The guest list is the only part of wedding planning where every decision has a person attached to it, and where the person can eventually find out.
Cut by category, never by name. A category is a policy you can say out loud without embarrassment. A name is a verdict you cannot defend to anybody, including yourself at 2am. "We are not inviting coworkers" is a rule. "We are not inviting Sarah" is an insult, and it is one that travels.
Every technique below is a way of turning a painful individual decision into a boring general one.
Why the name-by-name version goes wrong
When you cut person by person, three things happen, and all three are bad.
You are forced to rank your affections, which is not a thing anybody can do accurately or comfortably, and you will get it wrong in ways that surprise you later.
You lose the ability to explain yourself. If somebody asks why they were not invited, and the honest answer is "we went through and picked," there is no version of that sentence that lands well.
And it leaks. Guest lists always leak, usually through a parent, usually at a family event, usually eleven months later. A policy survives leaking. A ranking does not.
The category approach protects everyone, including you, because it makes the answer to "why not me" a structural fact rather than a personal one.
The number comes before the names
Do not start with a list and shrink it. Start with a number and fill it, because a list you have written down has already created obligations in your own head.
Your number is the lower of two things: what the venue physically holds, and what you can afford. The second one is the one people skip. Take your catering, bar, rentals, favors, stationery and seating chart, divide by headcount, and you will land somewhere between $120 and $250 a guest fully loaded. At $180, a list of 160 rather than 140 is $3,600, which is a photographer upgrade or the entire contingency line.
That framing matters because it converts a social problem into an arithmetic one. You are not deciding whether you love these twenty people. You are deciding whether these twenty people are worth $3,600 more than the alternative uses of that money. Sometimes the answer is yes. It is a much easier question to answer honestly.
The category menu
Pick your policies before you look at a single name, write them down, and apply them without exceptions. Exceptions are what destroy a policy, because one exception makes every other exclusion personal again.
- No coworkers, or current team only, or nobody you would not invite if you left the job tomorrow.
- The one-year rule. Nobody you have not spoken to, by choice, in the last twelve months.
- The met-them rule. Nobody neither of you has met in person.
- Children, all or none, or all under-tens excluded, or immediate family children only.
- Plus-ones by relationship status rather than by person, covered properly below.
- Second cousins and beyond excluded, applied to both families identically.
- Parents' friends capped at a named number rather than left open.
- No reciprocal invitations. You do not owe an invitation because you attended their wedding in 2019.
That last one deserves a moment. Reciprocity feels like an obligation and is not one. Weddings are not a debt system, and the people who understand this best are the ones who have already been married.
Apply each rule to both sides symmetrically. Asymmetric rules are the single most reliable source of family conflict in this entire process, because they are visible and they look like a judgment about whose family matters.
The four lists and the allotment
Build four separate lists: yours, your partner's, and one for each set of parents. Keep them separate, because a merged list makes it impossible to see who is driving the number up.
Then allot. Two common splits, and the important thing is to choose one and say it out loud rather than let it happen by drift.
The thirds split gives roughly equal shares to you as a couple and to each family. At 140 that is about 47 each. The modern split gives the couple half and each family a quarter, so 70 for the two of you and 35 each. The second is more common when the couple is paying, and the first is more common when families are contributing.
Whichever you choose, give parents a number rather than a veto. A number lets them make their own painful decisions inside their own family, which is both fairer and considerably less work for you.
We have landed on 140 total, which works out at 35 for your side. That is yours to spend however you like, and you know your family better than we do. Could you send us your 35 by the fifteenth, with names and mailing addresses? If you need 38 we will find it somewhere, but we cannot do 60.
Naming both the number and the flexibility is deliberate. A hard number invites a fight. A number with a stated three-person tolerance invites a negotiation you have already won.
If a parent comes back over, do not cut their list yourself. Send it back:
This came to 52 and we have room for 35, so I have not touched it because they are your people and I would get it wrong. Could you take it down to 35 and send it back? If there are two or three you genuinely cannot cut, flag them and we will look at the whole thing together.
Plus-ones
The hardest category, because the fair-sounding answers are expensive and the cheap answers sound unkind.
The rule that holds up best is relationship-based rather than person-based: plus-ones for anyone married, engaged, living together, or in a relationship you could name. Everyone else attends without one. It is defensible, it is applied identically to everybody, and nobody has to be told they specifically were judged.
State it on the invitation itself, because ambiguity is what produces the awkward phone call:
We have reserved 2 seats in your honour. Please let us know by the third of April if you will be joining us.
Naming the number on the card ends the question before it is asked. A card addressed to one person and an RSVP with room for two is an invitation to a misunderstanding.
If somebody asks for one anyway, the category is your answer, and it should be delivered warmly and without a negotiation:
We would have loved to, but we had to draw a line to fit the venue and we have gone with partners only across the whole list, on both sides. It is not personal and it is genuinely not something we can move. I promise you will know plenty of people there, and I will make sure of it at the seating chart.
Children
Children are the one category where a partial rule is worse than either extreme. "Immediate family children only" is defensible. "Some children" without a rule behind it is a war.
Decide all, none, or a stated bright line, and put it on the invitation and the website in the same words both times. Then, if you are excluding children, solve the actual problem you are creating, which is childcare:
We love your children and we have chosen an adults-only celebration so that parents can genuinely relax. If it helps, we have arranged a sitter at the hotel from 5pm, and there is a list of local options on the website.
The offer of a sitter converts the most common refusal into an acceptance more often than anything else on this page.
The B-list, run properly
A B-list is fine. A visible B-list is not, and visibility is entirely a function of timing.
The arithmetic: send your A-list invitations eight weeks out with an RSVP deadline at four weeks. Send B-list invitations as declines arrive, and stop sending them at three weeks out. After that point the invitation is obviously late and the recipient can do the maths as well as you can.
Send save-the-dates only to people who are definitely invited, because a save-the-date is a promise and withdrawing one is far worse than never sending it. This is the single most common way a B-list becomes visible.
Keep the wording identical between waves. Do not apologise, do not explain, and above all do not say "a space has opened up," which converts a lovely invitation into a demotion notice in six words.
When somebody finds out
Occasionally it lands anyway. What matters is that you have a structural answer, delivered quickly and without over-explaining.
You are right that we did not, and I am sorry you heard it that way rather than from me. We capped at 140 for the venue and ended up with rules we applied to everyone on both sides, which meant a lot of people we love did not make it. It genuinely was not about you, and I would love to take you to dinner after.
Three things that message does. It confirms rather than deflects, since denial makes it worse. It names the structure rather than the person. And it offers something real, because the underlying hurt is about the relationship rather than the party.
What not to do: blame your partner, blame the venue in a way that is untrue, or promise them a place at something that does not exist yet.
Keeping the list true
A guest list is not a document, it is a live count that four vendors depend on. Your caterer, your venue, your stationer and your seating chart all read from it, and they read it at different moments.
Organise by household rather than by person. That one decision fixes most counting errors, because your invitation count, your save-the-date count and your final headcount are three different numbers derived from the same list, and only household-level grouping produces all three correctly.
Track RSVP status as four states rather than two: invited, yes, no, and pending. A blank cell is not a no, and treating it as one is how people get missed.
Our Wedding Guest List Tracker is $19.99 and is built on exactly this shape: household-level grouping, full contact capture ready to export for invitations and thank-you cards, RSVP status with live totals, meal and dietary selections in one place to hand to your caterer, plus-one and children columns that do not double-count, and seating-ready columns so the chart builds straight out of the confirmed list. It works in Google Sheets and Excel, arrives instantly, and both of you work from the same live version.
Can somebody else answer from it?
Our six-foot test asks whether a guest can read a sign from six feet. The guest list version: can your partner open the file and give the caterer a number, today, without calling you?
Test it with the three questions vendors actually ask. How many confirmed. How many still pending. How many vegetarian. If any of those takes more than thirty seconds, the list is not doing its job.
The usual culprits are the same three every time. Plus-ones recorded in a note rather than as their own row, so the headcount is quietly wrong. Households split across separate lines, so invitations get double-counted. And dietary requirements sitting in an email thread rather than in the file, which is where they are on the Tuesday your caterer needs them.
Before It Prints
The guest list gates your seating chart, and the seating chart is the one sign that cannot be ordered early. Everything else can be finalised at eight weeks. The chart waits on RSVPs, which is why it is the last thing printed and the thing most likely to be rushed.
Plan for that rather than being surprised by it. Set your RSVP deadline at four weeks out, chase non-responders in the following three days rather than waiting, and lock the chart at three weeks. Chasing early is the whole trick, because the people who have not replied by the deadline were never going to reply to the deadline.
Size the chart to the count you are actually going to have. An Aria 01 seating chart at 24x36 is $95 as a poster, $175 on gatorboard, $345 in acrylic and $395 in gold mirror. If your count pushes past what fits, 24x47 takes the poster to $145 and the gold mirror to $510. Knowing your number early is what stops that decision being made in a hurry.
Every gathurr order includes a Proof Review at no charge, and on a seating chart it matters more than anywhere else, since it is the sign with the most proper nouns on it. We check spelling, spacing and sizing by hand before production, and it is included with every order.
The Cheat Sheet
- Cut by category, never by name. A policy survives being repeated. A ranking does not.
- Start from a number and fill it, rather than writing a list and shrinking it.
- Fully loaded, each guest is $120 to $250. Twenty extra guests is roughly $3,600, which makes it an arithmetic question rather than a moral one.
- Choose policies before you look at names, write them down, and allow no exceptions.
- Apply every rule to both families identically. Asymmetric rules are the main source of conflict.
- You do not owe an invitation because you attended their wedding. Reciprocity is not a debt.
- Build four lists and allot a number to each. Give parents a number, never a veto, and send an over-long list back rather than cutting it yourself.
- Plus-ones by relationship status, not by person, and print the number of reserved seats on the invitation.
- Children: all, none, or a stated bright line. If none, offer a sitter and watch the refusals turn into acceptances.
- B-list works if it is invisible: save-the-dates only to certain invitees, identical wording between waves, nothing sent after three weeks out.
- If someone finds out, confirm it, name the structure, offer something real, and do not over-explain.
- Organise by household, track four RSVP states, and keep dietary needs in the file rather than in your inbox.
- The seating chart is the one sign gated by RSVPs. Deadline at four weeks, chase immediately, lock at three.
Keep reading
The guest count drives the money, so How to Build a Wedding Budget That Survives Contact With Reality is the companion piece, including the per-guest math used above. For where the RSVP deadline sits among everything else, The 12-Month Wedding Planning Timeline plans in dependency order, and The Wedding Vault covers keeping the file findable and current.
Once the count is settled, Seating Chart vs. Escort Cards covers how to display it, and What Wedding Signs Really Cost in 2026 puts the chart in the context of the whole signage budget.
The tools: the Guest List Tracker at $19.99, the Budget Tracker at $29.99 since your count drives your cost, and The Wedding Vault at $97 for all nine, which come to $222.91 bought separately.