How do you plan an NFC deployment from scratch?
Backwards from the gesture. Who taps, where they are standing, what happens next, and what happens when it fails. Every hardware question is answered by those four, and every deployment that starts with the tag arrives at them late and expensively.
The order that works
The gesture first, then the object, then the surface, then the tag. The tag is the last decision and the cheapest one.
Working in that order means every technical choice has a reason behind it that someone can restate. Working the other way means choosing a tag and then discovering what it has to survive.
It also means the expensive decisions are made while they are still cheap to change.
The four questions
Who taps? Staff who will learn it, or members of the public meeting it once. That single answer sets how much margin the design needs.
Where are they standing? A tap needs somebody to pause, so the deployment can only exist where a pause already happens.
What happens next? A page, an action, a record. And on which platforms, since they differ.
What happens when it fails? Because a proportion of taps will, and a deployment with no answer excludes those people silently.
What the answers decide
Public taps mean a large tag, a printed target and a web address. Staff taps allow smaller tags and more specific content.
The location decides the surface, which decides whether you need on metal construction, which is the largest single line in most tag budgets.
The next step decides whether the tag holds a link or content, which decides the memory and therefore the chip.
And the failure answer decides the printed fallback, which is cheapest to add before anything is designed.
The usual sequence, and why it fails
The usual sequence is: buy tags, write a URL, stick them up, discover the surface is metal, discover half the audience sees nothing, discover the campaign address is now dead.
Each of those is a decision that was available at the start and was made by default.
The correction is not more technology. It is spending an hour on the four questions before spending anything at all.
And writing the answers down, because the person installing the tags in three months will not be in the room where they were decided.
Frequently asked questions
What should I decide first?
Who taps and where they are standing. Every hardware question follows from those two, and neither costs anything to answer.
When do I choose the tag?
Last. It is the cheapest decision and the one most constrained by everything above it.
Do I need a pilot?
For anything beyond a handful of tags, yes. Ten tags answer questions that a thousand would answer expensively.
What is the commonest planning mistake?
Starting from the tag. It forces every other decision to accommodate a choice made with no information.