BPO Showcase Videos

Modules designed for BPOs – A series of short and to the point showcase videos

There is no such thing as a standard way of handling documents and related business processes. Although there are similarities, every organization adopted some unique methods and practices.

A BPO taking over such processes, needs flexible tools that are easy to configure and maintain and are affordable.

Below showcase videos are inspired on real implementations. Some tough and unusual requests like separation on last page to show the flexibility of the tools.

All modules used in the videos were designed for Kofax Express, a production scan software with integrated VRS, searchable PDF generator, bar code reader and unlimited page count.

For confidentiality reasons, the documents used in these videos have been adjusted and are not the real documents used by our customers. Videos are short and to the point and only take a few minutes to watch.

Extracting floating loan account numbers from attorney letters

These letters are related to mortgages. Because the letters are sent from lawyers all over the country, the layout varies a lot. The data is there, but it’s just all over the place. Traditional zonal extraction is not going to work.

We will use MetaTool to read the full text of each letter and then find the account number by looking for a pattern. We first show the end-result. Then we setup the job from scratch.

We show the configuration step by step and in a few minutes you’ll know how to use MetaTool to extract floating data. You can use the same approach to extract dates, customer codes, invoice numbers etc.

Any data with some kind of pattern can be extracted regardless where it is positioned on the page or in the document.

Working with challenging background patterns on value paper

Fully automated import, processing and export with AutoBites and Kofax Express

In this CaptureBites showcase we will automatically extract the unique number from a set of Pfizer shares with MetaTool and store the shares in full color PDF.

The blue patterned background makes OCR challenging. To make this work, we will use electronic drop out to suppress the blue background color.

The video also shows how to configure this job in a very concise way.

If the images are delivered on DVD or a thumb drive, we also show how to automate the import, processing and export with AutoBites. We show straight through processing of correctly extracted documents. Incorrect extractions are sent to an exception station and don’t hold up the correct documents.

Keyword document separation – property inspection reports

Document separation on last page

Sticking bar code labels or inserting separator sheets are both popular document separation methods. But what if these methods are not an option. For example when there is no space for a bar code label. Or to minimize doc prep and avoid inserting separator sheets.

We will use MetaTool to separate the documents based on words that only occur on the first page of each report.

Alternatively you can use words or a bar code on the last page of a document to separate on last page.

We’ll show how all this works with property inspection reports. The title of these reports only appears on the first page. And the keywords in the title make good separator words. The system can be configured to use back-up separator words. If one fails, the other can be used to trigger a separation point after all.

If the first page doesn’t contain any good unique elements, we can also split on the last page of the report and look for keywords there. For example the property reports always end with the invoice and we can use the word invoice on the last page to trigger separation.

Automatic indexing with OCR – Introduction to MetaTool

Kofax Express can be used to index data with OCR while you’re reviewing each document. However fully automatic OCR is not available.

Documents with a static layout like the inspection reports in the example used in the video have key reference data in fixed positions.

With MetaTool, you define OCR zones just once in your scan job. Then, you scan batches of documents and the data will be extracted automatically. If parts of the text are damaged, the OCR technology may not read the data. MetaTool will only “then” show a validation screen to correct the data.

Floating data extraction is covered in the first showcase video on this page.