Blanc Labs Implements Data Science Solution for Nekso Taxi Platform

Nekso Logo

About Nekso


Canadian startup Nekso is a taxi hailing platform that integrates passengers, drivers and established taxi companies into a mobility ecosystem based on the smartphone. The service provides a competitive alternative to Uber for taxi companies and their drivers, and offers better safety and security valued by passengers.





22 cities

4 Latin American countries





Blanc Labs Implements Data Science Solution for Nekso Taxi Platform


An Intelligent Transportation System, Delivers Insights and Business Growth

The Challenge — Making Sense of Data


As Nekso’s operations rapidly grew, so did the ever-increasing amount of data they were collecting. They understood this data could give them a stronger decision making foundation to maximize customer benefits, manage KPIs and fuel service growth. But they were struggling to derive truly actionable insights from the vast amount of data they were presented with daily.


As Ramón Garcia, CMO of Nekso, described, “Our goal was to bring all of our information together and then make sense of it – and that is actually a very challenging problem.”


The Solution — Creating a Scalable Data Architecture


Nekso needed data intelligence expertise and sought the help of Blanc Labs as their technology partner. “We decided to go to Blanc Labs to help us with our data intelligence. We were impressed with their development acumen, their data science expertise and their understanding of our business,” noted Garcia.


The Blanc Labs Data Intelligence team knew the first step to tackle was a data architecture system that could integrate data onto one platform. This integrated system would help Nekso derive robust insights while offering agility and speed as they scaled operations.  It would also improve the service and effectiveness of internal IT operations.


A few key objectives were established for the data architecture:

  • Capable of ingesting data from multiple sources
  • Designed for scalability and performance
  • Be secure, traceable, and auditable
  • Able to support rapid development and deployment of new applications
  • Able to support real-time data analysis with automated feeds to a decision-making system


Building a Data Lake


The core of their solution is data lake built on Hadoop Distributed File System (HDFS) hosted on the Amazon Web Services (AWS) cloud platform (Diagram 1). Data is ingested into this lake from internal taxi operations, as well as external sources impacting their service such as: local events, social media, weather records, and traffic patterns. The data streams come in several formats: structured data (ride request history), unstructured data (social, text, etc.), and semi-structured data (logs). Data is transformed and analyzed then presented in a customized Nekso dashboard. The data pipelines were designed to be flexible so that new transformations can be triggered seamlessly on an on-demand basis.


Nekso Data Architecture

Diagram 1: Nekso’s Data Architecture


Visualizing Business Intelligence Results


A visualization and reporting dashboard was set up providing Nekso management with rapid awareness of service performance and KPI monitoring to drive business improvements.

“A month into the Blanc Labs implementation of our new data intelligence solution, things changed for us. Data started to improve our strategy meetings and accelerate our development with more informed decision making. Meetings now start with more fulsome target KPIs presented on the screen, then we dig in.”
— Azam Mohabbatian, VP Strategy

Internal users now have access to data and tools that fit their various job requirements and expertise, they can independently explore data, perform ad-hoc queries, and compose their own reports.


The dashboard gives deeper and richer management information such as marketing campaign performance, customer information about why passengers cancel rides, and visualization of user growth patterns.


Predictive Analytics


The Blanc Labs team implemented advanced predictive analytics to forecast trends for management decisions and actions. Predictive analytics is now being used to make the Nekso service smarter:


  • Forecasting and planning: Identifying temporal patterns in passenger ride requests to forecast ride demand. Nekso’s Business Development team can better manage and expand their driver base to increase revenue in the best new markets.
  • Operational efficiency: Tracking ride data to forecast and more efficiently position the driver base to meet demand for rides. They can also more easily identify reasons behind “out of the ordinary” changes in ride behaviour and take proactive measures to maintain and improve customer satisfaction.
  • Optimizing marketing campaigns: Advanced cluster algorithms are utilized to personalize marketing promotions to customize rider segments that can be tested and tracked more effectively. This has helped Nekso expand their customer base, improve the ROI of campaigns and grow revenue.


Intelligent Transportation System


Nekso uses both advanced machine learning techniques and predictive analytics to enable recommendation models as part of an intelligent transportation system. The system learns from customer behaviour, traffic patterns, and social events to enable smarter decisions about scheduling and dispatching drivers (Diagram 2.) Apache Spark was chosen to power the recommendation engine.


Diagram 2: Real-Time Recommendation System for Driver Location

Diagram 2: Real-Time Recommendation System for Driver Location


Being able to intelligently predict the level of demand for rides within a geographic location and calculate the optimal route for drivers improves overall efficiency for drivers and taxi companies, and improves the customer experience.


The predictive algorithms provide recommendation models based on explicit, implicit, and historical data. Variables such as driver location, time-of-day, season, weather, customer input, and ride request history are integrated into the analysis providing personalized, context-aware recommendations for where a driver should go.


Nekso found the model could accurately predict areas of high demand 93% of the time. By advising drivers to move to high demand areas, or where demand may arise, passenger wait time went down significantly.


Other recommendation models include predicting when to send passengers a push notification asking if they will need a taxi and predicting their preferences, such as favourite make and model of car. These capabilities are expected to grow over time. Real-time recommendation models like this can offer high value and can be applied across many business situations and industries.


Looking Ahead


Blanc Labs helped Nekso aggregate their information, dramatically enrich it, and gain strategic and actionable business insights for improvement. The results directly impact their revenue growth and improve user satisfaction of passengers, drivers and taxi companies. Nekso will continue to partner with Blanc Labs to refine its data intelligence solution to enhance its competitiveness as a viable user alternative to Uber.  “Blanc Labs helped us augment our IT operations with specialized data science expertise. We will continue to partner with them to leverage value from data to accelerate our growth,” said Garcia.


We have entered a new phase of the digital economy where organizations must analyze and use vast amounts of diverse data as a competitive imperative. Early adopter organizations are benefiting from improved business decision-making and operational efficiencies, as well as better customer experiences. The data architecture outlined here, while created for the specific needs of Nekso, can address the big data and analytic requirements of many organizations.


We would like to partner with you


Contact us at to discuss the data intelligence or digital transformation needs of your organization. Our talented data scientists, architects and software engineers can analyze your business needs and deliver solutions and value from your data to help you compete in the digital economy.