Sigmoidal Success Story
reduction in response times
documents processed as of 2019
What was the business problem?
For years, queries sent to finance departments had to be processed manually. This was extraordinarily time-consuming, extremely prone to human error, and often took employees away from more business-critical tasks.
Digitizing information has several advantages a business can gain on several grounds. Businesses can track their processes better, can provide better customer service, improve the productivity of their employees and reduce costs.
Our client, approached us to aid in the development of a tool that helps finance departments provide fast, reliable answers to the most common queries they receive.
What savings and profits does the client achieve?
Our client was able to automate the process. Manual procedures were made a lot faster and smoother. SIgmoidals’ tool was able to identify when important information is missing, and quickly email a request to fill in the blanks when needed. It was for this reason, that finance departments who utilized it saw a 90% reduction in inquiry response times.
How did we accomplish it?
Sigmoidals’ tool utilizes natural language classification (NLC) to assist in the data extraction process – providing relevant data points to a variety of finance-related questions. This ensures questions are always sent to the right person or department within a matter of minutes – not days.
The tool recognizes inaccuracies or insufficient data, or if your question requires a more detailed answer. If any of these instances occur, it quickly intervenes and sends it to the right person or department. The employee there then figures out what is missing, provides the information, and sends it back to the tool. This enables the process to proceed automatically or, in more complicated instances, it enables the right people to spend the time needed to resolve the issue properly before moving it forward in the process.
If any information is missing, then a request for supplemental materials will be sent to the relevant people almost instantly to keep the process moving.
How did we boost the project with ?
We utilized architecture for reproducible research to repeatedly analyze large, complex datasets, achieving impressive levels of accuracy.
Collaborating with our client we focused on building communities of practice, providing skills and technical novelty, maximizing the effectiveness of the cooperation.
Sigmoidal employed data bias monitoring optimizing new and existing models for performance accuracy. We made sure to perform bias check during training data creation to validate if the data is free of any selection bias.