From a box of business cards to a CRM table in one evening

Business cards have a short shelf life. A week after the trade show you still remember who's who; a month later you're holding a card with a stranger's name on it, no idea why you kept it. So most business cards end up in a drawer, because retyping a hundred contacts into a CRM is a half-day chore nobody ever finds time for. And yet the most valuable piece of information on a business card is often not the printed part — it's your pencil note in the margin: “follow up in September,” “interested in training.” No automatic card scanner reads that part. AI that can read handwriting does.

A typical scenario

Sales rep Ray comes back from a three-day industry trade show with 94 business cards. About a third of them have a note penciled on. The usual routine: the cards sit on his desk for three weeks, he retypes the twenty most promising ones, and tosses the rest — so seventy contacts the booth paid for go to waste.

The new routine takes him one evening. He photographs the cards on his phone into a folder, kicks off processing, and goes make dinner. He comes back to a table with 94 rows: name, company, title, email, phone, the margin note transcribed into a “context” column, two sentences of research per company, and a draft follow-up email for every contact. He spends two hours going through the drafts, deletes forty as not worth pursuing, edits and sends fifty-four himself. By Thursday he has three meetings booked. The difference from last year isn't the quality of the emails — it's that they got written at all.

How to do it

1. Scan the cards into a single folder. Phone photos are enough — card by card, or ten at a time laid out on a table. A scanning app will crop and straighten the shot, which improves read accuracy (scanning with your phone). Move the photos into one folder on your computer and don't sort them — that's the whole point.
2. Point Claude Cowork at the folder. Cowork is a desktop-app mode for working with files without writing code: you pick a folder, and Claude can read documents, images, and scans inside it. It only has access to the folders you explicitly give it — and you can revoke that anytime. Contacts are personal data, so do this on a paid or business account with contractual data protection, not a free chat, and put nothing but business cards in that folder.
3. Ask for extraction that includes the handwriting. From each card you want name, company, title, email, and phone — plus a “context” column transcribing whatever's handwritten on it. Explicitly ask that illegible bits be flagged as uncertain rather than guessed at. The output should be a table, in CSV or Google Sheets, with columns named the way your CRM's import expects.
4. Have it add a mini-research pass on the companies. A second pass: two or three sentences per company — what they do, roughly how big, anything recent. That produces a “prep” column so you're not going into a call blind. Ask for sources on the research so you can tell what's verified from what's a guess.
5. Have it draft follow-up emails, not send them. One short email per contact, grounded in the card's note and the context of the meeting — so “we talked about training for your team” instead of a generic “pleasure meeting you.” A hard rule applies here without exception: drafts, never a mass send. Every email gets reviewed, edited, and sent by a human.
6. Import and close the loop. Upload the table into your CRM, and for each contact set a follow-up task with a date drawn from the note (“follow up in September” is a date, not a feeling). Delete the card photos once imported — you don't need them anymore, and they're someone else's personal data.

The best tools

• Claude and Claude Cowork — goes through a folder of scans, reads printed text and handwriting, and produces a table plus draft emails without you writing a single line of code.
• Your phone's scanning app — cropping, straightening, and batch capture; the better the source image, the fewer corrections in the table.
• Google Sheets or Excel — the middle step where you check and tidy up column headers.
• Your CRM — almost all of them support CSV import; check ahead of time what column names it expects.

What you get out of it

• Time: one evening, most of which isn't spent by you, instead of half a day of retyping.
• Return on investment: a trade show booth costs money. A contact left in a drawer wastes that spend — this process recovers it.
• Quality of outreach: an email that references a specific line from your conversation gets a completely different response rate than a mass template.
• Speed: a follow-up sent within a week works. A month later it's a cold contact.

Pro tip

Write the note on the business card on the spot, and write it so it still makes sense a month later — not “interesting,” but “looking for Excel training for 12 people, deciding by October.” AI will read it, but it won't invent what you should have remembered. And one legal note worth not skipping: following up after meeting someone in person at a trade show is a legitimate business contact, but the rules on unsolicited commercial communication still apply — write to people individually, not in bulk, say who you are and where you met, and remove anyone who asks on the first request. The difference between building a relationship and spam is whether the other person can tell you're writing to them specifically.