How many spare badges should you prepare?
Enough to cover five separate sources of demand, rather than one percentage applied to the headcount. Counting them separately gives a number you can defend to a budget holder, and one you can refine after each event from what you actually used.
Five sources of demand
Dead on arrival. A small proportion of any batch never worked. Verifying before the event turns these into a known number rather than a surprise.
Failures during the event. Creased badges, mostly, and the rate rises with the length of the event and the badge construction.
Lost badges. People lose them, particularly at multi day events with an evening in between.
Name and detail changes. Substitutions on the day, misspellings caught at the desk, companies sending someone else.
Walk ins. Often the largest and the most commercially important, and entirely absent from a pre printed stack.
Measure rather than guess
Each source has a different rate and they do not scale together. Walk ins scale with the event's openness, failures with the badge and the duration, losses with the number of days.
So a single percentage of the headcount is a guess dressed as a calculation, and it is usually wrong in both directions at once.
Count each category separately at the event, in the replacement log. Five numbers instead of one.
Next time, those five numbers are your estimate, and they improve every event rather than staying a rule of thumb.
Spares are the cheap line
A blank badge is a small cost against the consequences of running out: a person turned away, a queue, or a member of staff improvising at a door.
Where badges are generic and assigned at the desk, spares are not even spares. They are the same stack, and running out means running out of badges rather than running out of the right badges.
That is the strongest argument for the generic approach: it converts five separate estimation problems into one inventory number.
Where badges are pre printed with names, spares must be blank ones plus a way to produce a printed face on site, which is a different and larger requirement.
Managing the stack
Verify every spare before the event with the rest of the batch, so the replacement bench is not the place you discover a dead badge.
Keep them at the replacement bench rather than at the doors, so replacement happens away from a queue.
Keep them from the same batch as the main run, since a top up ordered later may be a different product with the same name.
And count what comes back. Unused spares are the cheapest possible outcome and they tell you what to order next time.
Frequently asked questions
Is a percentage of headcount a reasonable rule?
It is a guess dressed as a calculation, because the five sources scale differently. Counting them separately gives a number you can defend.
What if I run out during the event?
Someone is turned away or waits. That is why spares are the cheap line and why generic badges, which are all spares, are easier to size.
Should spares be pre printed?
If badges carry printed names, you need blanks plus on site printing. That requirement is the hidden cost of personalised badge faces.
Do I verify the spares too?
Yes, with the main batch. The replacement bench is the worst place to discover a badge that never worked.