Why one acceptance rate is the wrong tool
Every rule of thumb about wedding attendance is a single percentage applied to a single list. The trouble is that the list is not one population. A guest who can be home in bed by midnight is making a decision about one evening. A guest who has to buy a flight, book two hotel nights and take a day off work is making a decision about a few hundred dollars and their annual leave. Those are different questions, and they get different answers at very different rates.
Split the list by what attending actually costs the guest and the arithmetic changes. In the defaults on this page, 120 local guests at 85 percent contribute 102 people, while 20 distant guests at 40 percent contribute 8. Blend those into one rate and you get a number that is right on average and wrong for every planning decision that depends on it, because when the mix shifts — and it always shifts, as soon as you add or cut a block of people — the blended rate quietly stops applying.
The range matters more than the number
An expected headcount is a mean, and a mean without a spread is not a plan. Each invitation is closer to a coin weighted by that group's rate than to a fraction of a person, so the spread of the total follows from the group sizes and rates directly. That is what the plus-or-minus figure in the results is: two standard deviations either side of the mean, which covers roughly nineteen outcomes in twenty.
Two things follow from it. First, a narrow range means your uncertainty is genuinely small and you can commit to numbers. Second, the range is widest when rates sit near 50 percent, which is exactly the colleagues-and-acquaintances group — a small group that nonetheless does more than its share of the damage to your confidence. If your list leans that way, hold more contingency, or move that group to a later invitation round once you know where the count is landing.
Plus-ones are a separate count, not a multiplier
An open plus-one is an invitation to a person you have not identified. It should be counted as its own line rather than folded into the group rates, because its take-up rate is unrelated to the invitee's own acceptance rate: someone who is definitely coming may or may not bring anyone. Named couples are different. If you addressed the invitation to two people by name, that is two guests in whichever group they belong to, and it should not appear in the plus-one field at all.
The wedding guest guide covers this from the other side, which is worth reading if you are trying to decide who gets one. From a purely numerical standpoint, twenty-five open plus-ones at a 60 percent take-up is fifteen extra dinners, and at most per-head prices that is a meaningful line in the budget rather than a rounding error.
Venue minimums and why trimming stops working
A minimum headcount, or a food and beverage minimum expressed in dollars, changes the shape of the cost curve. Above the minimum, every guest you cut saves the per-head charge. Below it, every guest you cut saves nothing, because the venue bills the minimum either way. People discover this at the worst possible moment, usually after a difficult conversation about who has to come off the list.
The useful output is the gap. If your expected headcount is eleven below the minimum, you have eleven seats already paid for. Adding a guest costs nothing until you cross back over, and cutting one returns nothing. Some contracts allow the shortfall to be spent on other things instead, and some do not — that is a contract term and it varies, so read yours and ask before assuming either way. Once you have a headcount you trust, take it to the wedding budget calculator, where the per-head effect compounds through service charge and tax.
The number the caterer needs is not this one
Caterers work from a guaranteed count given a set number of days before the event, and that guarantee is a floor you pay for. This calculator gives you the number to plan with, not the number to guarantee with. Those should differ: the guarantee usually sits at or slightly below the expected figure, because late additions are almost always accommodated while late reductions are almost never refunded. How far below, and how late you can move it, is written in the contract rather than fixed by convention.
What to do with the outputs, in order: use the expected figure for the budget, the high end for the tent, seating layout and parking and restrooms, and the low end only to see what the venue minimum would cost you if the replies come in soft. Sizing physical capacity to the expected figure is how people end up with a room that works only if the mean is exactly right.
Questions people ask
What acceptance rate should I actually enter?
Start from the travel each group faces and adjust from what you know. Guests who can be home the same night accept far more readily than guests who need flights and hotel nights, and the gap between the two is usually wide. Rather than searching for an authoritative number, put in your first guess, then look at the range the calculator produces. If the range is tight enough that the decision you are making does not change across it, the exact rate did not matter. If it is not, that tells you which group to firm up first, usually by asking a few people directly before the invitations go out.
When should I re-run this?
Twice. Once before invitations, to size the list and check it against the venue minimum, and once as soon as roughly a fifth of the replies are in. That first tranche is informative because it is not random: it is weighted toward the people who were always coming, so it tells you most about the groups you were most confident in. What you are really looking for at that point is a group replying far below your assumption, because that is the one that will move the total, and it is still early enough to send a second round of invitations if you want to.
Does the range assume replies are independent?
It does, and that is the main limitation. Real guest lists come in clusters — a family decides together, a friend group coordinates, one couple deciding not to travel can take three others with them. Clustering makes the true spread wider than the figure shown here, not narrower. Treat the range as a floor on your uncertainty rather than a bound on it, and if your list is heavy with tight-knit groups travelling from one place, widen it yourself.
What happens if more people accept than I have seats for?
That is the case the high end of the range exists to warn you about. Physical capacity is not elastic in the way a budget is: a tent, a seating plan and a restroom count either accommodate the number or they do not, and fixing it late costs more than building the headroom in from the start. Size capacity on the high figure and cost on the expected figure. If the high figure will not fit the venue at all, the list is too long for the room regardless of what the mean says.
Should children be counted the same way as adults?
Not quite, and there are two reasons. Their acceptance is tied to a parent who has usually already decided, so the rate is high and the uncertainty is low. And they are frequently billed at a reduced rate or not at all, which means they affect the headcount and the seating without affecting the per-head cost the same way. This calculator keeps them as their own group so the acceptance behaviour is right; if they are billed differently, work out the adult cost here and add the children separately in the budget.