Thesis Maps: Founder Chats, No. 2
Ramesh Panuganty: CEO of MachEye
Thesis Maps previously explored Jamin Ball’s vision for the future of DataOps and the analytical engineer. Today, we chatted with Ramesh Panuganty, CEO of MachEye, whose company is one we’ve identified to watch within the “modern business intelligence tools” space.
Thesis Maps (TM): As a serial entrepreneur, how would you say your prior experience as a founder has shaped your approach to building MachEye?
MachEye (ME): My first company – a cloud management platform (CMP) – was launched in 2008, back when the word “cloud” didn’t exist. I believed in infrastructure on an “as-you-consume-it” basis – and that’s what we now refer to as the cloud. My second startup, Drastin, created a category called “conversational analytics” and was acquired by Splunk. My third company focused on Natural Language Generation (NLG) and generated educational content intended for SAT and ACT preparation.
MachEye was created with the vision to become a “personal business companion”. We combine a Google-like search experience and a YouTube-like audio-visual experience to discover insights in enterprise data. We shattered the paradigm of “what you ask is what you get,” on which the entire premise of business intelligence (BI) was built. Business users ask questions that subject matter experts help answer. But how can they ask questions on the unknown unknowns – the things they don’t know or anticipate?
TM: What customers have you had the most traction with from an industry perspective? Who are your buyers typically from a functional perspective within companies?
ME: In terms of buyer persona, our current customer base falls into two broad categories. There's a lot of hype about being data-driven, but we believe that every company needs to be insight-driven. Being data-driven is meaningless unless the data is translated into meaningful insights.
The first category is business buyers who struggle to understand what their data can tell them. These customers love the idea that we can drive decisions with insights. We’ve closed multiple customers in just two business days after the first meeting, because they understood the value proposition. The second category is the IT buyer; people who are responsible for bringing innovation into the organization and improving processes or technology. We’ve had successes in both categories.
From an industry perspective, we’ve had diverse horizontal traction, from consumer products, to retail, to education, and so on.
TM: Can you dive deeper into your typical customer persona and talk through a couple use cases?
ME: Our typical customers are overwhelmed by new data on a daily basis and need to make sense out of it for many end users/consumers. Consider the supply chain function in a consumer products company, or completing a transactional analysis of daily purchases at an e-commerce company.
In terms of example use cases, a business user can explore data on products sold, distributors, etc., and get insights about average daily volumes, store locations, and so on. Different users such as account managers, sales managers, or territory managers, etc., get relevant insights.
One unusual BI use case for us is an EdTech customer whose end users are teachers that had never previously used any BI product like PowerBI or Tableau. Due to the sudden switch to online learning last year, they’ve had to analyze a lot of data on student submissions and their performance on exams. They can get these answers and insights very easily in MachEye.
TM: So someone who is an SVP, for example, is likely not interested in the same level of detail as someone who is a regional GM. How do you think about contextualizing the outputs for different audiences based on their levels and their function within a company?
ME: Good question. I compare our product sometimes to Flipboard or news.google.com. The same way they customize content, we try to do the same with enterprise data using two parameters. The first is the user’s persona and the metrics or KPIs that they’ve shown interest in. And the second is data governance: we account for what people are expected to see and the data they have access to. Each person gets insights in their own customized way depending on their persona, behavior, and likes and dislikes.
TM: I know you have something called a “Data Quality Index” and use that as an effort to avoid a “garbage in garbage out” scenario. It’d be great to hear more how you think about that, especially with AI doing the heavy lifting.
ME: I'm glad you noticed that finer detail. No other BI product uses data quality for data analysis – they think it’s up to the user to decide what to do with poor quality data. I think it’s essential to measure data quality to produce meaningful answers. In fact, the concept is so new that there are no industry examples of putting data quality on a scale. We came up with this quantifiable measure!
My inspiration was the California wildfires and the Air Quality Index, using a scale from “Good” to “Unpredictable”. We assign qualitative ratings using multiple characteristics for every table, column and row, such as clarity, completeness, and consistency of data.
Consider a scenario where you want to know the top five stores that sold the most lawn chairs. If only 40% of the “store name” column is populated in the data, we’ll show you the top five stores within the available store names, but more importantly, we highlight that 60% of store names are missing in the data.
TM: Operating a business in any industry is essentially just pulling one of a handful of levers to react to different events that happen. It seems like with enough context MachEye could perhaps even serve as a recommendation engine for “next best actions” for users to actually go out and act on within their businesses. How do you think about that?
ME: Let’s think of a retail customer scenario: if we find that lawn chair sales are increasing because of a campaign in California, we tell the user “The spring campaign you launched six weeks ago is driving sales up” in an audio-visual format. We also compare this with East Coast sales that are lower where the Spring promotion is not active. We don’t call these “predictions” or tell the user to launch the Spring campaign on the East Coast; we analyze what’s happening in the data and why there’s movement in business metrics, and present that in a way that makes it easier for users to decide on the next best action. We could tweak our NLG engine to present the content as either analysis or as action-oriented items, but we keep a careful balance.
TM: That makes sense – you don't want to be too heavy-handed in your recommendations. Knowing what MachEye is capable of and knowing the responsibilities that are maybe being taken off an existing business analyst, how do you see the role of a business analyst in the future?
ME: Most business analysts at our customers’ organizations are overwhelmed with their day-to-day jobs. Business users keep asking questions, and analysts keep answering them, without being uplifted to go beyond the question-and-answer relay race. We empower analysts to bring in more data into the hands of business users, democratize AI, create business metrics, define business-specific vocabulary, configure data governance policies, and make the organization insight-driven. We help them look better than ever and make insights available to everyone in the organization.
Let’s imagine a user asking for West Coast sales, but no data definition is available for “West Coast”. Analysts can step in and create this vocabulary within MachEye as a business-specific definition. We inform the analyst about usage needs and the analyst can essentially become a “data custodian” like never before. Analysts can now add more value to the business because they stay on top of business needs. They are no longer just answering questions coming from the business – they can now steer them.
TM: What's one or two things that surprised you about your users and their reactions to the product as you've iterated it?
ME: Some customers initially asked, can we try this with an Excel sheet? So, we allowed them to upload their own CSV data. But then we saw that imported data isn’t valuable because it remains static. How much analysis can they really do? They might play with it for a day, but it doesn’t reflect changes in the business and can’t really make an organization insight-driven.
TM: What's one or two features that you're most excited about on your product roadmap over the next six months that are currently under development that you're able to talk about?
ME: Our goal for the next six months is to make it very easy to onboard new customers on the cloud. We want to make it just a matter of minutes for somebody to sign up and start consuming insights wherever their data is.
TM: You raised your seed round late last year in October, led by Canaan Partners. It'd be great to hear a little bit about what it's been like working with your new investors and how they've been able to support you as a founder, and support your team.
ME: Our investors have been great; they believe in the problem we are solving, the market opportunity and perhaps most importantly, the vision of the company. They have been a huge help in making sure that our go-to-market approach is strong, and that we have the correct team and the leadership for scaling up.
TM: And so, as you think about what the next rounds of funding look like, how do you hope to round out your cap table? What attributes of an investor do you think are going to be most important to MachEye's success?
ME: We are not looking for an investment at this point, but an ideal next round investor would be somebody who is very passionate about bringing a paradigm shift to the enterprise world and improving data experiences. We are the only company that are presenting data experiences as audio-visuals.
Enjoyed this post?