THE CHALLENGE: SORTING OR IDENTIFYING INFORMATION

Imagine if you could quickly and accurately sort, identify and redirect Information, even if you’re not a subject expert.

Many businesses scan documents that contain information they need –whether it is to archive documents, or to provide immediate access to data that employees need. They most likely have systems in place that have specific routines to sort, identify, classify, or categorize unstructured information. Organized storage and retrieval of these documents is essential. Most of these systems require a lot of manual work up-front, and continual refinements. The time spent entering the data needed can be staggering, yet less than 60% of the information that passes through is accurate. Scanning and OCR capabilities are restrictive even though they require extensive word lists, they can’t handle simple typographical errors.

THE SOLUTION: CONTENT ANALYST

Content Analyst looks at everything it is fed based on context, so it can accurately send information to the correct location regardless of title, or word repetition. Because our software is adaptive and learns as it performs, it can correctly assign a location simply because it “remembers” a document type it has seen before.

Solutions in Action:
Scaling a daily mountain of mail
Imagine you are an agency that has decided to implement that is implementing Content Analyst as a “front end” to your email receiving system.

Content Analyst will analyze every incoming email, and scanned postal mail, and figure out which department should receive it – and in some cases, which other agency should handle the query.

Life before Content Analyst at this customer
Your agency receives 2000 queries a day on average. You had two people dedicated full-time to reading mail and email and forwarding them to the right departments. That translated to 1000 documents per reviewer, or 120 an hour. As one might imagine with this type of volume, there were numerous mistakes, and the workers found the job overwhelming and tedious. To add to the problem, you found that news releases highlighting your involvement in a given “hot” topic with the public could double the amount of correspondence you received overnight. It could take a week to sort through the backlog – meaning you now had a second round of correspondence, complaining about their lack of response to the first one. Agencies requiring critical information didn’t always get it on time. And you were being buried alive in an avalanche of paper.

Implementing Content Analyst
You decided on an integration strategy (using one of Content Analyst’s partners). The process was actually very simple. To get Content Analyst started, they took 20 or so representative “sample” documents for each type of recipient, and let Content Analyst read these documents. What Content Analyst does is look for patterns, concepts, and associations. After as few as 20 documents, it “knows” the kinds of queries that a particular department should receive. It turns this knowledge into an active vector space (like a sample library) and can then compare any new documents it receives against this library – or libraries, since you have hundreds of departments.

Life after Content Analyst
You anticipate a savings of over $120K in overhead costs right away, in addition to solving the problem of handling backlogs. You look to save another $60K a year in overtime expenses as your correspondence backlog evaporates, eliminating 2nd attempts/requests from customers. As an added bonus, your customer approval ratings have improved.