Why do bigger tags not always read better?
Because a phone excites a small region and area outside it contributes nothing. Past roughly card size the extra surface adds no coupling, while it does add thickness, cost and sensitivity to whatever the tag is mounted on. The curve flattens well before tags stop growing.
The curve flattens
Going from a very small tag to a card sized one is a large improvement, and it is the improvement people generalise from.
Going from card sized to twice that is not. The gain shrinks with each step, and beyond a point it is not measurable with a phone.
So the intuition that more area means more range is correct at small sizes and misleading at large ones, which is exactly the range where people start paying more for it.
The reader sets the ceiling
A phone coil is small and its field is concentrated around it, falling away quickly with distance.
A tag much larger than that coil has most of its area outside the useful field. Those turns enclose space the phone never fills, so they contribute nothing to the energy collected.
With a dedicated reader carrying a large antenna the answer changes, because the field extends over a larger region and a large tag can use its area. This is why industrial deployments and phone deployments have different sizing rules, and why a specification borrowed from one fails in the other.
Where bigger is actively worse
Detuning. A larger antenna covers more of whatever it is stuck to, so a large tag on a partly conductive surface is disturbed more than a small one in the same place.
Thickness. Bigger tags are often more robustly built, and every added layer holds the antenna further from the reader.
Placement. A large tag has to go where there is room, which is often not where people naturally present a phone. A small tag at the right spot beats a large tag at the wrong one.
And aiming. A large tag with no marked target invites a vague approach, while a small tag with a printed symbol gets a deliberate touch. The gesture is worth more than the area.
The practical sweet spot
For phone reading, something in the region of a card is where the useful gains stop. That is why cards and card sized labels dominate deployments aimed at the public.
Below that, size buys real reliability and is worth paying for. Above it, spend the money on the mounting, the surface treatment and a visible target instead.
And if the requirement genuinely exceeds what a card achieves with a phone, no tag size closes the gap. That is a reader problem, and it is solved by changing the reader rather than the tag.
Frequently asked questions
Is there a size beyond which nothing improves?
Not a single number, since it depends on the phone. The practical observation is that gains become hard to measure once the tag is comfortably larger than the phone's coil.
Why do industrial tags come so large?
Because they are read by dedicated readers with large antennas, where the extra area does contribute. Sizing rules from that world do not transfer to phones.
Should I buy the biggest tag that fits?
Up to about card size, yes. Beyond that, put the effort into placement, surface and a printed target instead.
Does a bigger tag help on metal?
Not by itself, and it can suffer more. What works on metal is a tag built with an isolating layer, at whatever size fits.