Every budget line that matters — plates, rooms, chairs, transport — prices off one number: how many people actually come. And Indian weddings forecast that number worse than any other event type, because the RSVP barely exists as an institution here, invitations flow through family channels no spreadsheet tracks, and asking directly reads as rude in half the country's cultures. You cannot fix Indian RSVP culture by November. You can forecast around it.
Why there is no universal percentage
Any page that tells you "expect X per cent of invitees" is guessing, and we will not print a number we cannot stand behind — attendance rates swing enormously with factors that have nothing to do with enthusiasm. The honest drivers, in rough order of force: distance — a hometown wedding draws its city almost fully, while a destination wedding filters precisely by who will spend a flight and two nights; the date's day-job cost — weekday muhurthams thin the working generation and barely touch the retired one; invitation channel — a hand-delivered card from the family patriarch obligates attendance in a way a forwarded WhatsApp does not, and lists built by forwarding inflate fastest; function tier — the reception outdraws everything, home functions draw exactly the inner circle, and everything else sits between; and the season's competition — in the compressed muhurtham weeks, your guests have other weddings the same day, and attendance splits in ways no planner controls.
The method that actually works
Forecast from your own family's history, not from the internet. Take the last two or three comparable weddings in each family — same city, similar scale — and ask the person who paid the caterer what the invite-to-attendance gap was; that ratio, adjusted for this wedding's distance and date, beats any published percentage because it encodes your community's actual behaviour. Then segment the list instead of averaging it: the must-come core (immediate family, the houses that would notice absence), the likely tier (active relatives and friends), and the courtesy tier (the broad invite that signals respect but travels poorly) — and apply different expectations to each. For destination weddings, replace forecasting with commitment where the money is: the room block converts attendance from a guess into bookings with names on them, which is one more reason the block deserves the attention our NRI playbook gives it.
Engineering the RSVP you will not get by asking
Direct RSVP requests underperform; logistics questions do not. "Which airport are you arriving into?" and "veg or non-veg for the plane of the family?" and "how many rooms should we hold for your side?" all return commitment data while reading as hospitality — the wedding-website form asking travel details gets answers the reply-card never would. Route the asking through the family's natural nodes: one aunt's phone census of her branch outperforms fifty individual messages. And time a soft reconfirmation wave two weeks out through those same nodes, which is when plans have actually crystallised.
Absorbing the error you will still have
However good the forecast, build the tolerance into contracts rather than into anxiety: the caterer's final-count confirmation as late as the contract allows, with a stated rate for extra plates; seating for the forecast, stacked chairs for the surprise; hampers and gifts ordered with a ten per cent buffer; and the venue minimum negotiated to your realistic count, not your invite count — the single place where over-optimism converts directly into money. Plan the number honestly, buy flexibility at the edges, and the gap between list and room becomes a rounding error instead of the budget's biggest miss.
