[dropcap size=”2″]RN[/dropcap]Ranjit Nair: The key tenets of our approach are that analytics needs to be actionable and timely. This means that analytics needs to be tied to some business outcome so that it is useful and that it is customized for the industry or organization. Also, insights should be delivered in time. This means that if an organization is not ready for real-time decision making there is no need for the insights to be delivered in real time, but if the decision is one where acting in time is critical there is no excuse for the insights to not be available in a timely manner.
We believe that analytics needs to be democratised so that you don’t need to be a data scientist to be able to interpret the analytics we present. We are committed to ensuring that our analytics and the data we process is public and cannot be used to harm anyone.
AIM: What is your approach to face the challenge of meeting the needs of so many clients across vast geographies with limited resources?
RN: Our approach is to focus on a fixed set of sectors and particular pain points. This allows us to not spread ourselves too thin and yet be very specialised and useful to our clients. We also believe in automating a lot of low level work so that the amount of manual intervention for tasks like data collecting, curation and analysis is reduced.
AIM: What are the key differentiators in your analytical solutions?
RN: We are focused primarily on analysing stakeholder actions in social media, emails, chats and surveys as they relate to business decisions. Our key differentiators lie in the strength of our proprietary NLP algorithms and the actionability of our insights.
AIM: Please brief us about the size of your analytics division and what is hierarchal alignment, both depth and breadth.
RN: Our engineering team is about 35 people and analytics is about 25. Our hiring is currently in our analytics team.
Our analytics team is divided into sector based teams for sectors like Pharma, Media, BFSI and CPG. Within each team we have people responsible for data curation, subject matter expertise and data science.
AIM: What are the next steps/ road ahead for analytics at your organizations?
RN: We are currently focused on adding more people with both subject matter expertise as well as data science in our analytics team. We expect our analytics team to be at least 100 people by next March.
AIM: What are a few things that organizations should be doing with their analytics efforts that most don’t do today?
RN: Organisations should be looking at integrating their analytics more closely into their decision making processes and systems. This could involve integration with CRM systems, transactional data and campaign delivery systems.
AIM: What are the most significant challenges you face being in the forefront of analytics space?
RN: The biggest challenge is easily finding high quality data scientists.
AIM: How did you start your career in analytics?
RN: I started with a PhD in Computer science (Artificial intelligence) from USC. This involved getting familiar with a lot of statistical techniques that are at the heart of data science and analytics.
AIM: What do you suggest to new graduates aspiring to get into analytics space?
RN: I have three suggestions:
I. Work with real data. Toy examples and text book examples won’t cut it. You need to work on problems yourself especially focusing on defining the problem worth solving.
II. Make your fundamentals strong. Build on your knowledge of analytics in a systematic way. This way you’ll have a full tool set and won’t be trying to solve all problems with just a few tools in your armory.
III. Work with a domain expert. That way you’ll know if you’re working on problems that are useful and actionable.
AIM: What kind of knowledge worker do you recruit and what is the selection methodology? What skill sets do you look at while recruiting in analytics?
RN: We look for people who are highly analytics and are able to define a problem based on a vague initial description. We ensure that the person has strong fundamentals or is capable of grasping new concepts quickly.
AIM: How do you see Analytics evolving today in the industry as a whole?
RN: Today analytics is a buzzword. Everyone calls what they do analytics. This results in a lot of noise but I see that as customers become more evolved you’ll see analytics companies coming out of the clutter.
AIM: What are the most important contemporary trends that you see emerging in the Analytics space across the globe?
RN: The biggest trend I see is the emergence of analytics companies that are focused on particular verticals or particular problems. This will disrupt the body-shopping model of analytics that we see to a large extent today. Also increasing is the emergence of technology solutions that integrate the analytics outcomes into the customer’s systems and processes.
To know more about Germin8, visit www.germin8.com[divider] [spoiler title=”Biography of Ranjit Nair” style=”fancy” icon=”plus-circle”]
Dr. Ranjit Nair is the CEO of Germin8, a products company focused building software products that enable decision making in the face of large volumes of potentially incomplete or inaccurate information.
Germin8’s main product is a Big Data analytics product called Explic8 which helps companies leverage textual conversations from their stakeholders in social media, emails, chats, and surveys. Explic8 is currently being used by the Marketing, Customer Care, Sales and Corporate Communications departments of several large Indian enterprises.
Ranjit has a PhD in Artificial Intelligence from the Computer Science Department at the University of Southern California. Before starting Germin8 in 2007, he worked at Honeywell Research Labs in Minneapolis as a research scientist. He serves as the President of University of Southern California’s Alumni Club of Mumbai, and member of the Advisory Board of USC’s Viterbi School of Engineering. Ranjit is deeply interested in product development, entrepreneurship, decision theory and analytics. His other interests include running, cooking and public speaking.[/spoiler]
Try deep learning using MATLAB