Developed a customer segmentation model leveraging Facebook Open Graph
data for a client with millions of facebook followers. The model helped our client
identify the profile of Facebook members that would become the most productive customers.
Through this segmentation, the client was able to increase ROI on paid acquisitions bids
by over 250%.
Developed a customized optimization algorithm requiring sub-second response times for a major entertainment company to minimize wait times. The algorithm evaluated multiple competing objectives and considered millions of combinations of product offerings to select the best ones to present to consumers.
Enhanced an optimization algorithm to combine products directly mailed to millions
of users from multiple suppliers into postal bundles that significantly reduce overall postage costs
by taking advantage of bundled discounts based on firm address, carrier route or zipcode.
Utilized multiple user attributes (browsing history, usage patterns, geographical location, etc) to segment customers in buckets. Created similarity matrix to group similar users together. Tool delivers customized online content to a site user based on the user attributes.
Click Through Rate increased 20% with a content recommender system that delivers customized content to members of an online travel community by taking into account their browsing patterns and other user attributes. Utilized A/B statistical testing to continuously improve quality of personalization.
Offer recommender system utilizes a variation of the Gale-Shapley Nobel prize winning algorithm to present the most appropriate offer to millions of consumers while ensuring broad exposure to thousands of offers. This algorithm could be extended to a multitude of uses including online dating, job searching, news content, etc.
Analyzed profiles of current customers to determine profile of potential
customers that are most likely to convert. Numerous attributes including business size,
location, unique characteristics, etc. Developed Propensity to Convert score that was used
to prioritize sales effort.
Measured lifecycle of members and how much they engaged with various types of emails
over their lifecycle. Identified thresholds for email volume that drove significantly higher opt out
levels which was used to set contact rule guidelines.
Created and deployed a dynamic price optimization system by analyzing customer conversion data
and combined with near real-time competitive pricing information.
Increased exposure to client's brand by ensuring press release content is geared to targeted audience.
Generate insights for numerous ad-hoc projects such as: top trending pages, pages with the largest growth share, user
growth, top voted pages, etc.
© 2018 Mazliah Analytics. All rights reserved.
