Jul 20, 2018

posted by Mark Milankovich

How a Product Recommendation Engine Works

How product recommendation engines work Yuspify blog

Are you considering displaying personalized product offers real-time and suggesting related, relevant and complementary products? Forget the worries of looming astronomical expenditures would come with hiring product recommendation engine developers! Licensing or building a recommendation engine was the privilege of monstrous e-stores 15 years ago. But nowadays it’s very affordable for smaller eCommerce stores.

offline product recommendation engine

Implementing Personalized Recommendations serving every single customer in real time would be very expensive for offline retail units.

Online shops with over 20-25,000 page views per month cannot avoid deploying recommendation engines any more. After adjusted and configured, these product recommendation engine algorithms helped by artificial intelligence will significantly boost CTR! Accordingly, they will increase average order value, revenue, conversion, and other important metrics even in the first few weeks.

Engines like Yuspify employ clusters of product recommendation algorithms that analyze the events made by each unique visitor. To serve the most up-to-date offers to the store’s visitor, the tracking system works in real time. The engine is ready to refresh the offers in any second if it detects any change in the customer’s behavior.

E-stores using Yuspify can realize an average 10 % revenue through recommendations. The following factors can influence this result:

-Number of recommendations,

-General page design (Sometimes it can impede the widgets’ successful integration and performance)

-Fixed placement of recommendation widgets that you can not modify. (Because you can’t access the code or the store has strictly determined the placements.)

-Extraordinary traffic in campaign periods.

Product recommendation engine - cora romania

Cora.ro french hypermarket chain having 11 stores in Romania and 1 million page views / month saw 10% more visitors and 7% more revenue after Yusp recommendation engines had been integrated.

Shopping user experience also can see a positive effect which ends up in better retention ratio and higher customer satisfaction. Needless to say, what happens if the store fulfills the customers’ outspoken, hidden or subliminal demands. They will be definitely more willing to make completed purchases.

The published figures about Netflix’s engine shed the spotlight to the added value of the recommender systems: the engine belonging to this video giant produces a billion dollar plus revenue per year. This monstrous engine sorts more than 250 million subscribers to almost 2000 taste groups. Because of the abundance of tags and attributes the algorithms are able to author even very sophisticated taste groups: ‘Users that prefer movies about vigilant cops in a bible-belt environment.’

According to a VentureBeat-published study 77% of the digital natives cannot imagine their online existence without personalization. They certainly would like to see personalized experiences, products, digital contents or social contacts around them.

Amazon’s often-cited statistics says 35% of their total revenues arrive through those products that their customers found via recommendations.

Product Recommendation Engine metrics

The revenue generated through recommendations is the most crucial metric when evaluating a recommender system. The industry standard measurement counts ‘revenue through a recommendation’ when the recommended product is clicked and bought within 24 hours. A product recommendation engine dashboard should handle at least the following important metrics:

% of Revenue Through Recommendations – These are cornerstone metrics. The quotient of the revenue generated through recommendations / the total revenue.

Gyogyexpressz Yuspify analytics

Increased recommendation CTR on gyogyexpressz.com — Yuspify dashboard. After 9 months CTR (click-through rate) has gradually reached the 250% performance increase in November, delivering an 11.5% annual average click-through rate which made up an average 12,7 % of the monthly incomes.

CRR ( Conversion Rate from Recommendations) – The Conversion Rate of the customers who clicked on recommendations. Juxtapose this to the general conversion rate and to the CRnR ( Conversion Rate from non-Recommendations)

GMV/1000 Recommendations – The average revenue sent by 1000 recommendations. It arrives from those customers that bought products via recommender containers.

Number of Recommendation Clicks – The number of products clicked by those visitors who actively used recommendations while they were roaming in the store. You can subdivide this total to sub-reports like ‘number of products viewed on XYZ category page’ or on ‘ABC product page’ etc.

CTR – the commonly used ratio: The number of users clicking on the recommendation widgets compared to the total number of customers viewing the page where the same widget shows up.

The Difference Between Recommendations and Personalization

People even in the e-commerce sector often use the terms of “recommendation” and “personalization” interchangeably. Personalization is a broad category within the website optimization and e-stores also apply them in the practice of recommendations.

Product Recommendations represent 2 subsets:

-Personalized Recommendations. The engine utilizes the user profile.

-Non-personalized Recommendations (by utilizing the data mass of item attributes and other purchases).

Using personalization in recommendations is not the obvious choice every time, in certain cases, item-based recommendations entail more conversion. Well-streamlined recommendation engines launch fallback scenarios: ‘Does the customer have enough history to get personalized offers that might imply a higher probability of conversion? Alternatively, shall we apply item-only recommendations without personalizing it to the user?’

book recommendations

inspired by browsing history

Personalization cannot be done only with recommendation engines, but a 100% personalized site is always supported and powered by recommendation engines.

Summary

An affordable and efficient product recommendation engine solution helps e-Commerce sites to boost sales and to blow up conversion rates. Personalized recommendations have spread across multiple e-Commerce website types and channels to support online sellers. It helps stakeholders to understand their customers’ prevailing intentions and preferences. Consequently, the engine will display the most appropriate products in real time. It’s important to highlight that a decent recommender system’s presentation layer flows seamlessly into the store design. Moreover, it integrates the recommendation widgets to make them look like an organic part of the store.

Recommendation engine providers often open free trial periods ranging 7 to 30 days. This trial period is usually enough to win over the cold start problem.  As a result, you can start the accumulation of recommendation-driven revenue even after a few weeks as this case study shows. Before contracting a solution provider you are eligible to get a cent-perfect breakdown of your costs in the proportion of served recommendations – or, if you choose Yuspify you need to pay only a small % based on the revenue generated by recommendations.


Ready to get started? Yuspify offers a 30-day free trial. Check it out yourself, it’s on us!

Sign up for email updates

Get regular tips and updates from the world of eCommerce and big data.

We take your privacy seriously. No spam. See our Privacy policy here.

Recent Blogposts