When not to buy a scanner

This site is about scanning hardware, so it should be able to say when the hardware is the wrong answer. A dedicated scanner solves one specific problem — the per-item effort of capturing a lot of paper — and if that is not your problem, the device will sit unused and the paper will pile up exactly as before. There are four situations where buying one is a mistake, and they cover a large share of people who are thinking about it.

When most of your receipts were never on paper

A scanner cannot help with a document that arrived as an email. If the majority of your purchases produce a confirmation in an inbox rather than a slip in a hand, then your capture problem is a filing problem and no hardware touches it. Buying a scanner in this situation produces a device used a handful of times a year, and a pile of digital receipts still sitting in an inbox.

The test is easy: look at a recent month and count. If paper is the minority, spend your effort on the digital side and handle the occasional slip with a phone.

When the volume is genuinely low

A feeder’s whole value is unattended batch throughput. If your paper arrives at a rate of a few items a week, you will never load a tray — you will scan them one at a time, which is the flatbed and camera use case, and a camera is already in your pocket.

There is a specific trap here. Low volume plus a dedicated device produces the worst of both: the device is not worth the desk space, so it gets put in a cupboard, and a scanner in a cupboard captures nothing. Meanwhile the phone you would have used is always present. The friction argument matters more than the specification, as the multifunction printer as a receipt scanner makes in a different form.

When the real problem is the habit, not the equipment

This is the big one and the hardest to admit. If there is a pile, the question is what stopped you capturing each item when it arrived — and the answer is almost never that the scanning step was too slow. It is more often that there was no established destination, no obvious moment for it, or no confidence that the scan would be findable afterwards.

A faster scanner does not fix any of those. It makes the fast part faster and leaves the parts that were actually blocking you untouched, which is why people buy a device, digitise a backlog in a burst of enthusiasm, and then stop. If you are not currently capturing anything with the phone in your pocket, buying a scanner is a bet that you will start once the equipment is better, and it is a bet that mostly loses.

The honest sequence is: build the habit with the phone for a month, and then buy hardware if and only if the capture step turned out to be the bottleneck. That way you are buying a solution to a problem you have measured rather than one you have assumed.

When a phone is genuinely sufficient

For a single receipt, a phone camera is competitive with a dedicated scanner on everything except throughput. It has no length limit in the mechanical sense, no feed path to jam, nothing that wears out, and no driver to expire. It is also better than a scanner at one specific thing: because nothing touches the paper, it can capture items that would be destroyed by rollers — torn slips, fragile paper, anything with a staple in it.

The place it is weaker is repeatability. There is no fixed distance to the page and no controlled lighting, so results vary with how you hold it and what the room is doing. A fixed arrangement — the phone in a stand, a consistent background, a light off to one side — removes most of that variability, and the shallow-angle lighting can actually recover faint print that a scanner’s straight-on illumination washes out, as scanning faded thermal paper describes.

When buying is the right answer

To be clear about the other side, because the four cases above are not universal. Hardware earns its place when several of these hold: paper arrives faster than you would photograph it individually; the items are reasonably uniform; you already capture reliably and the per-item effort is the thing limiting you; you have a backlog large enough that batch feeding is qualitatively different from one-at-a-time; and the device would live within arm’s reach of where you sit.

If that describes you, the useful next step is establishing what a device would need to handle by running your own worst paper through whatever you already own: testing a scanner with your own paper. And if the backlog is the whole problem rather than the ongoing stream, outsourcing the backlog alone may be better value than buying a device to do it once — with the caveats in what you ship to a scanning service.

The purchase that most often disappoints

A capable device that lives somewhere you are not. Every other mistake on this list is recoverable; this one produces a scanner that works perfectly and captures nothing, and it is the most common outcome of an otherwise well-researched purchase. Placement is a bigger determinant of how much paper actually gets captured than any figure on the specification sheet, and it is the variable nobody researches.

Before buying, decide precisely where the device will sit, and check that the place has a socket, a data connection if it needs one, room for the paper going in and coming out, and is somewhere you sit anyway. If you cannot answer that, the honest answer is to wait, use a phone, and see whether the problem you thought you had is the problem you have. And if you do buy, buy for longevity rather than speed, because a modest device that still works after two operating system changes captures more paper than a fast one that stops: the driver is the part that expires.

None of this bears on whether a digital capture is an acceptable substitute for a receipt, or how long the paper has to exist alongside it. Those requirements vary by country and by the kind of entity keeping the records, and they are questions for your own tax authority or an accountant in your jurisdiction — worth settling before you decide the phone photograph is enough.