“Neu.ro PaaS decreased our NLP development time by >50%.”
THE OPPORTUNITY
Current State of CRM:1
Revenue Grid is at the forefront of CRM market transformation to AI-driven sales solutions powered by predictive analytics. They call this approach Guided Selling.
In the past, CRM software companies focused on streamlining the sales process and maintaining better, easier-to-use records of contacts, customer interactions and sales team performance.
Today, deep learning is driving an AI Transformation of CRM. Algorithms are now able to derive intelligence from vast amounts of data that are created over the lifetime of customer interactions. What was primarily a passive tool for scheduling and record keeping is becoming a productivity multiplier for sales teams.
“By 2020, 30% of all B2B companies will employ some kind of AI to augment at least one of their primary sales processes.”-Gartner Research
THE CHALLENGE
As an AI-focused company, Revenue Grid has always relied on in-house resources and personnel to develop their AI solutions and to manage the infrastructure, resources and toolsets required across the entire ML lifecycle.
Previously, Revenue Grid had committed to the Azure platform for hosting their code base and serving their SaaS products and had fully integrated their product with Azure’s many service offerings.
But by mid-2020, Revenue Grid found themselves developing multiple new ML products simultaneously, stretching the capacity of their internal team. As these projects scaled, the company recognized that their aggressive GTM plan could face significant delays.
To de-risk delivery, Revenue Grid sought specialist outside resources to develop and deploy a new ML feature – specifically an ML-automated system for analyzing large volumes of emails, extracting target features, and providing insights based on the results.
Revenue Grid had several important parameters for evaluating vendors, including:
- A seamless collaboration of internal and vendor teams
- Data security
- Co-location of data with existing Azure GPU cloud resources
After a thorough evaluation based on these parameters, Revenue Grid turned to Neu.ro.
Revenue Grid’s existing secure Azure environment
Developed and implemented Neu.ro NLP+ solution for Automated Lead Generation, including Text Clusterization and Sentiment Analysis - on Neu.ro MLOps Platform running natively within Azure.
Implementing robust MLOps and ML tool interoperability via Neu.ro resulted in >50% reduction in Revenue Grid’s AI development time
Developed and deployed entirely within the client’s secure environment in Azure cloud
>50% reduction in the client’s AI development time
THE SOLUTION
Neu.ro began the process by installing the Neu.ro MLOps Platform directly into Revenue Grid’s development environment on Azure Cloud. Within days, the two teams were able to seamlessly collaborate, securely sharing all project assets and co-locating training data with Revenue Grid’s Azure-based GPU computational resources. This allowed the teams to protect sensitive data without circumventing the company’s existing security measures.
The pipeline included developing a custom automated data-labeling system based on the latest state-of-the-art research in deep learning that allowed for the identification and categorization of target features within email threads. Neu.ro used a state-of-the-art NLP algorithm that was subsequently trained, tuned, and tested with the data extracted. Finally, Neu.ro deployed the model within the Revenue Grid environment, where it continues to learn on the company’s real-time inference data.
Neu.ro further provided a customized model based on the latest state-of-the-art research in deep learning that allowed Revenue Grid to identify and categorize target features within email threads and automatically label data so that the system can continuously learn based on real-time inference data.
In 45 days, Neu.ro was able to:
- Develop and deploy entirely within the client’s secure Azure cloud environment
- Provide a customized model based on the latest state-of-the-art research in deep learning
- Reduce Revenue Grid’s development time by over 50%