Increasing Conversions Through Effective Product Recommendations

TAGS: AI CX Merchandising Online Shopping Retail

Making product recommendations relevant and engaging is key to conversion success.

I love product recommendations. They can be so useful, so interesting and have so much potential, but are often ill thought out and badly executed.

This makes me very annoyed, and it should make you annoyed too because it is losing money for your business.

On an almost daily basis I ask my clients “Why are you recommending these products?" In 9 out of 10 cases, you can bet I'm met with the words, “I don't really know."

A little bit of TLC can make your product recommendations go from a Shakespearean tragedy, to an Oscar winning performance.

We've all experienced shoddy recommendations at some stage or another during our time shopping online. If you have experienced inferior recommendations, you'll know how annoying they are.

This article will show you how to increase your conversion by using effective product recommendations for online retail.


But why are Poor Product Recommendations so Annoying?

They're annoying when the recommendations made are completely irrelevant or aren't suitable based on your buying patterns. If you've shopped at one of the biggest UK health & beauty stores on the high street, you will be familiar with being asked if you want to buy a multitude of discounted irrelevant items whilst you're at the checkout.

No, I don't want to buy a celebrity perfume for a tenner. Why would I, when I'm buying Lemsip?

It makes a shopping experience which should be enjoyable, or at least easy, instead frustrating.

For product recommendations to be commercially successful and useful to your customer, it's vital to avoid this kind of behaviour. A behaviour which clearly shouts, 'I'm not recommending this to you because I think you will like or need it and I know you as a customer', but instead says, 'we've bought this in massive volumes and need to push it at discount to absolutely anyone who will take it. We need to reach our targets this month. Please buy it.'


So how do you Avoid this Behaviour?

It's important all product recommendations have some context. Having context instantly makes the item, and reason for the suggestion, understandable. It's the difference between a product being perceived as old tat, or tantalising.


How do you Know if the Recommendations are in Context and how do you Encourage People to Buy them?

You must carefully think about where the recommendations are placed on the site. At what stage in the user journey will your customer be presented with recommendations? Then think, what is the most logical product, at this stage in the user journey, my customer will be interested in?

For example, if you're browsing black evening dresses and you've clicked through to a product detail page, what is best to recommend here? Other black evening dresses or would you cross sell outfit suggestions?

Until a choice is made, it could be argued that as a logical starting point, it's best to assume the customer is still browsing and you should help them to come to one decision before you suggest another item.

Always try to imagine how your online retail recommendations would appear in an in-store environment. If you walked up to a shop assistant and said you were interested in a black evening dress and they started directing you to handbags and shoes, you'd probably by correct in assuming the sales assistant would soon be on the receiving end of a P45.

You'd never willingly provide a service like this in a store, so don't employ strategies online which supply the same bad experience.

Once a user has made their initial choice, by adding a product to basket, this is the perfect opportunity to recommend complementary items.

(Cue good sales assistant thrusting relevant bags and shoes in your face.)

At this stage in the journey, a customer may or may not be interested in buying further items, but offering them up isn't seen as such a hindrance. It's just good customer service. Many customers will be prepared to at least browse these items to see what you suggest, and if not, they'll pass them up with minimal frustration and disruption to their experience.

Recommendations are a particularly welcome addition when the customer is purchasing a product which requires other necessary items to complete a task. This is especially true for home DIY and building projects. If a customer is fitting a shower and they have just added a shower enclosure to their basket, it is useful to show them the shower trays and additional fittings which are essential to complete the job.

Then that takes us on to the basket page. The checkout is always an area of contention when it comes to product recommendations.


Is it Dangerous to Recommend Product on the Basket Page? Does it Distract and Confuse Customers? 

I think it's important to A/B test this with your customers and see how they react, rather than avoiding basket recommendations purely because you've heard it is a distraction. 
Basket pages are an excellent opportunity to recommend complementary items to your customer before they complete their purchase. Either cross sell products or recently viewed items are a no-brainer here. Give your users the opportunity to add items of interest to their order, before it's too late.

Of course, it's important to make sure not to lead users away from the basket page. The last thing you want to do is steer someone away from completing their purchase.

Make sure recommendations open in a small pop up window which allows users to select their colour, size, and add to bag, without the need to navigate away from the page. Another option is to only recommend products which are one-size, or have no size, so they can click one button to add them to their cart.

Basket recommendations are a great way to increase average order value. If you offer free delivery over a certain price threshold, reference the current basket value and then recommend products within the price range needed to make the customer eligible for free delivery.

It's a win/win, the customer gets an item which they feel is almost 'free' and you meet your average order value targets.


Whatever strategic recommendations you choose to deploy, continued A/B testing is critical to ensure

a) the strategy you've chosen truly is working for you and your customer, and

b) you continue to ensure as your client base grows and trends change, your strategy is still up-to-date with your needs and demands.

There is no one method which works for everyone across the board as every site's product and customer are different, even if the contrasts are only small they can be the difference between success and failure.

Conduct regular usability testing to ensure your hypothesis has a solid basis. In other words, always remember to “check yourself before you wreck yourself", as a wise man called Ice Cube once said.

I'm sure he was most likely referring to product recommendations when he wrote that.


You may also like


Learn more about our Search Solution


Learn more about our Personalisation Solution


Learn more about our Merchandising Solution

Ready to see our platform in action?