My Paperless Office, Part II
Until now, my implementation approach can be summed up succinctly as: "scan forward." I picked a date out of the air and started scanning (or throwing away) any piece of paper my hands come in contact with.
Today, I started the second phase of my paperless office implementation. I installed Subversion on a server and started getting serious about scanning in my existing records: tax returns, invoices, notes, corporate filings, corporate minutes, and assorted detritus cast off by a small business. Since this morning, I have filled up a 50 gallon trash can with shredded documents; I'm still working on filling up another one.
I can honestly say that the process of emptying out a 5 foot tall metal filing cabinet is very cathartic.
The work flow is pretty simple:
First, try to cull the documents before you scan. First go through your documents and try to discard as much as you can. Just ask yourself, questions like "will I need this in an audit? Will I ever need to retrieve this? Why am I saving this?" Sure, disk space is cheap, but looking back a year from now, will you really need a magazine clipping or old electricity bill?
Next, scan the documents, and rename them if they are important enough. The scanning software will automatically create a generic file name with an incremented number appended. For things like bills and statements from previous years, I just take them as they are, and drag them to the appropriate filing cabinet folder.
The question is, how should you organize your virtual filing cabinet? The choice is yours, but I settled on a scheme that organizes files by category, year, and subcategory. For example, a receipt for a $20 item would be placed in the receipts folder in the following directory hierarchy:
Biz --> FilingCabinet --> Accounting --> 2007 --> receipts
This is convenient primarily because my business uses the calendar year for accounting purposes. Should I get audited (or if my accountant has questions), I could just copy everything from 2007 downwards and give it to the auditor on a USB drive or CDR.
Next, of the documents you scan, decide what to through out, what to file away, and what to shred. Any personal information should go into the shredder, while any contracts or leases should go back into the filing cabinet. The rest should just be tossed or recycled.
Lastly, for each batch I will add the files and commit the changes to my subversion repository.
Here are my recommendations/observations/tips:
- Shredding takes up the most time. My SnapScan S510 will blast through a stack of documents, while hand feeding documents into my noisy shredder takes forever. As a result, I scan documents in batches, dumping the scanned documents into a box for later shredding.
- I also append dates to files in the format of YYYYMMDD. For example, a bill from OfficeMax, with a billing date of January 1, 2007, becomes: OfficeMax_20070101.PDF. When you pull up the directory listing, the files will be automatically sorted, and you can tell what you are looking at without opening it.
- For records like bank statements that have a low probability of being necessary (but required for record keeping), I just scan an entire year's worth into a single PDF. It simply isn't worth my time to scan each month independently and rename them.
- When you reconcile your "online" statements (everyone still reconciles monthly, right?), download the PDF and put them into your repository. At least twice I have lost access to historical statements online.
