How many guests should we invite?

Let the budget decide the number, then divide it up. Across 281 couples who built a real list in English on ChungDoi, the typical list is 28 guests and the average is 46, pulled up by a handful of very large weddings (the biggest was 462). Those are the names entered into the system, so the true list is usually larger once both sets of parents add theirs.

Write one rule and apply it to everyone, rather than deciding guest by guest. The common rules are: anyone married, engaged or living with a partner; anyone in the wedding party; or anyone who will not know another guest. Whichever you pick, put it in writing before you start inviting, because the exceptions are what create arguments later.

If 200 people say yes, how many seats do we need?

More than 200. Across 12,243 RSVPs to English invitations, 95.8 percent said yes, and each yes averaged 1.57 people. That makes 200 confirmations closer to 310 seats. The multiplier varies enormously by culture: Vietnamese invitations average 1.19 people per yes, Spanish ones 1.91, so use the number for the crowd you actually invited.

When should the guest list be final?

Set the total 3 to 4 months out, collect names 2 to 3 months out, send the first wave 6 to 8 weeks out, invite from the B list 3 to 4 weeks out as declines come in, and give the venue a final headcount about a week before. The list will change several times, so start early and lock late.

How to Build a Wedding Guest List

Most couples sit down to write the guest list expecting one spreadsheet with two people filling it in. A week later it is obvious that it was never one list. There is your side and their side, and behind both of those there are the parents, who have their own people to invite and their own reasons. Every one of those lists lives with a different person, and only that person knows how the names are actually spelled.

The tool for this: you can build the list on the guest list tool, track who has replied, watch the headcount update itself, and export it to a spreadsheet whenever you need to hand it to a venue.

In short: decide the total from the budget first, then hand out quotas, never the other way around. Sort every name into A, B and C before you send anything. And remember that one "yes" is not one seat: on English invitations it averages 1.57 people, so 200 confirmations is closer to 310 chairs.

The list is really four lists

Even at a wedding where the couple hosts and decides, four sets of names show up:

Your people. School, university, work, the group chat. Only you know who is currently in and currently out.
Their people. The same, on the other side.
Your parents' asks. Relatives you have not seen since childhood, their oldest friends, sometimes colleagues. They are invited because of a relationship that predates you.
Their parents' asks. The same, and usually the list you have the least visibility into.

The reason this matters is not politeness, it is arithmetic. Categories 3 and 4 are the ones that grow when you are not looking, and they are the ones you cannot check for spelling, duplicates or plus-ones yourself. Ask for them in writing, early, with a number attached.

Start with the number, not the names

The venue's per-head cost sets the guest count. Everything else is a consequence of it. Work out what you can spend on catering, divide by the cost per person, and that is your ceiling. Then split the ceiling before anyone writes a single name:

Who	Quota	Who owns it
You	50	You
Your partner	50	Your partner
Your parents	40	Your parents
Their parents	40	Their parents

The split does not have to be equal, and it will not be. One family has more cousins; the other has more old friends. A 10 to 15 percent overshoot is normal and fine as long as the total holds. What is not fine is collecting 300 names first and then trying to cut back to 180, which is how people end up cutting friends to keep acquaintances.

What the lists actually look like

Some numbers from English-language invitations on ChungDoi, from couples who got far enough to build a real list:

The typical list is 28 guests, and the average is 46. The gap is a handful of very large weddings pulling the average up; the largest single list was 462 people.
37 percent of lists still have fewer than 10 names in them, which usually means the real list is sitting in a spreadsheet or a notes app somewhere and only part of it has been moved across.

Read the typical number as a starting point for the conversation with your families, not as a target. What it is genuinely useful for is calibration: if your list feels enormous, it is probably more ordinary than it feels.

Sort into A, B and C before you send anything

Not every name carries the same weight, and pretending otherwise is what makes cutting painful later.

A list. The wedding is not the wedding without them. Immediate family, closest friends, the people you would move the date for. Invite first, in whatever form feels most personal.
B list. Genuinely want them there, would not reshape the day around them. Current colleagues, good friends you see a few times a year. Digital invitation is completely fine.
C list. Invite if there is room. Former colleagues, distant relatives, friends you mostly keep up with online. Send this wave once the A and B replies are in.

Doing this at the start rather than at the end means that when the venue caps you at 120, you already know where the line falls and nobody has to argue about it at 11pm.

Write down the plus-one and kids rules

These are the two decisions that will otherwise be re-litigated for every single guest. Pick a rule, write it in one sentence, apply it to everybody:

Plus-ones: married, engaged, or living together is the most common line. "Anyone who will not know another guest" is the kindest. "Everyone gets one" is the most expensive.
Children: all welcome, family only, or none. Whatever you choose, say it on the invitation itself rather than hoping people infer it.

A written rule is also the only thing that lets you say no to an exception without it feeling personal.

Getting four people's lists into one place

The hard part is not writing names, it is merging four sources without typing all of them twice.

A shared spreadsheet. One file, one tab per person. Cheap, familiar, and works well if everyone involved is comfortable with it. The catch is the last step: you still have to move every row into whatever you are using to send invitations.

Reading names over the phone. Slow, but sometimes the only realistic option for a parent who does not want to be handed a link. Budget half an hour per 25 names, and expect to fix spellings later.

Bulk import. If the list already exists somewhere, paste it in or upload the file rather than retyping it. On ChungDoi, "add multiple guests" takes a pasted list (one name per line) or a spreadsheet upload, up to 1,000 people at once, and generates a personal invitation link for each name.

A co-hosting link. Send relatives a link that lets them add their own guests directly, with no account and no login. It is the only method that keeps parents' names out of your inbox entirely, and the one that saves the most time on lists over 100.

Whichever route you take, the destination is the same: every name on the list becomes its own invitation, addressed to that person, so nobody opens a card meant for someone else.

Catching duplicates before they get two invitations

Duplicates are close to guaranteed once four people contribute, because families overlap exactly where you cannot see it. Four ways to catch them:

Sort by normalized name (lowercase, accents stripped) so near-identical spellings land next to each other.
Match on phone number, which is faster and more reliable than matching on names.
Ask directly at the family meeting: "who is inviting Aunt Margaret?" Five minutes there saves an awkward phone call later.
Let the import screen do it. When you paste a list or upload a file, names already on the list are shown struck through and skipped automatically.

The timeline

3 to 4 months out: agree the total and hand out quotas.
2 to 3 months out: collect the actual names, sort into A, B and C.
6 to 8 weeks out: send the first wave to A and B.
3 to 4 weeks out: invite from the C list against the replies you have.
1 week out: final headcount to the venue.

One yes is not one seat

This is the arithmetic that catches people out. Across 12,243 RSVPs to English-language invitations, 95.8 percent said they were coming, which is high enough that you can plan against the replies rather than holding your breath. But each yes averaged 1.57 people, and it has been drifting up (1.52 last quarter, 1.59 this one).

So 200 confirmations is not 200 chairs, it is closer to 310. That is more than ten extra tables, and it is the single most common reason a venue has to set another one during the reception.

The multiplier is also very culture-specific, so do not borrow someone else's. On Vietnamese invitations it averages 1.19, because cards go to individuals. On Spanish ones it averages 1.91, because they go to households. If your list spans both, count those groups separately.

Keep the list afterwards

Do not delete it once the wedding is over. It is the basis of your thank-you notes, it records who gave what for when their own wedding comes around, and it quietly becomes the address book the two of you use for the next decade. Exporting it to a spreadsheet takes one click, and it is worth doing the week after.

Where these numbers come from

Figures come from ChungDoi invitations written in English, as of 15 August 2026: 281 lists with 10 or more guests, and 12,243 RSVP responses. Party size counts the person replying, and ignores obvious mis-entries above 10. These are guests entered into the system, not people who walked through the door, so a real list is usually longer. We re-run the numbers every quarter and update the article when they move.

Once the list is settled, the next question is which invitation to send it with. The current design ranking is a reasonable place to start, or browse all templates.