Online personalised recommendations: 7 types and how to use them

As a child, do you remember being confused as to why your mum would always go to the same small number of shops when you were out shopping? As you’ve got older, you probably realised it was because each of those shops offered a highly personalised service. Maybe they greeted her (and you) by name and they would recommend products that they thought your mum would like.

With the growth of ecommerce – 693 billion GBP in revenue in 2019 and as many as 87% of UK households making online purchases in the 12 months up to 2020—that highly personalised service and attention to detail has disappeared. Right?


With advances in automation and AI, most online retailers now offer personalised product recommendations and a targeted service that recognises returning customers, their buying behaviours, and what they want and expect from an online shop. 

Read on to learn more about:

What are personalised product recommendations?

The benefits of personalisation in ecommerce

The top 7 types of product recommendations

Best practices in making successful recommendations

Measuring success in product recommendations

Features to look for in a recommendation solution

Start making personalised recommendations with PureClarity today!

Case studies

The takeaway

You also might like this


To view the Ebook please fill in the form below

What are personalised product recommendations?

The clue lies in the word personalised. This isn’t some algorithm throwing out product suggestions randomly. The suggestions are based on previous customer behaviour on that particular website. That means any purchase history, but also items that you may have viewed or that you may even have put in your shopping cart before, for whatever reason, abandoning the cart. 

A good example is shopping on Amazon. Standard SEO may bring you to a landing page, but whenever you view any product, from a book to a torch, you will always see a list of other products with the title; “Customers who bought this item also bought.”

In this case, the Amazon AI is identifying patterns of behaviour on its online store when it comes to buying a certain product. It could be as simple as customers who buy a hammer usually bought a box of nails at the same time. 

Those recommendations are based on general historical data on buying patterns. But Amazon will also offer you more personalised recommendations based on your own browsing and buying history. Put simply, what Amazon and other ecommerce businesses are doing is predicting the products you may like based on the data they’ve collected over your interactions with them.

Want to optimise your digital customer experience?

The benefits of personalisation in ecommerce

Decrease in cart abandonment

Cart abandonment is the bane of the lives of all ecommerce brands. Personalisation achieves two things that can help reduce those abandonment figures; better engagement and a better customer experience (CX). 

Using everything from recommendations to exit pop-ups, you can make a customer’s journey through your sales funnel smoother and make buying decisions easier.

Increase in the average order value (AOV)

If you’re providing your customers with specific information targeted directly at them, they’re more likely to spend more on each visit. For example, if they’re purchasing a laptop, giving them suggestions of related products (laptop bag, mouse, headphones), will make them more likely to buy some of those products at the same time.

Increase in online sales

By personalising your interaction with customers, you’re likely to see increased conversion rates. In fact, 86 percent of customers and 96 percent of retailers say that personalisation impacts purchasing decisions. Providing customers with information and products relevant to them makes them more likely to buy.

Increased time on-site

When you offer a great user experience and make customers feel more engaged with your brand, then they’re likely to spend more time browsing your site. And those factors also mean they’re more likely to come back to your site again (and again). Of course, the more time a customer spends on your site, the more likely they are to make a purchase. 

Increased competitive edge

As the ecommerce sector constantly grows, so does its competitiveness. Unless you’re in an extremely niche market, your marketers are likely to have hundreds of competitors fighting for the same customer base. By offering a highly personalised service, you appear friendly to customers and can gain a competitive edge.

Improved digital customer experience (DCX)

Ideally you want a long-term relationship with your customers that increases CLV (customer lifetime value). By personalising every aspect of their journey, from emails to targeted ads, you provide a better digital customer experience which in turn makes them more likely to establish a long relationship with your brand.

The difference between custom and AI driven product recommenders

Many ecommerce platforms offer integrated product recommenders. While these may be beneficial in the early days of switching strategies, they lack the benefits provided long-term by PureClarity’s AI driven recommenders

  1. Real-time: Standard recommenders are static and based on fixed rules, so may show duplicate content to customers. PureClarity’s AI driven recommenders monitor real time data including current shopping behaviour, current and previous searches, products viewed and bought (including items that may have been placed in their cart and later abandoned), and overall trends. 
  2. Placement: Standard recommenders limit placement and you can only have recommendations on your home page or on landing pages. PureClarity’s AI-driven system allows you to place recommendations at any touchpoint of your customer’s journey, from landing pages to emails. 
  3. Relevance: Standard recommenders may duplicate content and may also show recommendations not relevant to what the customer is browsing or searching for. PureClarity’s AI identifies the right products to recommend throughout a customer’s journey, avoids any duplication, and ensures they are relevant to that particular customer.

The top 7 types of product recommendations

  1. Homepage recommendations

While some visitors may go directly to a landing page via a search engine, an ad, or a social media post (or social proof), your homepage is still your main entry portal, especially for new customers. 

New customers present a challenge, as you have yet to gather any data on them. Tools such as PureClarity are equipped, however, to collect and analyse user data in real-time. Good practice here is to highlight your best sellers and most popular products as a starting point.

  1. On-site search recommendations

A good search option is a necessity for any online retailer. However, you can take it to the next level by incorporating intelligent recommendations to your search option, and also making it “typo friendly”, so it can recognise incorrect spellings and still suggest products. Your search bar is a great way to guide your customers through your funnel.

  1. Category recommendations

Your category pages can be a crucial driver when it comes to discovery. By listing all the relevant products within a category, you can then recommend products within the general category and subcategories, making it easier for customers to find a particular product they’re looking for and to discover other products.

  1. Product page recommendations

Recommendations on product pages offer the ideal opportunity to upsell and cross-sell. You can offer either alternative products (for example, a cheaper alternative) or you can offer complementary products (for example, a camera bag to go with that nice new DSLR).

  1. Checkout recommendations

One of your aims is to increase the AOV, and this can even be achieved at the end of the customer’s journey when they reach the checkout process. You can recommend complementary products or you can remind customers of recently viewed products in case they’ve slipped their mind.

  1. Email recommendations

Opportunities to recommend products don’t end once the customer has left (whether having made a purchase or not). Now that you have their details, including browsing and purchasing behaviour, you can leverage that info with templates of personalised and targeted emails that can encourage shoppers to return to your site.

  1. Order confirmation recommendations

This is perhaps the most under-utilised step in the customer journey, yet recommendations at this stage present a great opportunity to engage with the customer and to offer more products. Amazon has mastered this stage, offering both products based on what the customer has looked at and products based on a customer’s purchases.

Best practices in making successful recommendations

  1. Identify your target audience

There is little point in just randomly throwing product recommendations without knowing who you’re targeting. Just as you conducted research to identify the overall audience for your brand, so you should be identifying the different groups you will make recommendations to. 

Through identification and segmentation, you will be able to recommend more relevant products to your different customer groups.

  1. Continuously test your campaigns

A good campaign is not necessarily a great campaign, and even a great campaign may lose strength over time. You should be continually testing your strategies and establishing benchmarks to work from. 
Identify what works well and what produces disappointing results. If there seems to be any issues, then consider using A/B testing to highlight weaknesses and strengths.

Measuring success in product recommendations

Metrics and analytics are your road signs to success. They tell you what parts of your business work and what parts fail. What are the specific metrics that show your product recommendation engine is working efficiently?

  1. Browsing time

If a customer spends only a few minutes on your site before leaving, then they’re not engaging or not having a positive online shopping experience. As you expand your product recommendation strategy, you should be measuring the average browsing time. 

The time customers spend on your site, and the number of recommendations they click on, are good measures of your strategy working.

  1. Email open and click-through rates

Email marketing should be an integral part of your product recommendation strategy. There are two main metrics to consider here. The most important is your CTR (click-through rate) as that will tell you how many people are actually following any links contained in the email. 

How many people just open an email is less important, but can be used as a comparative metric to see if tweaking subject lines means more people read the mail.

  1. On-site click rates

Actual clicks on recommendations on your site help measure the level of interaction with customers. Clicks in Recommendations (CIR) can help gauge how effective your recommendation strategy is. Measure CIR regularly so you can see if there are issues with particular groups of recommendations.

  1. Increase in AOV

It’s natural to want your customers to spend more on each visit. If you see an increase in average order value (AOV) after implementing or changing your product recommendation strategy, then you know the strategy is working. Especially if both AOV and CIR have risen by a significant amount.

  1. Increase in AOV

Increased conversions is always a good thing, as it means greater revenue. Ideally, you want to be able to measure the orders that contain one or more items that came from recommendations made during the customer journey. Seeing what proportion of your total sales contained recommended items is a great indicator of how well your recommendation strategy is working.

Features to look for in a recommendation solution

So, you currently don’t utilise recommendations in your ecommerce store. Now you want to integrate a recommendation solution to increase sales and conversions. What features should you be looking for?

  1. Customer segmentation

Not all customers are the same, and to completely personalise for each individual could be costly and time-consuming. By segmenting your customer base into different groups based on factors such as location or types of products they show interest in means they’re seeing content more relevant to them and more likely to lead to a possible sale.

  1. A/B testing

Trends change and people change, so what works in terms of content, images, and other aspects of your product recommendation strategy may also change over time. Utilising a solution that is continuously A/B testing on all elements of your campaign means that you can make adjustments to improve engagement.

  1. Real-time recommendations

Ideally, you want to provide real-time recommendations to your customers. There’s little point in recommending a product based on something they bought a year ago. So you want your solution to be tracking recent and current behaviours, and making recommendations based on that data. 

  1. Upselling and cross-selling

One of your primary aims with personalised and targeted recommendations is to increase AOV and revenue. The best way to achieve this is through upselling and cross-selling. 
Upselling may often be combined with an offer to tempt the customer to spend more (for example, a more expensive pair of trainers with free shipping). Cross-selling can offer related products (for example, offering peripherals when someone buys a laptop).

  1. Multi-channel personalisation

Modern ecommerce stores usually take an omnichannel approach, so you want your personalised recommendations to be the same, or similar, across every channel you utilise. Look for a solution that engages with customers across every channel they may use, making them feel like individuals rather than just another consumer.

Start making personalised recommendations with PureClarity today!

Why should you consider PureClarity as your solution when it comes to integrating personalised recommendations on your ecommerce site? Companies like Amazon may have blazed the initial trail but it’s solutions like PureClarity that make intelligent personalisation accessible for all businesses.

AI-driven personalisation

PureClarity’s AI-driven recommendation solution combines the data collected from customers’ historical behaviour and their current behaviour so that it can choose the most relevant products to show them. The AI tracks what recommendations it shows, as well as where and when it shows them, so that the customer is not shown any duplicate content. 

Main features and functions of the PureClarity AI recommender include:

  • The AI engine personalises the best recommendation strategy for every visitor as an individual. 
  • The PureClarity Cold Start feature can work even with sparse data sets.
  • No manual intervention needed, and your PureClarity system can work 24/7. 
  • The AI can identify if you have added any manual recommenders so that it doesn’t duplicate any offers. 
  • To achieve maximum conversion, the AI engine places recommendations in the best position on any relevant pages. 
  • Whether you operate a B2B business or B2C, the engine is optimised for both so it can implement different recommendation strategies according to your customers’ profiles. 
  • As the system is fully customisable, you can have recommendations placed anywhere on your site. 

Intelligent product recommenders
Why should you consider PureClarity as your solution when it comes to integrating personalised recommendations on your ecommerce site? Companies like Amazon may have blazed the initial trail but it’s solutions like PureClarity that make intelligent personalisation accessible for all businesses.

PureClarity’s AI and advanced machine learning ensures that visitors to your site see any and all recommendations at the right time and in the right place. 

  • Custom product recommenders. You may have a particular brand or line of products you want to highlight. PureClarity lets you customise and design your own recommendations so that customers see those products. 
  • Ensure that different sections are working efficiently by A/B testing the various recommenders on your site. 
  • Monitor performance with real time analytics that give you a real insight into how things are working. You can customise what metrics you measure, including CTR and conversion rates. 
  • Place your recommenders in any category, sub-category, or specific landing page where you think they will have the most effect.

Integrations for content

You can let the PureClarity AI engine decide where it thinks is the best place to place content, you can add manually, or you can use a combination of the two. Given you will be likely using an omnichannel approach, you want those recommendations to appear at every possible touchpoint. These can include:

  • Carousel banner. A set of rotating banner ads (like a slideshow) that appears on your home page. You can show up to five images and accompanying text displaying relevant recommendations. 
  • Search bar. Having an intelligent search bar that both recognises any errors and also suggests relevant products is an ideal tool to guide your customers to different products. 
  • Live chat. Even during a live chat with a customer, our real time AI can present recommendations relevant to the customer’s behaviour and needs. 
  • Email. With easy integration with most email providers, you can offer recommendations in any email you send to subscribers. You can also track important metrics such as click-throughs, so that you can gauge effectiveness of campaigns and adjust in future if needed.

Case studies

Comfy Homes AI-driven product recommenders
Comfy Homes AI-driven product recommenders

Comfy Homes

With their business increasingly moving online, Comfy Homes needed to optimise their online presence. One of their main priorities was to replicate the service they had offered in their physical stores. 

Working with PureClarity ensured they were able to achieve that, and that any visitor to their website had a personalised journey, only seeing products or promotions relevant to them. 


Within 60 days of starting to use PureClarity, Comfy Homes saw a 5% increase in their AOV and 20.8% of all orders were due to PureClarity.

Harry Hall

Harry Hall is one of the UK’s leading suppliers of equestrian products. They already had an established online presence and knew the importance of a personalised experience. However, before switching to PureClarity, they were working with another solution provider and were dissatisfied with the results. 

When they looked at PureClarity, they found the pricing was far more competitive and also liked the access to the PureClarity team for support and advice in optimising their personalisation strategy. 


Since switching to PureClarity, Harry Hall has seen a 10% increase in their overall conversions 

And a 17.5% increase in revenue purely driven by PureClarity product recommenders.

Comfy Homes AI-driven product recommenders

The takeaway

Online shoppers still expect the high level of personal service they experienced in physical stores. While that may sound daunting at first, advancements in technology means that it is—relatively—easy to achieve. With so much competitiveness in ecommerce, optimising that personal experience is crucial to keeping an edge over your competitors. 

Although many ecommerce platforms offer basic recommenders, they have severe limitations that could reduce the quality of the customer experience rather than enhancing it. With PureClarity, you get a product recommendation service that does enhance that experience while boosting your conversion rates and revenue. 

Why not check us out on social media, book a demo, or try our 30-day free trial today?

Instagram, Facebook, LinkedIn, Twitter, and YouTube

Did someone say increased conversion?

Start a no-commitment 30-day free trial and enjoy all PureClarity features on your website.

Increase conversion