Party headcount calculator
Estimate how many people will actually turn up to your party from the number you invited. Free, instant, and nothing you type leaves your device.
Count invitations sent, not people. A family of four is one invitation.
Changes the typical reply rate and how many people arrive per invitation.
Best guess
29to38people
From 30 invitations, expect roughly 18–24 to say yes, arriving with about 1.6 people each once partners, children and siblings are counted.
That is a spread of 9 people, which is the difference between catering comfortably and running out. No calculator can close that gap, because the answer is not a formula. It is whatever your guests would tell you if you asked them.
Reply rates here are rules of thumb, not measurements, and real ones vary a lot by how you invite people and how much notice you give. Nothing you type leaves this device.
Stop estimating. Let your guests confirm their own numbers.
An estimate is a guess about people who could just tell you. A party on MakeTheRSVP gives you one link to send: your guests answer in a tap with no account, they say how many they are bringing, and the total counts people rather than replies. Free for up to 25 guests, with no card.
How to use it
- Enter how many invitations you are sending. Invitations, not people: a family of four is one invitation.
- Pick the kind of party. It changes both the typical reply rate and how many people arrive per yes.
- Read the range, then look at the spread underneath it. That number is the one worth acting on.
Why the number of replies is not the headcount
This is the mistake that leaves hosts short. Eighteen people replying yes to a kids' party is not eighteen people at the door: it is eighteen children, plus the parents who stay, plus the younger siblings who come along because there was nobody to leave them with. The same gap shows up at a shower as partners, and at a wedding as plus-ones nobody confirmed.
Counting replies tells you how many people answered. Counting people tells you how much food to buy. They are different numbers, and only one of them is useful on the day.
When an estimate is genuinely fine
If the party is small, the venue is your house and the catering is a supermarket run, an estimate is fine and you do not need anything else. It stops being fine when a number has money attached to it: a caterer's per-head price, a venue minimum, a restaurant booking, or a cake that has to be ordered on Tuesday. That is the point at which guessing gets expensive and asking gets cheap.
Related free tools
Building the actual list of names? There is a tool that keeps it in your browser and exports it as CSV. Once you want real answers rather than an estimate, read what a kids' party is worth asking and how to chase the people who have not answered.
Questions about party headcounts
As a rough rule, expect 50-70% of invitations to a casual adult party to say yes, 60-80% for a kids' birthday, and 75-90% for a wedding, where people plan around the date months ahead. Then multiply by the number of people arriving per invitation, because a yes is rarely one person. Those are rules of thumb rather than measurements, and the honest answer is that the range is wide enough to matter.
It depends far more on how you asked than on the occasion. A link people can answer in one tap gets a much higher reply rate than a phone number to text or an email to reply to, because replying costs nothing and can be done immediately. Notice matters too: three to four weeks tends to get more answers than three months, when people put it off and forget.
Because a single number would be a false confidence. The spread between the low and high end is usually the difference between catering comfortably and running out, and no formula can close it. The only thing that closes it is asking the people who know.
Most hosts do, and it is reasonable when you are confident about the reply rate. It gets expensive when you are not: over-inviting a party with a hard capacity, like a restaurant table or a venue, turns a good reply rate into a problem rather than a relief. If capacity is fixed, collect real answers before you send the second wave.
It saves them in this browser on this device, so they survive a refresh, and it sends them nowhere. There is no account and no server call. Clearing your browser data clears them, and they will not appear on your phone.