Digitizing Business Cards Proves Easy With Plustek Scanner
Piles of business cards can be rapidly turned into useful contact data-assuming their artwork is not too fancy. Even then, there are work-arounds.
Piles of business cards can be rapidly turned into useful contact data-assuming their artwork is not too fancy. Even then, there are work-arounds.Going paperless is one thing, but what about those little pieces of paper that people give you? If you're like me, you end the day at a conference, in your hotel room, at your laptop, going through the business cards that people have handed you, trying to decide which are worth the drudgery of entering their information into your contact file. Such is SMB life without an admin.
Business card readers can help, such as the recently announced $129 MobileOffice S800 from Plustek Technology Inc. Powered by its USB connection, it's only 2x4x1 inches (the coiled USB cord took up more room) and ran fine off a laptop.
I found that it did cut the drudgery almost to nothing, but you did still have to edit the results. Character recognition was generally good, but it seems that every card literally uses a different font. Therefore, as soon as you got complacent it would render a lower-case L as an upper-case I and wreck an e-mail address.
After scanning a card the interface presents you with a monochrome picture of the card, and a list of the data fields it extracted via OCR and AI. (These include company name, the first and last name of the person, their title, address elements, one fax and two phone numbers, Skype, e-mail, URL, and any note you attached.) You can edit it or just feed in the next card.
It was very good at picking out phone/fax numbers, e-mail addresses, and URLs. (Of course, those are not free-form text.) As for the rest of the text, it was fairly good at picking out names and business titles. However, it sometimes assumed that the subject's first and middle name was the first name, and the professional title was the last name.
As for the corporate names, these were often rendered with cutesy fonts and logos, giving the software more trouble than any other data field. Sometimes it mistook the "We're the Best" promotional text for the name. (Remarkably, the rest of time it was smart enough to ignore that text.) But at least a third of the time the software was stumped and made no entry into the corporate name field.
But that actually is where things got fun, because you could select a field on the card image and drag it to the data field and drop it. The software would then perform OCR on the selected area, and often read it correctly. This lets you pull the right text out of a jumble.
But there were some things that defied OCR, especially hand-drawn logos and off-beat fonts. Some color schemes didn't help: yellow on white, and white on light blue were especially problematic. In other words, a small amount of manual input proved unavoidable.
Once the data was entered, it could export it to a long list of personal organizer software, plus Outlook, Excel, and comma-separated-value text files.
On the whole, it was easy to set up and use. The scanning and reading almost kept up with my fumbling efforts to sort the cards and feed them in, and together took about ten seconds per card. Editing the data via drag-drop was actually fun. It lent itself to processing all my accumulated cards, rather than discarding most. And that's a good thing-you never know what bit of information will later prove useful.
But what struck me was that business cards ultimately represent an effort to escape digital conformity by conveying distinction and personality within a 3.5x2 inch format. After a while I found myself feeling guilty about what I was doing, since it rendered the people behind the cards into bland data, despite their efforts to escape that fate. When a card truly stumped the software I felt like applauding.
But when it was over I had their cards-and their data. It was the best of both worlds.