Amazon Lookout for Metrics Evaluates Business With Machine Learning

Using Amazon’s machine learning technology, Lookout for Metrics automatically checks business metrics, including revenue, web page views, active users, transaction volume, and mobile app installs.

Yesterday, Amazon announced the availability of Lookout for Metrics, a fully managed service that uses machine learning to track key factors affecting companies. Launched at re: Invent 2020 last December, Lookout for Metrics is now available to most Amazon Web Services (AWS) customers through the AWS console and support partners.

Organizations analyze metrics or KPIs to keep the business running efficiently and effectively. Business intelligence tools have traditionally been used to manage data collected from disparate sources, but it is difficult to identify anomalies in it. Traditional rule-based methods look for data outside of numeric ranges. The problem is that these ranges are usually static and do not change based on conditions such as time of day, day of the week, seasons, or business cycles.

Using Amazon’s machine learning technology, Lookout for Metrics solves this problem by automatically checking business metrics, including revenue, web page views, active users, transaction volume, and mobile app installs. The service also helps diagnose the root cause of anomalies such as unexpected drops in revenue, abandoned shopping carts, spikes in payment transaction failures, an increase in new user signups , and more.

Customers can connect Lookout for Metrics to 19 popular data sources such as Amazon CloudWatch and Amazon Redshift, as well as SaaS apps from Salesforce, Marketo, Zendesk, and more. Lookout for Metrics collects and prepares data, selects the most appropriate ML algorithm, begins to detect anomalies, groups related anomalies together, and summarizes possible causes. For example, if your website traffic suddenly drops, Lookout for Metrics can determine that the cause was an unintentional shutdown of an ad campaign.

Lookout for Metrics also connects to notification and event services, allowing customers to create alerts or actions, such as registering a ticket or removing an incorrectly priced product from a retail website. When the service begins to return results, developers have the ability to provide feedback on the relevance of the detected anomalies, which the service uses to improve its accuracy.

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,” says Swami Sivasubramanian, VP of ML for AWS. “But catching and diagnosing anomalies in metrics can be challenging, and by the time a root cause has been determined, much more damage has been done than if it had been identified earlier.”

At the time of this writing, Lookout for Metrics is available at select Amazon locations in the US, Europe and Asia. In the next few months, it will appear in other regions.

Stay tuned with Software Focus!