Consider how daunting a task processing 40,000 document pages per month would be — a task that involves placing them online, and interpreting each piece of text and numeral so it’s searchable. Now imagine processing the same document load with the majority scribbled in hand written text, inks of different colors and sometimes with writing utensils as arbitrary as color crayons.

This was the task that faced Joel Perkins and Andrew Booth late last year as they toiled desperately to digitize Georgia’s Government Transparency & Campaign Finance Commission’s regular onslaught of paperwork — documents that consisted of financial disclosures and campaign contribution forms for public office elections.

As the consultants managing the Commission’s IT infrastructure and code, Perkins and Booth — who are CEOs for IT consulting companies Inserv360 and Jaxified LLC, respectively — were asked specifically to update the Commission’s eFiling system by December 31, 2013. The assignment allowed a mere two months to meet state legislation requirements of House Bill 143, a bill that required local candidates and elected officials to file paperwork with local filing entities, translating into a deluge of faxed handwritten documents.

Prior to the change, filing had been done through the Commission’s website, which involved a data entry process, a method that made online searchable documents simpler to produce.

"The challenge was now you're accepting hand written documents from people all across the state and you have to be able to read the different handwriting types,” Perkins said.

Though the two are both adept at, and in fact, enjoy solving complex problems, this IT challenge was difficult due to a tight budget and tighter deadlines.

“We like to look outside the box even if something is difficult and figure it out,” Perkins said. “The stress came in when we realized that OCRing these documents [making document data legible through Optical Character Recognition] was going to be far more difficult due to the different hurdles we were running into.”

With the clock ticking and pressure mounting, the two tried different open source program code and filtered through assortment after assortment of software vendors to solve the problem. Notwithstanding, results were poor and time kept shrinking. The remedy didn’t come until a final Google search led the two to Captricity, the Berkeley, Calif., based data capture firm that combines machine and human intelligence to create data.

“When Andrew found Captricity and ran that first test — and I remember we were sitting right across the desk from each other —  he just looked up and said, 'You won't believe this! It was literally 100 percent readability,” Perkins said.

According to a Caprticity blog post, since Jan. 1, the Commission has received more than 6,800 faxes, many with 10 or more pages. Current estimates expect 40,000 pages to be processed per month during the Commission’s seven annual filing periods in 2014. Pages are faxed to Captricity’s platform, and then to the Commission’s eFiling system and database.

The timely solution represents an obvious win for the Commission, but equally so, a win for Captricity, a civic tech startup that was part of Code for America’s 2012 Accelerator program. Georgia is now the second major jurisdiction announced as a subscriber to the civic tech company’s platform that can convert information from email attachments, image scans, faxes and mobile phone photos to legible data.

Prior to Georgia's deployment, Captricity served the Food and Drug Administration as it processed a massive backlog of paperwork — a project that’s brought the backlog down to zero and is ongoing, said CEO and Founder of Captricity, Kuang Chen.

Continuing its work, on Tuesday the company announced the launch of an updated version of its platform — which Georgia used in its beta form — called Captricity for Enterprise. The update builds upon the former version first developed through Chen’s dissertation work at the University of California at Berkeley, where he received his Ph.D. in computer science. Specific enhancements include a new advanced programming interface for developers and additional methods to extract data from multiple sources.

"Now we've taken on the challenge of enterprise-level data capture head on, so the improvements include what we like to call ‘multi-channel input,’ which is essentially to be able to take documents from all sorts of ways, including scanning it with your scanner in your office, but also grabbing email attachments that are documents or grabbing right from an eFax line,” Chen said.

As momentum for Captricity builds, the company, like many other civic tech startups, is hoping to break into the often tumultuous world of government contracts and their associated labyrinth of state and federal procurement processes. As a start up, Captricity is an anomaly, it being a company that’s collaborated with both state and federal jurisdictions, which is a territory typically reserved for multimillion dollar IT firms.

Speaking for the Commission, Perkins conjectured a major advantage of Captricity and civic startups like them is their direct relationship with clients and a heightened degree of creativity and innovation.

“I think when you deal with the smarter startups, they’re hungry, they're breaking the rules, the customer service is outstanding and this because they want to make that connection,” Perkins said. “They want to get their product out there and they want it to work.”

Government Technology Staff Writer Jason Shueh Jason Shueh  |  Staff Writer

Jason Shueh is a staff writer for Government Technology magazine. His articles and writing have covered numerous subjects, from minute happenings to massive trends. Born in the San Francisco Bay Area, Shueh grew up in the east bay and Napa Valley, where his family is based. His writing has been published previously in the Tahoe Daily Tribune, Amazon Publishing, Bike Magazine, Diablo Magazine, The Sierra Sun, Nevada Appeal, The Union and the North Lake Tahoe Bonanza.