PaperStackThe instinct with scanning is to slow down and do it properly: flatbed, lid closed, one page at a time. That instinct is right for some documents and overkill for most of them. Here's how to tell which is which, without turning every scan into a production.
A flatbed gives you the best scan you can get
If you're comparing options, a flatbed scanner wins. The page lies flat against the glass under even, controlled lighting, with nothing feeding or gripping it that could skew the image or crop an edge. A sheet feeder and a phone camera both introduce something a flatbed doesn't: paper moving through rollers that can tilt it slightly, or hand-held lighting that varies from one shot to the next. For a document where every detail on the page has to be legible, a flatbed is the right tool.
Most of what you scan doesn't need that
A flatbed is also the slowest option, one page at a time, lid up and down for every sheet. Running an entire week's mail through it that way is how scanning stops happening at all. For day-to-day paper, favor whatever gets the stack digitized quickly:
- A sheet feeder or ADF, if your scanner has one, for anything more than a couple of pages. It's faster than a flatbed by a wide margin and plenty sharp for reading and searching.
- A phone scanning app for a single receipt or letter when you don't want to walk over to a scanner at all. Good enough lighting and a steady hand get you a scan you can read back later.
Delivery notes, routine mail, most receipts: none of it needs archival quality, it needs to be searchable and legible. Treating a flatbed as the only acceptable way to scan just means you scan less.
Save the flatbed for when you actually need it
The reason a quick, rougher scan is safe as your default is the same reason it's fine to keep the paper after scanning it: the original is still in the box. If a document later turns out to matter more than expected, or a scan comes out harder to read than you'd like, pull the original and redo it on the flatbed then. That way the decision about scan quality only has to happen once, for the one document that actually needs it, instead of upfront and under time pressure for every sheet in the stack.
Getting double-sided scans out of a single-sided scanner
If your scanner or feeder only does one side at a time, you can still get a double-sided result without scanning page by page. Scan every front page first, in order. Then flip the whole stack over as one block, not sheet by sheet, and scan every back page onto the end of the same file. What you get is one PDF with all the fronts in order, followed by all the backs, but in reverse, since flipping the stack as a block reverses their order.
That file isn't in reading order yet, but you don't have to sort it by hand. PaperStack's free duplex scan sorter takes that exact file and interleaves it back into front, back, front, back order, entirely in your browser. It also catches the common case where the last sheet was single-sided and its "back" is just a blank page, and offers to drop it before you download the result.
The short version
Use a flatbed when a document genuinely needs the best scan you can produce. Use whatever's fastest, a feeder or your phone, for everything else. And if a rougher scan ever turns out not to be good enough, that's what the box is for.