See the full publication here.

JANUARY 2021, CASE ,HBS CASE COLLECTION

ABSTRACT

Autonomous business monitoring platform Anodot leveraged machine learning to provide real-time alerts regarding business anomalies. Anodot’s solution was used in various industries in order to primarily monitor business health, such as revenue and payments, product usage and customer experience. Every day, Anodot used 30 types of learning algorithms to analyze 6.2 billion data points and 428 million unique metrics. By 2019, Anodot’s platform tracked more than 400 million metrics daily, driving four billion autonomous decisions that were translated to less than 1,000 alerts for all its customers. This highly accurate monitoring led to a low incidence of false positives, or false alerts, and customer satisfaction was high. Since Anodot’s tool had the ability to identify granular business anomalies in real time, such as an unexpected drop in e-commerce sales for particular products or markets due to a technical glitch, fast detection and resolution of the problem meant that the potential financial damage could not be easily measured. The management team contemplated several strategic issues: How could they help their customers realize the value of Anodot? They had been working on several tools to show the value in different stages of the sales cycle and post-sale, but it was still hard to measure the actual financial value. In 2019, Anodot had adjusted its strategy to focus on client verticals and use-cases that would benefit most from Anodot. Would this make the sales process any easier? An improved product-market fit, combined with an ability to measure Anodot’s value, could increase conversion and retention. Should they narrow down the use cases even more? As the team was thinking about their next funding round, it was important to prioritize their efforts.

You'll believe it when you see it