Data monitoring solutions are required for timely decision-making

9 Nov 2020

A successful business needs accurate data on its customers, products, and markets. Many companies recognise the importance of data in developing and preparing business strategies. However, proper data management and analytics is a herculean task for most companies. Therefore there is a trend of outsourcing some parts of that task, namely metrics observability, data quality, and data diagnostics.

Data monitoring solutions are required for timely decision-making

Companies that want to be competitive in the 21st century should have proper data analytics tools since the right insights can increase productivity, facilitate smooth operations, reduce security risks and drive better decision making. This is where Orbiter, a new Y Combinator-backed automated data analytics platform, comes in.

The company is co-founded by Winston Zhang, Victor Zhang and Mark Wai, recognized experts in business, product management, and data science who have worked with Fortune 500 companies like Tesla, DoorDash, and Facebook. "As business and product leaders, it often felt impossible trying to manually keep up with different dashboards and changing metrics,” explains Winston Zhang, co-founder of Orbiter. According to him, the other reason they felt the need to introduce a machine learning and statistics powered data analytics platform was because automation tooling was severely lacking for data and product folks. Unlike engineers who have plenty of tools for automated monitoring, alerting, and diagnostics (e.g. DataDog, Sentry), data scientists and product managers have few options. They built Orbiter to solve this problem. Orbiter is an innovative data monitoring company that uses statistical and machine-learning models to analyze business or product metrics for abnormal changes in near real-time.

Especially in times of crisis, unusual changes to the business or product should be monitored 24/7 to identify deficiencies that may have severe customer consequences. According to Winston Zhang, this is one of the reasons why a Machine Learning Alert (MLA) system can save the day. "Applying machine learning and statistical models to alerting not only improves accuracy but also reduces the significant overhead that is required to set up traditional alerting systems with fixed thresholds. Once the normal behavior of metrics is understood, it becomes easy for a well-trained system to flag abnormal changes immediately for the team’s attention,” he says.

“The MLA is an automation function. It can serve business use cases, such as alerting our customers of abnormal trends in conversion or sign-ups, but also serve internal use cases such as data quality observability,” Zhang said while talking about MLA and the service that his company provides.

Zhang believes that ML platforms like Orbiter will enable data and product leaders to never miss an issue impacting their customers or revenues again. Most recently, Zhang’s startup, which is backed by Y Combinator, has been working closely with a logistics company Doordash, which provides door-to-door delivery of food from restaurants. The enablement of automated monitoring for key product metrics ensures that DoorDash is able to continually deliver a delightful experience for their customers. By equipping statistical modeling to generate alerts, it’s possible for companies like Doordash to stay on top of real-time trends without staring at dashboards all day long.