Recommender System and Bidding Algorithm for Marketing Agency

Machine Learning Platform Helps Retailers, Brands Deliver Personalized Shopping Experience For Increased Conversion Rates
DOWNLOAD THE FULL CASE STUDY

ABOUT THE CLIENT

FusePump is a London-based digital marketing agency providing data-based solutions for retailers and brands.
The company connects customers with real-time prices and easy purchasing through its BuyNow button, and partners with clients to develop custom solutions for unique business challenges, such as SkyePlatform, which offers personalized content for retail websites. The algorithms created by Rediscovery.io are at the core of SkyePlatform.

PROJECTS HIGHLIGHTS

  • DELIVERABLES:
    • Recommendation engine algorithm
    • Bidding algorithm
  • TIMELINE:
    • 2 weeks: Proof of Concept
    • 4 months: Working model
    • 6 months: Final models and test deployment
  • TECH USED:
    • Python, MongoDB
  • MACHINE LEARNING TECHNIQUES:
    • Collaborative Filtering

Intro

Digital marketing company FusePump needed its own recommendation engine to satisfy a big client, but lacked the internal capability to build machine learning algorithms that learn customer preferences and recommend relevant products.

Executives turned to Rediscovery.io, as a company specializing in developing bespoke machine learning solutions for business and academia.

Rediscovery.io built the core Machine Learning engine behind our product recommendation system for a major client, and did brilliant work in designing the bidding algorithm’, said FusePump Product and Project Manager Dwayne Manzano. ‘I highly recommend Rediscovery.io

The challenge

Build a Recommendation Engine For Online Retailers to Deliver Personalized Product Suggestions and Increase Conversion Rates

FusePump asked Rediscovery.io to build a product recommendation engine able to:

  • Analyse user preferences and recommend relevant produces based on online activity.
  • Use anonymised data to not compromise customer privacy.
  • Continuously learn from new data and new customers.

FusePump also requested the product recommendations be at the core of a bidding platform that would allow brands and other retailers to compete for ad space.

The Decision-Making Process

Proof of Concept
An important milestone was the early delivery of a proof of concept, a simple prototype with limited functionality that demonstrated that the algorithm works and helped non-technical employees understand how machine learning translated logic into code.

Development and Iteration
Throughout the course of the project, Rediscovery.io collaborated with FusePump’s internal development, marketing, design and consulting teams to ensure the product aligned with the clients needs and was well integrated with FusePump’s existing infrastructure.

The Solution

RECOMMENDATION ENGINE WITH COMPETITIVE BIDDING

Algorithm 1: Personalized Product Suggestions
kyePlatform’s core functionality is powered by a collaborative filtering algorithm that analyzes user preferences to suggest relevant products to online customers. To protect customer privacy, data is stripped of all identifying information and stored in a secure database. Basic workflow:

  1. The algorithm analyzes a customer’s purchase history and estimates how similar items are to each other.
  2. A customer’s historical purchases are matched with similar products and assembled into a list of product recommendations.
  3. When a customer visits the retailer website, the top ranking item is shown as an ad.

Algorithm 2: Competitive Bidding for Ad Space
Bids are weighted by the bidding algorithm, which builds a score base on multiple factors:

  • Likelihood of customers to purchase the product.
  • Net margin of the retailer on the product.
  • Size of the bid.

Results

The algorithm created by Rediscovery.io sit at the core of SkyePlaftorm, allowing retailers and brands to deliver a more personalized experience for customers and in turn increasing revenue and customer satisfaction:

• Estimated 47% increase in conversion rates as compared with static banner ads.

• Shoppers have a more engaging experience due personalized advertisements and products.

• Retailers maximize the value of their ad space and generate new revenue.

• Brands have a unique opportunity to influence shoppers with a high intent to buy. This allows brands to increase sales and reduce their advertising budget.

• More efficient use of advertising space for retailers.

What are you waiting for?

If you want to leverage Artificial Intelligence, get in touch and see how we can help.
CONTACT US

Some of our case studies