation. This will never work. Merchandisers know things that no recommendation “black box” is ever going to know. For example, they will be aware that a particular camera is going to be replaced by a new model three weeks from now. Opt instead for a solution that allows merchandisers to manually override certain recommendations and fine-tune them as required. For example, a merchandiser may want to increase the rate at which they recommend overstocked products, but still restrict the recommendation to those people who are most likely to be interested in them. A successful recommendation system does not try to replace the merchandiser, but instead provides the tools to do “bionic” merchandising online.

3. Put recommendations everywhere

Today’s consumers have multiple channels for shopping, so retailers need to recommend their products to them everywhere that they are: online at a retailer’s website, online at an affiliate site, looking around a store or reading an e-mail on their iPhone. The recommendation solution provider should enable the extended use of personalised recommendations across all customer touchpoints, both online and offline, and throughout the customer lifecycle. This could be from the minute they view a retailer’s online advert, land on a homepage, browse a website, or a few days later when they get a promotional e-mail revisit a website. If retailers constantly recommend relevant products to consumers, no matter where or how they are engaging with their brand, they will increase customer loyalty and encourage repeat visits.

4. Get the message right – be specific

When retailers add clear, understandable language to each recommendation, conversion not only increases more significantly but customer service requests go down. A good recommendation engine allows messages to be added so that customers know exactly why those products are being recommended to them.

Mistakes are often made in the way retailers communicate recommendations to site visitors, presenting products in a way that turns off potential buyers. It is important always to explain clearly where the recommendation came from and why it has been recommended it to this particular person. Instead of “May we recommend”, which sounds like an attempt to unload excess inventory, messages like: For example, “People who purchased this HD television player also purchased this stand,” justifies its relevance and is more likely to result in a sale.

5. Avoid classic mistakes

There are some simple pitfalls to avoid when making product recommendations:

• Recommending a product a client is already looking at.

• Recommending controversial or inappropriate products that may damage a brand, such as X-rated films in a children’s film category.

• Not giving customers control over their recommendations (for example, not allowing them to tell you that a product is not of interest).

These mistakes are avoidable if retailers use a next-generation recommendation platform that uses all four techniques of segmentation, collaboration, personalisation and similarity.

Personalised product recommendations are the start of a dialogue with customers. The end goal may be more sales, but to get there, the customer relationship needs to be nurtured through thoughtful, relevant recommendations that are truly useful to customers.

Darren Erik Vengroff is chief scientist at RichRelevance.

Read More

Register for Newsletter

Group 4 Copy 3Created with Sketch.

Receive 3 newsletters per week

Group 3Created with Sketch.

Gain access to all Top500 research

Group 4Created with Sketch.

Personalise your experience on IR.net