carrot carrot carrot Change Centers x cognizanti collaborators create-folder Data Science Decisive Infrastructure download download edit Email exit Facebook files folders future-of-work global sourcing industry info infographic linkedin location Mass Empowerment Mobile First our-latest-thinking pdf question-mark icon_rss save-article search-article search-folders settings icon_share smart-search Smart Sourcing icon_star Twitter Value Webs Virtual Capital workplace Artboard 1

Please visit the COVID-19 response page for resources and advice on managing through the crisis today and beyond.


Algorithms That Learn Are the Stars of the Self-Service Show

The Challenge

The larger a company’s customer base, the more important self-service tools become. For one telecom giant, self-service requests generate about 800,000 annual application programming interface (API) calls across its framework to resolve various customer issues. This self-help system reduces costs and improves customer satisfaction.

When these API calls fail, customers suffer and costs increase, so the company strives to predict and prevent failure as well as ensure successful API execution. The telecom already has an API management solution, but it asked Cognizant for help developing a predictive analysis algorithm to proactively determine API failure and thus prevent future system failures.

Our Approach

Cognizant’s analysis describes how the most commonly used APIs behave, as well as the specific failure patterns for each, to help prevent future interruptions in service. We leverage random forest algorithms, neural networks and confusion matrices to improve predictive capabilities, enable analysis and optimize model output.

Our technology reads machine and application logs. It uses predictive algorithms to identify device behavior patterns and outcomes such as failures. These self-learning capabilities can alter the predictive algorithms to make the best use of the latest data feeds as well as leverage advanced analytical methods such as event classification, text mining, association rules, correlation mapping and time series extrapolation.

Predicting Failure Makes Customer Service Consistent

By predicting and preventing API failure for self-service tools, we help the telecom directly improve customer satisfaction. Improved self-service increases the volume of requests that can be resolved by customers and saves more than $100,000 a year for just one of the close to 70 APIs the company uses.


reduction in management costs per API


increase in speed of insight generation

Reduced expenses

for troubleshooting, infrastructure overhead and deployment effort