Why is Kyligence needed now? What has changed in the market overall that makes it either possible or necessary?
LH: People want a better analytics experience. They want more than fast analytics at petabyte scale; they want more than to accelerate data-to-insight but, fundamentally, they want large-scale, intelligent automation of data management and analytics. They want an answer to the huge human effort that is being expended today to collect, cleanse, and organize data for effective, widespread use.
The data analytics market has changed from decision support systems to democratized analytics—an “analytics civilization,” if you like. We are moving away from ivory tower analytics that only supports a few decision makers and moving toward a world of citizen analysts—where everyone can access and use data in their daily work—the so-called data-driven business. This requires an abstracted, standardized data view for all users with automatic high performance and high concurrency.
Data will continue its exponential increase, generated by a growing variety of data sources, structures, and types. Previous notions of data warehouses, data marts, and decision support systems must evolve to adopt agile methodologies, machine learning, and self-tuning/self-healing strategies. Kyligence provides a clear path in that direction.
How has the pandemic impacted the use of Kyligence—has it had any impact?
LH: Our business grew in the fiscal year 2020—almost 200% in annual recurring rate and more than 130% in net dollar retention. The practical impact for us has been similar to that of other companies: less travel and more remote work, meetings, and proofs of concept. Actually, we were able to realize some valuable efficiencies for product development and customer support. Less travel has meant more time to focus on delivery and creating value for customers. Remote work and Zoom meetings have, in a small way, increased cloud adoption. Our cloud revenue increased many times that of fiscal year 2019, which we think is the best evidence of a bright future for our cloud business.
Who are the target customers in terms of types of organizations and roles within companies?
LH: We are focused on financial services industry, manufacturing, telecom, retail, and high-tech companies, including SaaS vendors. Kyligence attracts attention from CEOs, CIOs, chief data officers, directors of analytics, architects, data engineers, data analysts, data scientists—and anyone who needs data for their business.
What are the use cases?
LH: Use cases vary by industry—for example, clients in the financial services industry typically want to build a unified analytics platform reporting services, dashboards, and other in-house analytics. Retailers are more focused on supply chain analytics. Our SaaS clients are building out their data products to leverage Kyligence as their data-as-a-service platform for analytics dashboards for their subscribers.
One particularly interesting use case is at Ping An Bank [in China]. It built a single unified metrics platform that centralized all KPIs, and specifically defined more than 4,000 atomic metrics and more than 500 dimensions covering most of its business operations. It reduced more than 30% of the human effort in data preparation, decreased data project implementation to less than 3 days, and has already replaced more than 25% of its analytics workloads from the legacy platform it was using. The system supports more than 1,500 users and increases day-by-day. These are the types of efficiencies that can change the fortunes of a business.
Looking ahead, how do you see Kyligence differentiating itself from competitors?
LH: We will continue to pursue open source technologies to build the core of our solution and to continue to build a global community of developers and users. We will also continue our focus on delivering on three key values to business users: simplification, automation, and cost-effectiveness.
Simplification means we will continue to improve the user experience for non-tech people so that they can more easily use and benefit from their data. Automation is about how to optimize systems and data pipelines using machine intelligence and to seek better solutions to lowering the cost of any analytics workload.