Automatic Incoming Mail Classification: fast, efficient and accurate

Insurance companies are usually considered flagships for excellence in internal organization, structure and overview. Unfortunately, they are not immune to the tremendous challenges of big data, and are becoming increasingly inundated with vast quantities of information. The amount of notifications and messages which bombard the average insurance company, be they letters, e-mails, faxes or the now ubiquitous social media posts, is equivalent to around 25,000 pages every single day. Rapid processing is an absolute must. The fact that this seems impossible is generally not due to shortcomings in the competence of the staff. The sheer amount of mail makes it humanly impossible to forward the right information to the pertinent department in time for the customer to receive a prompt response.

Generally insurance companies are all facing the same ongoing issue: every day the staff opens letters, reads e-mail and fax messages, analyzes and sorts this mail and then forwards it to the respective agent. Not only does this process seem to take an eternity, it is also precariously fragile; employee sick leave or a sudden storm resulting in a peak period of inquiries leave it vulnerable to bottlenecks and an inability to handle the incoming data. In addition, the amount of incoming mail is constantly growing and the employees are becoming buried under an avalanche of data. What is needed is a system that can take over this task, optimize and speed up the distribution process and allow the staff to focus on the customer.

Experts have long been looking for ways to make this process more efficient in order to improve customer service and free up the staff. The incessantly growing amount of daily incoming mail makes this topic more relevant today than ever. The astonishing thing, however, is that not only do potential answers to this problem exist, but fully designed and ready-to-use solutions are already available. Some companies are already one step ahead and are actively pioneering the development of these systems. They are face to face with companies in the insurance sector, with its inherent flood of incoming mail, and the challenge of optimizing the handling process now and in the future. Currently only a few insurance companies have considered this issue at all and are looking for a new approach. While there is a handful of companies which reap the benefits brought about by these products, the majority are still buried under the immense mountain of data. The solution to this is simple, quick and effective: automated incoming mail classification.

Self-learning Systems

Many managers reading this are probably now wondering how a computer program can sort documents. How can it know which piece of mail needs to go to which agent? How can it suddenly be so smart?

Intelligent incoming mail classification systems really do master all these skills. They pursue a new approach in which they analyze documents semantically, thereby actually understanding the content, and then in a broader sense through "predictive analytics", determine how the document should be classified. Predictive analytics can best be described as learning from the past for the future. Personal experience is to humans what predictive analytics is to software. The software solution recognizes certain patterns in these "experiences" and uses them to reach conclusions about future behavior and developments. Using this automatic classification, each letter is forwarded directly to the relevant department. There, the documents can be immediately processed and dealt with. This optimizes the entire handling process. Another advantage of this system is that it learns from its mistakes. If a document was incorrectly classified and then manually corrected, the system remembers this. The longer the system is in use, the more information the application collects and the more accurate the classification becomes.

Moreover, to an intelligent incoming mail classification system, it doesn’t matter whether the data that arrives at the company is structured, e.g. a completed online form, or unstructured, for instance in the form of an e-mail text. The system treats both types of mail equally, so that no customer or business case is at a disadvantage due to the communication form used.

Flexibility is one of the strengths of an intelligent system. Whether an e-mail, a scanned letter or the ever-present social media post - every piece of written information is analyzed, sorted and classified. Instead of fearing the new developments that the "Internet of Things" will bring in its wake, insurance companies are armed and prepared to see the future and big data as an opportunity and a positive challenge.

Straightforward Implementation

The initial configuration of automated incoming mail classification can be done with very little effort. After the integration into the company's own IT, the data sources can be linked by means of connectors. To make this process fast and seamless, the system offers more than 400 connectors for linkage to different data sources. Of course, these also include typical data sources such as network drives, Microsoft SharePoint, and a variety of ECM systems. Then the training begins. In training mode, the systems learn to sort already classified documents based on existing criteria. Thus, the "brain" of the system is trained and becomes more intelligent with each document it processes. The training phase is simple - pre-classified documents from the past days, weeks or months are used – the more input, the higher the learning success rate. The learning process itself requires only a few milliseconds per document. Should the system make an error, the manually corrected document is re-submitted to the system, which then saves the correct classification information.

Mastering the Challenges

One exciting challenge facing automated incoming mail classification is the variety of input items, which today encompasses far more than just letters in the mail. E-mails, often with attachments that consist of several pages, faxes and even posts in social media channels – the system has to be able to process and analyze all of these different sources. There is still the problem of the selection criteria used by the system to determine which department should receive which document. Using training documents which have already been classified, the system educates itself for the future. Because the system is self-learning, additional selection criteria can be added at any given time.

With their product Mindbreeze InSpire, the Austrian software manufacturer Mindbreeze has fulfilled all these requirements with one appliance. The solution is already being used successfully at major insurance companies and classifies documents on the basis of approximately 3,000 different characteristics queried per document.

Technology for the Future

It is not a question of if, but of when such systems will be standard features in every insurance company in order to optimize the workload. In light of the fact that the huge data quantities are only going to continue to increase, these applications are not a "nice-to-have" but rather a "must have". The many benefits of automated incoming mail classification speak for themselves and are proof that it is imperative to invest in the future. Besides the fact that the number of documents is steadily snowballing, the amount of time in which answers and responses are expected is decreasing. Customer satisfaction is critical for success in the insurance industry, and is likely to gain significance in the future. Automated incoming mail classification is today’s answer to the issues of increased efficiency and customer satisfaction.