Amazon Web Services Inc. today announced the general availability of Amazon Lookout for Metrics, a machine learning service designed to help enterprises detect and respond to important business events such as a sudden jump in user sign-ups. 

With the service, AWS is looking to address a number of technical challenges that normally make such business monitoring difficult. 

The first is the issue of important events going undetected because of visibility gaps. Usually, companies detect business changes by configuring an analytics system to generate an alert if a certain condition is met, for example when daily product sales drop below a certain predetermined level. The issue with such static alerting rules is that they can become outdated over time as a company evolves and metrics such as average daily sales volume change, which makes it difficult to reliably detect anomalies. 

Lookout for Metrics replaces static rules with artificial intelligence models that AWS says automatically adapt in response to new conditions. When analyzing sales activity, for  example, the neural networks can take into account seasonal demand fluctuations. In allowing more reliable detection of important events, AWS claims that its machine AI’s ability to take more factors into account reduces the number of false positives that customers receive. 

Another challenge Lookout for Metrics aims to address is the task of identifying anomalies’ root cause. The faster a company can uncover what caused an issue such as a drop in a given product’s sales, the sooner it can fix it. According to AWS, Lookout for Metrics speeds the task by automatically pinpointing the likely reasons behind a given business event. 

The service diagnoses problems using technology that AWS parent Amaozn.com Inc. relies on internally to analyze its own operations. AI algorithms estimate what could be causing an operational anomaly by looking through data collected from a company’s day-to-day business activities. Companies can connect Lookout for Metrics to operational data troves they store in AWS services and external software-as-as-service platforms such as Salesforce.

“From marketing and sales to telecom and gaming, customers in all industries have KPIs that they need to be able to monitor for potential spikes, dips, and other anomalies outside of normal bounds across their business functions,” said Swami Sivasubramanian, vice president of the Amazon Machine Learning unit at AWS.

The launch of Lookout for Metrics expands the growing lineup of managed machine learning services the cloud giant offers to its customers. AWS also provides Amazon Lookout for Vision, which helps manufacturers spot production issues, and Amazon Lookout for Equipment, a service that can identify malfunctioning industrial equipment based on sensor logs.  

Image: AWS