Do organisations looking to gain value from data-driven insights actually have a standard execution or estimation methodology? At a time when companies are struggling with an ever-increasing amount of data, business teams lack sophisticated analytics solution which covers end-to-end analytics development lifecycle and most vendors lack an outcome-driven approach.
Amidst all the noise about analytics, machine learning and AI, organisations look for a clear “data strategy” but the big aspects of it is patchy as it is built on legacy ideas and technology — ETL, MIS, reporting. Even though organisations have made considerable progress in capturing enterprise data, around 75% of this consist of unstructured data and it is growing at a rate of 65 percent every year.
Often, business leaders struggle to integrate data and identify business use cases. It is here that most organisations lack clarity and fail to formulate a standard framework. Analytics India Magazine caught up with Rishi Jain, CEO and co-founder of Infintus Innovations and technology evangelist, to discuss how Intelligent Enterprise.ai platform goes beyond simple tool features to deliver cost-effective, scalable analytics capabilities. The solution encompasses a “full stack” of analytics, domain expertise bundled in with Analytics App feature and automated reporting. It provides an end-to-end data analytics platform for users to directly integrate, analyze, and visualize data.
The technology platform can read and collate structured and unstructured data in real-time — enabling businesses to discover use cases to create business value. To allow users to extract value from data and meet KPIs, the platform has a added capability — Analytics Apps which can be used to explore and build relevant use cases. To date, 60+ apps for various domains have been created on the technology platform.
Analytics India Magazine: What was the core idea behind setting up IntelligentEnterprise.ai?
Rishi Jain: Data is growing in size, complexity and diversity. Just as Google crawlers go across the internet, collect, index and store the information and present it in the right context based on algorithms when user searches the information — we have a similar purpose for IntelligentEnterprise.ai to be able to crawl the enterprise data and documents, ingest, index, create algorithms and make it available for enterprise analytics and intelligence.
AIM: What do you think are the problems in Enterprise Data and Analytics solutions?
RJ: While building IntelligentEnterprise.ai, we spoke to over 100+ enterprise analytics leaders globally. There were 3 core issues — costs were high, time taken for information to reach decision makers was very long and the intelligence on unstructured data was underleveraged and painful.
I believe, the fundamental issue causing this is legacy and multi-software environment (ETL, Database, Reporting, Visualisation) and multi-skill (resource needed for multiple software) environment. Most organisations were trying to build modern AI/ML layers on top of legacy ETL/data warehouse foundation. It is not the right model.
AIM: How is IntelligentEnterprise.ai is aiming to solve these issues?
RJ: We are focusing on building a strong foundation for data in enterprise. If the data foundation is strong, analytics and intelligence becomes relatively easier. What I mean by building strong foundation is to bring data from structured (databases) and unstructured (files, images, voice) in one enterprise datastore and secondly doing it in as real-time as possible without impacting the performance of the transaction systems and thirdly doing it a configurable manner minimising need of coding.
AIM: Who are your competitors and how is IntelligentEnterprise.ai different from other solutions in the market?
RJ: While our purpose is different as stated above – to organize enterprise information (structured and unstructured) and make it available through human centric interactions; we do come across players in the value chain of data analytics — data extraction, transformation, visualisation, predictive analytics such as Tableau, QlikView, Sisense, Domo, YellowfinBI, Alteryx, Microsoft to name a few and other niche analytics providers.
We are different from most competitors architecturally in our approach to stream, clean and index the information. In addition, we also provide a no-coding approach, wizard-driven data management that can be done by data analyst.
Secondly we want our users to think in terms of “Analytics Apps” a way to standardise around use-cases and drive collaborative innovation across enterprises. 60-70% analytics needs across the enterprises are common, there is no need to reinvent the wheel on what inputs fields are required, what visualisations are needed for use-cases such as lead analytics, customer analytics, talent analytics, financial analytics etc.
AIM: You talked about the platform integrating data from multiple sources. Can you elaborate more on this?
RJ: Through the App Studio, data analysts can set the collection profile and collection rule for multiple sources from where the data is to be streamed. Then Data Streamer comes into action to understand the structure of the data and stream. Data Processor transforms and cleans the data and indexes it in NoSQL database. Data Orchestration engine keeps an eye on data flow across multiple threads and ensures that data is in sync with the source system. These components can be deployed across multiple servers to manage scalability and performance.
AIM: Can you talk about the need for developing Analytics Apps?
RJ: During our discussions with 100+ analytics leaders, we found that same use-case could mean very different to different enterprises. Hence, we felt there is a need of standardisation of use-cases in terms of input and output. That’s the genesis of Analytics App. It provides a touch and feel perspective of value. Once enterprises see a use-case is in terms of visuals, alerts, they can then start to connect with their business and customise in that context. It is like semi-cooked food getting delivered to be customised to taste and ready to eat!
AIM: Can you give a few examples of Analytics Apps that have been built on the platform?
RJ: Several Analytics Apps have been built on the platform across multiple industries such as insurance, healthcare, financial services, professional services, education, e-commerce and functions such as sales & marketing, call centre, customer service, finance, human resources and legal. A partial list is available on our website. We are getting a particular interest in doing analytics on unstructured data such as invoices, market intelligence, leases and contracts, medical reports.
AIM: Can you throw light on the solution’s interactive UI which allows users to meet their analytics and reporting requirements effectively?
RJ: We believe that users would like to interact with analytics like stock tickers, search bar and voice. That’s our ultimate goal. Meanwhile, platform provides hundreds of visualisations and basic NLP to interact with visuals using search bar. Visuals can be easily developed and edited by data analysts or data-savvy business users. Visuals can also be shared across or embedded in business applications.
AIM: In terms of cost-effectiveness, can you tell us how it is the most cost-effective and scalable solution in the market?
RJ: Since it is analytics-in-a-box full stack, it doesn’t require buying of any other license. You don’t need a technology staff to write code. Data Analysts can learn and configure it in around 4-6 weeks’ time. Therefore this eliminates over 50% costs in the value chain around multi-software and multi-skills need. More importantly, as data changes in the sources systems, the updates to IntelligentEnterprise.ai require little intervention, testing and support. So total cost of ownership, over a period of 3 years is 60%+ lower.
AIM: How does the platform reduce the burden of hiring big IT teams especially at small companies which are just getting started with analytics?
RJ: IntelligentEnterprise.ai is designed to be highly configurable with little or no need of IT skills. It needs a few data analysts for app building, data analysis, developing visuals and business alerts. For small companies hiring data analysts can be a challenge, therefore we offer data analyst effort bundled in our pricing plans. So users just have to install it and lean on us for analytics.
AIM: As you mentioned, most large enterprise already have some analytics tools and systems, how can they benefit from the platform
RJ: There are significant investments that large companies have made in analytics. Those will continue to exist. IntelligentEnterprise.ai can augment existing tools. For example, IntelligentEnterprise.ai can help with data streaming and management while Tableau and Qlikview can be used for visualization.
At the same time, the enterprise needs to look at its data and analytics strategy in terms or Renew and New. They need to Renew their existing analytics infrastructure and bring New tools to manage the growing data and drive innovation. IntelligentEnterprise.ai can be used as part of “New” strategy, it can lead the new use-case development and collaborate with existing infrastructure to augment it.
AIM: You mentioned about introducing human-centric interactions in the tech platform. Can you tell us how voice will become the new interface for analytics.
RJ: As per Gartner, about 30% of all searches will be done without a screen by 2020. Post Millennials (born 1997 – present) at voice, video and short messaging generation. 20 years ago we might not have thought of voice based search when even Google didn’t exist. Imagine, this generation and next coming to work in enterprises. While the need is imminent on one side, good news is that the technologies Computing power/AI/ML/Big Data are also maturing fast to enable it. Do we have any other choice?
Try deep learning using MATLAB