![Fisher#$%&’()*+,--./012 · 书!" Fisher#$%&’()*+,--./012———!"#$%&’() *+,1，- .1，/001，1232 （1．!"#$%&’()$*，+,!" 450001；2．+,-./012，+,!" 450002） 34 [45]6789](https://static.fdocuments.net/doc/165x107/5f6e2d4ba9949b038256b3e4/fishera-fishera-012aaaa.jpg)
01,1#2*3'45#2 !#$%&'()*'&'+,')#'$)-'.'$)./)012 6/*-%'()* Created Date: 3/12/2018 3:46:48 PM ...
2
Car-sharing Industry Leader, Zipcar, improves Revenue per Member by 15% by using Relativity6 Predictive Engine Case Study The Results Improved Revenue per Member by 15% Predicted lapsed customers with a propensity to purchase with 84% accuracy Predicted which specific vehicle would be booked with 70% accuracy The Challenge Zipcar is a car sharing service that offers rental vehicles to customers by the hour, day or week. Relativity6 worked with Zipcar's analytics team to determine which inactive customers had the highest propensity to make their next booking. Relativity6 was also tasked with predicting which specific make and model each Zipcar customer was most likely to book. The Solution Relativity6 fused our own external data sources with transaction data from all of Zipcar's long lapsed customers. Using Relativity6’s proprietary machine learning algorithms and semantic engine, we identified the segment of customers with the highest propensity to book a ride within the month of August 2017. Zipcar launched a highly targeted campaign to this cohort of users, as well a cohort of users from a control group. The segment identified by Relativity6 performed 15% better than the control group on revenue per member and was over 85% accurate in classifying lapsed customers with a propensity to make a booking. Additionally, our product recommendation engine was above 70% accurate. Technologies used: Semantic Engine Natural Language Processing Machine Learning Open-Source Intelligence Data Fusion Behavioral Listening
Transcript of 01,1#2*3'45#2 !#$%&'()*'&'+,')#'$)-'.'$)./)012 6/*-%'()* Created Date: 3/12/2018 3:46:48 PM ...
Car-sharing Industry Leader, Zipcar,improves Revenue per Member by 15%by using Relativity6 Predictive Engine
Case Study
The Results
Improved Revenue per Member by 15% Predicted lapsed customers with a propensity to purchase with 84% accuracy Predicted which specific vehicle would be booked with 70% accuracy
The ChallengeZipcar is a car sharing service that offers rental vehicles to customers by the hour, day or week. Relativity6 worked with Zipcar's analytics team to determine which inactive customers had the highest propensity to make their next booking.
Relativity6 was also tasked with predicting which specific make and model each Zipcar customer was most likely to book.
The SolutionRelativity6 fused our own external data sources with transaction data from all of Zipcar's long lapsed customers. Using Relativity6’s proprietary machine learning algorithms and semantic engine, we identified the segment of customers with the highest propensity to book a ride within the month of August 2017. Zipcar launched a highly targeted campaign to this cohort of users, as well a cohort of users from a control group.
The segment identified by Relativity6 performed 15% better than the control group on revenue per member and was over 85% accurate in classifying lapsed customers with a propensity to make a booking.
Additionally, our product recommendation engine was above 70% accurate.
Technologies used:Semantic Engine Natural Language ProcessingMachine LearningOpen-Source IntelligenceData FusionBehavioral Listening
About Relativity6
Companies use Relativity6’s Technology to improvecustomer engagement and to arm sales, marketing and service professionals with the tools needed to better serve their customers.
Relativity6 enables partners to better understand their customers’ behavior with support from world-class PhD data scientists and machine learning experts from MIT.
Relativity6 employs compliance and technical processes to meet or exceed data security and privacy regulations across industries and across borders.
Relativity6 proprietary algorithms predict when and what your most profitable customers will purchase, with above 80% accuracy.
Contact
For more information, please contact us.
![012&0#)3#!,3'45)+3,# 7*+.#83'90#8+99#:&'+.* · 10.07.2013 · Oxford Left Review 59 SDUHQF\ DV D V\VWHPLF FRHUFLRQ WKDW VHL]HV DQG WUDQVIRUPV DOO VRFLHWDO SURFHVVHV As a force that](https://static.fdocuments.net/doc/165x107/5e0d3513e02dd915042a04a3/012033453-78390899-10072013-oxford-left-review.jpg)
