An in-person event attendance rate is the number of people who checked in divided by the
number of valid registrations.

Skift Meetings puts no-shows at around 10 to 20% for paid
events and 40 to 60% for events with no registration fee, so a paid conference usually lands
somewhere near 80 to 90% attendance and a free one near half. Those ranges are a starting
point; the rate you plan against should come from your own check-in history.

## What is a good event attendance rate?

For planning, use the rate you forecast before signing contracts. The venue guarantee, the room block, and the audience number promised to sponsors
are all set against an expected turnout, and a miss of 10 points in either direction shows up
as money paid for empty seats or as rooms bought at the last minute.

Few published attendance figures come from research you can trace. The ones below do, and each
row keeps the context its source gave.
| Figure | What it measures | Source and date |
| --- | --- | --- |
| Around 10 to 20% no-show | Paid in-person events. Stated by the publication; no study is named | Skift Meetings, June 2026 |
| 40 to 60% no-show | Events with no registration fee. Same article, same caveat | Skift Meetings, June 2026 |
| 33% to 55% turnout | What one planner with 27 years of experience reports seeing at other events. An anecdote, not a sample | Skift Meetings, June 2026 |
| 34% of planners saw attendance decrease | Most recent in-person event compared with pre-pandemic, up from 20% a year earlier. 80 respondents, surveyed September 2025 | PCMA Convene, 32nd Meetings Market Survey |
| 50% of events behind registration pace | Registrations against each event's regular pace in 2025. Measures registration, not attendance | Freeman End-of-Year Trends Recap, reported by Trade Show Executive, January 2026 |
The two Skift ranges split on price: a registrant who paid has a cost for not showing up,
and one who filled in a free form has none. Complimentary passes for
speakers, sponsors, and exhibitor staff behave closer to free registrations than to paid ones,
even at a paid conference.

The same Skift article shows how far a single event can move from any range. One planner
expected 80% of registrations to show and planned for 900 people; all 1,100+ registrants came.
Kyle Jordan, director of meetings at INFORMS, told Skift that a 2025 international meeting
expected to draw 600 to 800 attendees brought in almost 1,200.

## How do you calculate no-show rate?

Both rates come from 2 counts: valid registrations and attendees.

**Attendance rate (%) = Attendees ÷ Valid registrations × 100**

**No-show rate (%) = 100 − Attendance rate**

A conference with 1,400 valid registrations and 1,176 check-ins has an attendance rate of 84%
and a no-show rate of 16%. When 2 teams report different rates for the same event, the gap
usually comes from what each counted as a valid registration.

For valid registrations, remove test records, duplicates, and cancellations before you divide.
A person who cancelled and got a refund is not a no-show. Decide whether transfers count
once or twice, and write the rule down so next year's number is comparable.

For attendees, use a check-in record: badge pickup, a scan at the door, or an onsite
registration desk. Badge print time works when nothing else exists.
[Event attendance tracking](https://eventiq.io/md/blog/event-attendance-tracking) covers what each of those records proves. Count each person once,
even if they scanned into 3 sessions.

Run the formula separately for each registration type. A conference where 15% of registrations are comps will show a lower blended
rate than the paid audience actually produced, and next year's guarantee will be set too low.

For planning, turn the segment rates into an expected turnout:

**Expected attendance = Σ (Registrations in segment × Attendance rate of that segment)**
| Segment | Registrations | Your attendance rate (history) | Expected attendance |
| --- | --- | --- | --- |
| Paid, member rate |  |  |  |
| Paid, standard rate |  |  |  |
| Registered in the final 2 weeks |  |  |  |
| Free: comps, speakers, sponsor and exhibitor staff |  |  |  |
| Onsite registrations |  | counted as attended |  |
| Total |  | blended |  |
To build the history behind the middle column, export registrations and check-ins for each
past event and join them on registrant ID or email. Group the results by registration type and
by how many weeks before the event each person registered. Skip any group with fewer than 30
registrants, because one or two people will swing its rate by several points. Put the finished
rates into your [post-event report](https://eventiq.io/md/blog/post-event-report-template) so they are there when the
next forecast starts. Run the same join from one edition to the next and it gives you
[attendee retention](https://eventiq.io/md/blog/attendee-retention).

## What is a normal webinar attendance rate?

There is no reliable public benchmark for webinar attendance. The figures that circulate most
often are published by webinar and event platform vendors from their own customers' data. None
of the research sources on this page publishes a webinar or virtual attendance rate, so the
usable number is one you build from your own sessions.

Start with a definition, because "attended" means less online than it does at a venue. A
participant who joined for 2 minutes appears in the same report as one who stayed for the
full hour. Pick a minimum time in session, apply it to every webinar, and keep both numbers:
people who joined and people who met the threshold.

1. Export the registrant list and the participant report for each webinar. Most platforms
   report join time and time in session.
2. Match participants to registrants by email. Count unmatched participants separately; they
   often joined from a forwarded link or a different address.
3. Calculate the attendance rate twice: joined ÷ registrants, and met the threshold ÷
   registrants.
4. Segment by promotion source and by how many days before the session each person registered.
5. After ten or more webinars, use the median of the last several as your planning rate and
   watch for a topic or time slot that sits well outside it.

The [webinar ROI guide](https://eventiq.io/md/event-roi/webinar-roi) follows attendees on to pipeline and closed
deals, and [virtual event ROI](https://eventiq.io/md/event-roi/virtual-event-roi) covers multi-session
online programs.

## How do you forecast event attendance?

Forecasting attendance takes 2 numbers: how many registrations you will end with, and what
share of them will show up. The second comes from the history above. The first is where most
forecasts go wrong, because registration now arrives late.

The Maritz Registration Insights Report analysed 360,000 registrations across 30 trade shows and
found that in 2023, 45% of registrants waited until less than 4 weeks before the event, more
than a quarter until the final 2 weeks, and 9% registered on site
([PCMA Convene](https://www.pcma.org/rethinking-early-bird-pricing-other-event-registration-strategies/)).
That is registration records from trade shows, not a survey and not association conferences, so
use it as a warning about shape rather than as your own numbers. In Kyle Jordan's words: "Our old
registration pacing models are not as reliable as they used to be."

The old model compared this year's registrations with last year's on the same calendar date.
If last year's curve ran earlier, every weekly check shows a shortfall that is not there. A
curve indexed by weeks before the event holds up better, and
[conference registration forecasting](https://eventiq.io/md/blog/conference-registration-forecasting) builds one in full.

**Step 1: Build the baseline curve from your own history.** For each of the last two or three
editions, calculate the share of final registrations you had at 6, 5, 4, 3, 2, and 1 weeks out.
Average them, with the most recent year weighted most heavily.

**Step 2: Calculate the implied final every week.**

**Implied final registrations = Registrations to date ÷ Baseline share for this week**

**Step 3: Adjust with a signal index.** Compare this week's leading indicators with the same
week last year, each as a ratio: net new registrations this week, registration page visits,
opens and clicks on event emails, hotel block pickup, and sponsor and exhibitor sign-ups.
Average the ratios. An index of 1.03 means demand is running about 3% ahead of last year at
this point. Cap the adjustment at plus or minus 5% until the final 2 weeks, so a single
noisy signal cannot move the forecast far.

**Adjusted forecast = Implied final × (1 + capped signal adjustment)**

**Expected attendance = Adjusted forecast × Your attendance rate**

**Step 4: Report a range.** Take the error your forecast had at the same week in past years
and put it around this year's number. Finance can plan against "1,200 to 1,280 registrations"
and see from the width how much is still open.

An illustrative example, with invented numbers: a paid conference whose baseline curve gives
38%, 46%, 55%, 64%, 73%, and 81% of final registrations at 6 weeks through one week out, and
whose own history shows an 85% attendance rate.
| Weeks out | Registrations to date | Baseline share | Implied final | Signal index | Adjusted forecast | Expected attendance at 85% |
| --- | --- | --- | --- | --- | --- | --- |
| 6 | 470 | 38% | 1,237 | 0.97 | 1,200 | 1,020 |
| 5 | 575 | 46% | 1,250 | 1.00 | 1,250 | 1,063 |
| 4 | 670 | 55% | 1,218 | 1.02 | 1,242 | 1,056 |
| 3 | 800 | 64% | 1,250 | 1.02 | 1,275 | 1,084 |
| 2 | 905 | 73% | 1,240 | 1.01 | 1,252 | 1,064 |
| 1 | 1,010 | 81% | 1,247 | 1.00 | 1,247 | 1,060 |
Suppose last year's curve ran earlier, with 48% of final registrations in hand 6 weeks out.
The calendar-date method divides 470 by 0.48 and reports about 979, roughly 21% below the 1,240
the weekly forecasts settle on. A team reading that number in week six could reach for an
unplanned discount, while the weeks-out curve had the event near 1,240 from the first check.

Four decisions fall inside those six weeks and depend on the forecast: the food and
beverage guarantee, the room block release, extra marketing spend, and whether to hold or extend
a price deadline. After the event, record the actual count against each weekly forecast and
update the baseline for next year.

## When should you cancel an event due to low attendance?

Cancel when the high end of your attendance forecast sits below the minimum the event needs, on
or before the last date when cancelling costs less than running it. Set both before
registration opens, so a single disappointing week does not decide them.

The minimum comes from 2 thresholds in your own budget. The financial one
is the attendance at which registration revenue and sponsor commitments still cover the costs
you cannot cut. The experience one is the turnout below which the room, the sessions, and the
exhibit floor stop delivering what attendees and sponsors paid for. The higher of the two is
your minimum.

Then map the decision dates. Venue, hotel, and catering contracts usually carry cancellation and
attrition terms that step up as the event approaches. List each date and what it costs to exit
on it. Review the forecast with cancellation as an option on each of those dates.

On each date, read the forecast range against the minimum.
| Forecast range at the decision date | Likely call |
| --- | --- |
| Low end above the minimum | Run the event as planned |
| Range straddles the minimum | Reduce exposure: a smaller room, release rooms from the block, a lower food and beverage guarantee, targeted outreach to lapsed attendees. Recheck at the next date |
| High end below the minimum | Postpone, merge into another event, move online, or cancel, whichever costs least under the contracts |
For a free event, run the same test with your free-registration attendance rate, not the paid
one. A registration count that looks healthy can still forecast a thin room when half of those
registrants are unlikely to come.

If the range falls below the audience you promised sponsors and exhibitors, tell them at the decision date and offer options while there is still time to act on them.

## Where EventIQ fits

EventIQ replaces nothing. It connects on top of the platforms you already run: registration and
ticketing (Cvent, Zoom, Swapcard, StubHub), CRM (Salesforce, HubSpot, GoHighLevel), and marketing
(Google Ads, Meta Ads, LinkedIn Ads, Mailchimp, Google Analytics). Platforms with an API outside
that list are connected on request.

Contacts are matched by exact email, and
every record keeps its result: matched, unmatched, or no email. Records sync on a schedule:
Swapcard every 15 minutes, Cvent every 30, Zoom every 2 hours, Salesforce every 4. The Event
Dashboard and the Portfolio Dashboard show them in one view.

Registrations and check-ins arrive as separate records, so the attendance rate is a division you
can run on them. The forecast is visible before the event; ask any vendor, including us, what it
is built on.

---

HTML version: https://eventiq.io/blog/event-attendance-rate