machine learning

Amazon’s managed service uses ML to detect anomalies in business metrics

The service also diagnoses the root cause of anomalies like unexpected dips in revenue, high rates of abandoned shopping carts, spikes in payment transaction failures, increases in new user sign-ups, etc.

Amazon Web Services (AWS) announced the general availability of Amazon Lookout for Metrics, a new fully managed service that detects anomalies in metrics and helps determine their root cause.

Amazon Lookout for Metrics is a new machine learning service that automatically detects anomalies in metrics and helps customers quickly identify the root cause. Lookout for Metrics puts the same technology used by Amazon internally to detect anomalies in its business metrics into the hands of every developer. Customers can connect Amazon Lookout for Metrics to 19 popular data sources, including Amazon Simple Storage Solution (S3), Amazon CloudWatch, Amazon Relational Database Service (RDS), and Amazon Redshift, as well as SaaS applications like Salesforce, Marketo, and Zendesk, to continuously monitor metrics important to the business (e.g. total revenue, gross margin, average purchase frequency, return on advertising spend, etc.).

Amazon Lookout for Metrics automatically inspects and prepares the data, selects the best suited machine learning algorithm, begins detecting anomalies, groups related anomalies together, and summarizes potential root causes. For example, if a customer’s website traffic dropped suddenly, Amazon Lookout for Metrics can help them quickly determine if an unintentional deactivation of a marketing campaign is the cause.

“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. 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,” said Swami Sivasubramanian, Vice President of Amazon Machine Learning for AWS. “We’re excited to deliver Amazon Lookout for Metrics to help customers monitor the metrics that are important to their business using an easy-to-use machine learning service that takes advantage of Amazon’s own experience in detecting anomalies at scale and with great accuracy and speed.”

The service also ranks the anomalies by predicted severity so that customers can prioritize which issue to tackle first. Amazon Lookout for Metrics easily connects to notification and event services like Amazon Simple Notification Service (SNS), Slack, Pager Duty, and AWS Lambda, allowing customers to create customized alerts or actions like filing a trouble ticket or removing an incorrectly priced product from a retail website. As the service begins returning results, customers also have the ability to provide feedback on the relevancy of detected anomalies via the AWS console or the Application Programming Interface (API), and the service uses this input to continuously improve its accuracy over time.

Amazon Lookout for Metrics helps customers monitor the most important metrics for their business like revenue, web page views, active users, transaction volume, and mobile app installations with greater speed and accuracy. The service also makes it easier to diagnose the root cause of anomalies like unexpected dips in revenue, high rates of abandoned shopping carts, spikes in payment transaction failures, increases in new user sign-ups, and many more—all with no machine learning experience required.

Organizations of all sizes and across industries gather and analyze metrics or key performance indicators (KPIs) to help their businesses run effectively and efficiently. Traditionally, business intelligence (BI) tools are used to manage this data across disparate sources (e.g. structured data stored in a data warehouse, customer relationship management data residing on a third party platform, or operational metrics kept in local data stores) and create dashboards that can be used to generate reports and alerts if anomalies are detected. But effectively identifying these anomalies is challenging. Traditional rule-based methods are manual and look for data that falls outside of numerical ranges that have been arbitrarily defined (e.g. provide an alert if transactions per hour fall below a certain number), which results in false alarms if the range is too narrow, or missed anomalies if the range is too broad. These ranges are also static, and don’t change based on evolving conditions like the time of the day, day of the week, seasons, or business cycles. When anomalies get detected, developers, analysts, and business owners can spend weeks trying to identify the root cause of the change before they can take action.

Machine learning offers a compelling solution to the challenges posed by rule-based methods because of its ability to recognize patterns in vast amounts of information, quickly identify anomalies, and dynamically adapt to business cycles and seasonal patterns. However, developing a machine learning model from scratch requires a team of data scientists that can build, train, deploy, monitor, and fine tune a machine learning model over time. Furthermore, a single algorithm rarely serves all of the needs of a business, which causes businesses to expend meaningfully more time and expense creating and maintaining multiple algorithms to solve different use cases.

Lookout for Metrics is available directly via the AWS console as well as through supporting partners in the AWS Partner Network to help customers implement customized solutions using the service. The service is also compatible with AWS CloudFormation and can be used in compliance with the European Union’s General Data Protection Regulation (GDPR). Lookout for Metrics is available today in US East (N. Virginia), US East (Ohio), US West (Oregon), EU (Ireland), EU (Frankfurt), EU (Stockholm), Asia Pacific (Singapore), Asia Pacific (Sydney), and Asia Pacific (Tokyo), with availability in additional regions in the coming months.

 

Leave a Reply

Your email address will not be published. Required fields are marked *