-
for those who care about e-commerce
Recommendation Engine
Leverage the power of your data and boost sales with RENG. Enhance customer experience and create engagement in your e-commerce website.
INTRO
Aim of RENG is to recommend your online customers best fit products, which they are likely to click, learn more about and buy.
RENG predicts users’ interests and recommends items very likely matching their needs.
Did you know?
35% of purchases on Amazon come from product recommendations.
75% of Netflix views come from product recommendations.
Exploit product recommendations with RENG, extend your customers' shopping basket, and ultimately increase your sales profit.
OVERVIEW
RENG has been primarily designed for medium to large e-shops with millions of product views per month. It was built as the horizontally scalable data-mining service with no centralized component and no single point of failure. Regardless of the amount of purchased items or portfolio of products, the performance of the RENG scales linearly with the size of the cluster.
RENG gets you covered no matter how big your data is.
RENG is highly optimized with compact and efficient data structures. Fully distributed computations are happening in real-time over the raw data. There is no need to train the model or run some pre- computations in batch.
Need hundreds of recommendations delivered per second? Need less than 30ms per request?
RENG speaks your language.
RENG provides a simple REST API which makes the request for recommended items for product 123 as easy as:
curl -X GET -H "Authorization: Bearer {{OAUTH_TOKEN}}"
https://reng-lb/v1/grp/123?rpCount=5
RENG integrates with any e-commerce solution with minimal amount of implementation effort.
RENG can be deployed on-premise or provided as the managed cloud-native service. We support primarily AWS, however we can adapt to customer needs if another cloud provider is required. We support and endorse Kubernetes as well. We do not charge any additional fee for hosted solution apart from the cost of consumed resources by RENG cluster.
Interested in more technical details? Feel free to contact us.
REFERENCES
Auto Kelly a.s. is the largest distributor and supplier of replacement parts for vehicles in Czech Republic. Auto Kelly a.s. is wholly owned subsidiary of LKQ Corporation - a leading provider (Fortune 500) of alternative and specialty parts to repair and accessorize automobiles and other vehicles.
RENG provides product recommendations to Auto Kelly’s e-commerce website in Czech Republic since 07/2018. RENG produces around:
3 000 000 recommendations per month
across 300 000 different products
based on data from 13 000 000 individual purchases
while 95% of requests are served within 40ms
Eager to see it in practise? Just visit autokelly.cz and start browsing. Any detailed product view includes also a list of recommended products.
What our client says
Using RENG is very easy, fast and simple. We've significantly increased the number of conversions compared to manually setting recommended products since launching RENG algorithm.
Stanislav Ivanega (Head of IT Development)
PRICING
RENG comes with three plans based on the size and usage of your e-commerce web site. For every option we provide period of first 90 days free of charge.
Standard
1199 €/mo
Premium
1499 €/mo
Ultimate
1899 €/mo
All packages include
Real-time monitoring
On-prem / Cloud / Kubernetes
Upgrade without loss of availability
Unlimited or custom solution
AUTHOR
RENG has been designed and developed by Michal Zerola. Michal holds a Ph.D. in computer science and has spent more than a decade building software solutions in both world class academical environment as well as industry leading companies. Having solid fundamentals and experience from both worlds, he believes in simplicity and perfection, also known as “elegance” among mathematicians. These are the core principles he has embedded into RENG as well.
Michal Zerola
Domousnicka 649/8, 197 00 Prague, Czech Republic. Registered in Trade Licensing Register at Mestsky urad Praha 19 since 10/2018 under the identification number 07454546.