Case Study

Improving tracking & forecasting for Canada’s leading billing software solution

We built a proprietary anomaly detection and billing model that improved accuracy by 3 percentage points, unlocking millions of dollars in savings for our client’s customers.

Moon Graphic

The Client

A company providing subscription and usage billing and rating solutions for data, voice, SMS and API usage. The company’s SAAS platform allows its customers to monetize in real-time any triggered event in the connected network of hundreds of millions of IoT devices.

The Challenge

Today’s Telecom, IT and Utility companies are looking for reliable billing solutions that provide accurate data reporting. The efficacy of usage-based billing models is predicated on collecting, analyzing and forecasting event information from millions of IoT devices in real-time. For decades, businesses have lost millions of dollars due to irregular tracking, anomalous data and an inability to forecast.

As a leader in billing and reporting software, our client was looking to improve their overall accuracy to help mitigate risk and drive cost-savings on behalf of their customers.

The Solution

Our analysis revealed that a 1% increase in tracking and forecasting accuracy could save millions of dollars for our client’s customers. Based on this insight, we built a state of the art, proprietary model that offered unparalleled accuracy versus the industry-standard, time-series tracking algorithms. We integrated the model using industry-standard web services within our client’s monitoring and production infrastructure, and improved the accuracy of the software from 90 to 93%, delivering a strong ROI for our client’s customers.

IP Generated

Crater Labs created a unique deep learning model capable of analyzing irregularly spaced, non-stationary time-series data, identify anomalies and predict when and with what confidence the next anomaly would occur.

Benefits & ROI

Our model improved anomaly detection and billing accuracy by 3 percentage points, unlocking millions of dollars in savings for our client’s customers.

The new solution enabled our client to detect issues faster, while mitigating risks and driving efficiencies on behalf of their customers.

Keep Reading

More Moonshots Worth Celebrating

Reducing Audit Risks in Tax Filings with Deep Learning

Crater Labs developed a neural network to highlight potential risk factors in R&D tax credit submissions, saving hundreds of hours in audit defence times.

Read More

Identifying Potential Security Access Violations

We developed a semi-supervised neural network to identify anomalous security access card swipes in a large corporation with numerous facilities throughout the world, previously hidden secure access violations.

Read More

Optimizing product placements in print flyers to maximize revenue

We combined a convolutional neural network with a custom region proposal network that provides key insights on how to maximize revenue with flyer layout, with a double-digit return expected within the first two years.

Read More