This article was first published on DataScience.com.
The worldwide enterprise software market is steadily growing and is forecasted to grow 7.6 percent this year, as Gartner predicts. The competition is fierce and, in the span of the next few years, AI-powered services will cannibalize revenues for 30% of market-leading companies across enterprises. It is no wonder then that companies are desperately looking for implementation guidelines to help them move into a new business landscape that’s heavily influenced by machine learning and automation.
I’ve been fortunate to talk to Oleg Rogynskyy, CEO of People.ai, who is utilizing machine learning to revolutionize how enterprise companies analyze and improve their workflows and sales. Rogynskyy is considered a machine learning veteran. Prior to People.ai, he started and scaled several companies focused on big data, two of which, including Semantria, were subsequently acquired. He’s made a career of solving business problems through machine learning and was kind enough to share his insights on what’s going on with the enterprise market right now and how companies can stand out. Here’s what we talked about.
Access to Workflows and Proprietary Data
Artificial intelligence is completely changing the way companies innovate and communicate their processes and services. But when it comes to building and running successful AI-powered companies, there are several aspects that organizations need to get right. Sometimes it’s getting access to proprietary data and other times it’s having the best algorithms in the world. So what is that key differentiator?
According to Oleg Rogynskyy, “Algorithms don’t make or break an AI company. It’s about having a private dataset with unique value, coupled with a powerful business workflow. The only way a true enterprise product will succeed is by seamlessly embedding itself into a complex enterprise workflow and making it easier. In our case, we’re solving the age-old problem of aligning sales and marketing around sales activity data and completely automating many of their workflows.”
Data as a Competitive Advantage
Many people claim that data is the new oil and a source of competitive advantage. There is even a discussion happening around new competitive moats in the form of Systems of Intelligence. Companies are constantly looking for a new competitive edge and it seems that data is already powering that.
As Oleg Rogynskyy explains, “Data collection used to be a great moat. A system of record is a unified place where data is collected — whether it’s financial, sales, employee, customer, or other data — in a single place. Think of Salesforce.com, Workday, or NetSuite. A system of intelligence is the next generation platform, one that will replace systems of record. It isn’t just collecting data, but enabling humans to act on it by surfacing valuable insights within the workflow or acting autonomously on our behalf. It’s an intelligent layer on top of a system of record.”
In the case of People.ai and other companies that integrate systems of record, this data is used to create a System of Intelligence. The defensive moat is the fact that companies use those integrated products with very proprietary types of data, combined with complex workflows, and tied directly to business results.
Beating the Cold Start Problem
Since most of the enterprise companies are reluctant to give away their data, how do entrepreneurs beat the cold start problem? Is it focusing on something that is not subject to a lack of data? Is it possible to create AI-powered solutions that can be used by any company without giving away years of their internal data?
As Rogynskyy told me, “Over the last decade we’ve seen companies become more open to giving away their data to some of the larger players, e.g. Salesforce.com, Oracle, Microsoft, Google, Workday etc. However, as a young startup, a large enterprise is naturally going to be uncomfortable giving you access to their private data. We got around this in two key ways — security and value creation. Understanding that our target market was the enterprise market, we focused on security from day one, building security and privacy protection into our platform by default. We also focused on doing everything possible to show that we are a security-centric organization — getting our SOC2 certification, frequent penetration testing, compliance with GDPR, etc.”
Though security and certifications have a huge role in the process of getting access to proprietary data, entrepreneurs have to be able to provide the answer to the most important question bothering enterprises: “What’s in it for us?” Ultimately, the benefit of your service needs to outweigh the risk. In other words, there needs to be a clearly defined upside.
Evolving Around Artificial Intelligence
The revolution is coming, so who is going to win and benefit from it the most? According to Gartner, by the year of 2019, startups will overtake Amazon, Google, IBM and Microsoft in driving the artificial intelligence economy with disruptive solutions.
There is still too much manual and menial work happening, costing companies tens of millions a year. It’s a waste of human talent, making highly educated and creative individuals repeat the same tasks again and again. No doubt that AI will help to phase out many of these ineffective manual workflows and let people concentrate on the business. A perfect example is how sales reps typically spend a third of their time manually entering data into a CRM, which is absurd and ripe for disruption.
According to Oleg Rogynskyy, “The winners will be nimble AI companies that can disrupt these workflows and operate with unique data sets which have very strong network effects. There should be a clear cyclical motion; more customers equals more data equals improved product equals more customers. It’s a process that benefits everyone involved and dramatically increases efficiencies for the customer and their employees.”
At the end of the day, companies can have the hottest AI technology in the world but if they are not addressing real challenges that businesses want to pay for, they don’t stand a chance. To differentiate, companies have to solve real problems that customers are facing every day.