How to Count Event Attendance: Visitor Counting Methods, Tally Sheets, and Real-Time Occupancy

What attendance means: total visits, unique visitors, and occupancy

“Attendance” can mean different things depending on how you count. For the same event, “3,000 total visits” and “2,400 unique visitors” can both be correct, and if you start counting without deciding which one you mean, you won’t be able to compare with last year or with other events. Start by understanding these three.

Shops often distinguish between “foot traffic” (people who came through the door) and “transactions” (the number of purchases at the register). If you also count people who simply walk past on a shopping street or outside a station, that is not attendance but pedestrian volume. Pedestrian counts are covered in how to do a traffic count.

Ways to count attendance and when each works best

There are five broad ways to count attendance. Each differs in accuracy, cost, staffing, and what it tells you, so choose one to suit the venue layout and your purpose, or combine them.

With one or two entrances and a flow of up to a few dozen people per minute, staff with tally counters are the simplest and most reliable option. You can count with tally marks on paper, but splitting them by time slot takes effort and recounting is hard. A phone tally counter that records the time of each press lets you turn the counts into a 15-minute table afterward.

What to decide before counting: location, categories, rules, staff

The accuracy of attendance figures is decided more by preparation than by the day itself. Settle these six points and on the day you can focus purely on counting.

1. 1

   1. Choose where to count

   If there are several entrances, put someone at every one or close the ones you won’t count. Even one uncounted entrance makes your numbers low by that amount. Define the count line with a landmark everyone can picture, such as “beside the reception desk” or “just inside the automatic doors.”

2. 2

   2. Decide whether to count exits too

   If you only need total visits, counting the entrance is enough. If you want to know how many people are inside (occupancy), or you’re managing capacity or entry limits, count people leaving at the exits as well. If the entrance and exit are the same door, one person can press two counters: “Entries” and “Exits.”

3. 3

   3. Decide what to count and how to split it

   Make counters only for the categories you need to report separately, such as adults and children, general visitors and guests, or individuals and groups. The more categories, the more mis-presses, so aim for three or four per person. Also decide here whether to count infants and whether to exclude staff, exhibitors, and performers.

4. 4

   4. Decide how to handle re-entry

   Whether you recount people who leave and come back makes a big difference to total visits. If you want a figure close to unique visitors, hand out wristbands or stamps at entry and don’t count people who show them on the way back in. If instead you want to see crowding including comings and goings, count every entry.

5. 5

   5. Decide the hours and intervals

   Decide whether to count from opening to closing or only at peak times. If you’ll compare time slots later, 15 minutes is the most useful interval. With a tally counter that records press times, you can switch to any interval, such as 5, 30, or 60 minutes, after counting.

6. 6

   6. Assign people and rotation times

   At an entrance with a constant stream, one person can concentrate on counting for about 30 minutes to an hour. Line up handovers with round interval times (every hour on the hour or half hour), and read the on-screen numbers aloud at each handover so it’s clear who counted up to where.

Tips for counting at the door: avoid misses and double counts

• Press the moment someone crosses the line: not when you see their face or when they reach reception, but when their body crosses the line you defined. Pressing early counts people who turn back.
• For clusters, look at the number before pressing: with groups of three or four, wait until they’ve all crossed the line and then press that many times. You miss fewer people than when you chase them one by one.
• Ask the leader of a large group for the head count: for groups of more than ten, ask the organizer how many there are. Decide in advance whether to use the “Groups” counter for the number of groups and press “Entries” once per person, or add the head count in a note.
• Confirm by feel: use vibration on press, or number keys on a computer, so you can press without looking at the screen. If you keep staring at the screen, you’ll miss people walking past.
• Fix mistakes immediately: when you notice a mis-press, undo it or press −1 right away. If you try to remember it later, you won’t know which category you got wrong.
• Call out numbers every 15 minutes: when there are several entrances, have each one report its count to the control desk by radio or chat at each interval. If one entrance is unusually low, you can check the cause on the spot (an absent counter, another door opened, and so on).
• Be especially careful just after opening and before closing: right after opening the queue surges in at once, and before closing exits bunch up. Put two people on the door for these periods only, splitting entries and exits, to keep the numbers stable.

Hand counts inevitably carry an error of a few percent. To keep it small, have two people count the same entrance separately for about 30 minutes and check the percentage difference; then you can state in your report how reliable the day’s numbers are. For example, if two people count 412 and 405, the difference is 7, or about 1.7%.

Calculating occupancy (people inside): entries minus exits

The number of people inside the venue at a given moment is the cumulative number of entries since opening minus the cumulative number of exits. If the venue is empty at opening, the formula is as follows.

Here is an example from an event that opened at 10:00, with entries and exits counted every 15 minutes at a combined entrance and exit. Occupancy is the number of people inside at the end of each slot.

Total admissions over the two hours were 577, but the venue was fullest at 11:30, with 364 people inside. If the venue’s capacity is 400, it reached 91% of capacity at 11:30. The slot with the most entries (10:45–11:00) and the moment the venue is most crowded (11:30) often don’t coincide; judge crowding by entries alone and you’ll put the peak about half an hour too early.

• Set an entry-limit threshold in advance: for example, at a venue with a capacity of 400, decide that “entry pauses when occupancy passes 340 (85%).” Staff can then act without hesitation. Check each venue’s permitted capacity with the facility manager.
• If entrance and exit are separate, decide how to combine them: have the exit counter use the same 15-minute intervals, and calculate “cumulative entries − cumulative exits” at the control desk. The numbers won’t match if clocks differ, so make sure everyone’s phone is set to automatic time.
• Check that the count approaches zero at closing: if a large occupancy remains after closing, it means exits were missed or there was an exit no one counted. It’s a useful check on how reliable your numbers are.

Reading a time-slot tally sheet and the peak hour

A time-slot table tells you when to add staff, when to schedule breaks, and whether to change the opening time next time. The key to finding the peak hour (the busiest hour) is to compare one-hour windows shifted in 15-minute steps, not clock-hour blocks. Using the entries from the example above:

Splitting by the clock into the 10 o’clock and 11 o’clock hours, the larger is 326 (the 10 o’clock hour). Shift in 15-minute steps, though, and 378 people came in between 10:30 and 11:30, 52 more than the best clock hour. When deciding how many reception staff or ticket windows you need, base it on 378 per hour, or queues will build at the peak.

• Convert to an hourly rate: 95 people in 15 minutes is a pace of 95 × 4 = 380 per hour. Compare that with how many people one receptionist can handle (say 6 per minute = 360 per hour) to estimate how many windows you need.
• Find the peaks for entries and exits separately: if you find the peak from entries plus exits, the example above peaks at 10:45–11:45 (335 entries + 156 exits = 491), which differs from the entries-only peak (10:30–11:30). To judge crowding at the entrance, compare the entries column alone.
• Look at shares too: comparing the proportion of children, groups, or invited guests with last time shows how your audience is changing.

Estimating when you can’t count everyone: sampling and area × density

At open-air events without gates, or when the flow is more than one person can count, estimate instead of counting everyone. Always report an estimated figure as an estimate, together with the method.

1. 1

   Sampling (count only part of the time)

   Count for 10 minutes from the top of each hour and multiply by 6 to get an hourly figure. 63 people in 10 minutes is 63 × 6 = 378 per hour. Starting at the same minute every hour makes time slots easier to compare. Where the flow comes in waves, such as train arrivals or show start times, count for 15 minutes or more, or make sure your count spans a whole wave.

2. 2

   Area × density (how many people are present)

   Estimate the number of people in a plaza or viewing area as occupied area × people per square meter. As a rough guide: about 1 person per m² when there’s room between people, about 2 per m² when shoulders touch, and about 3–4 per m² when the crowd is packed and hard to move through. If 60% of a 50 m × 30 m plaza (1,500 m²) is filled at 2 people per m², that’s 1,500 × 0.6 × 2 = 1,800 people.

3. 3

   Unique visitors from total visits (divide by visits per person)

   At a two-day event with 3,000 total visits, if 25% of survey respondents say they came on both days, the average number of visits per person is 1 × 0.75 + 2 × 0.25 = 1.25. Unique visitors can then be estimated at 3,000 ÷ 1.25 = 2,400.

Counting store foot traffic and using conversion rate and average transaction value

In a shop, combining the foot traffic counted at the door with the number of register transactions lets you break sales down into “people,” “share who bought,” and “amount per buyer.”

Breaking it down this way makes clear which lever to pull. With the same foot traffic, raising the conversion rate from 35% to 40% gives 192 transactions and sales of 192 × 2,300 = ¥441,600, an increase of ¥55,200. If foot traffic is low, look at signage and promotion; if conversion is low, product range and service; if transaction value is low, bundles and how recommendations are displayed.

• Count entries only: in a shop with a single door, counting both directions doubles your foot traffic. Press only for people coming in, or split into two counters, “In” and “Out.”
• Separate parties and people: in shops where families and friends often come together, counting both the number of parties and the number of people lets you read conversion correctly alongside transactions per party.
• Exclude staff and suppliers: don’t count people coming in and out for opening preparations or deliveries, or give them a separate counter.
• Compare by weekday and time slot: comparing the same weekday and time slot makes it easier to separate out the effects of weather and events.

How to state attendance in reports and announcements

The credibility of an attendance figure depends less on the number than on how it was counted. When you publish a figure in a report or press release, include the following so readers can compare it correctly with last year or other events.

• Which number: total visits or unique visitors, and whether staff, exhibitors, and infants are included.
• When: the dates, the hours counted (e.g. 10:00–17:00), and a per-day breakdown.
• How it was counted: tally counters at the entrances, ticket scans, sensors, or an estimate (with the method).
• Whether it’s an estimate: if part was filled in by estimation, the scope and calculation method.
• What it’s compared with: year-on-year, against target, or against other events counted the same way.

“Organizer’s figures” announced for large events are often total visits or estimates, and the counting method differs from event to event. When comparing them with your own figures, make sure you compare like with like.

Counting visitors with the Tally Counter (Multiple)

The Tally Counter (Multiple) is a tally counter that runs in a phone or computer browser. Because it also records the time of each press, simply counting entries and exits automatically gives you a time-slot table, the peak hour, an hourly rate, and the share of each category. No app installation or sign-up is needed.

1. 1

   Choose the “Visitors & customers” template

   Four counters appear: “Entries,” “Exits,” “Children,” and “Groups.” The unit is set to people.

2. 2

   Rename the categories to fit your count

   You can freely change the counters’ names, number, and colors. Split them into “Adult entries,” “Child entries,” and “Exits,” or delete Exits if you only count the entrance. You can add up to 12 counters. Entering a survey name (e.g. “Summer Fair, Main Entrance, Day 1”) also makes saved records easier to tell apart.

3. 3

   Hand out the same counters to every entrance

   “Copy this counter’s URL” creates a URL containing only the survey name, unit, and counter names (not the counts). Send it to the person at each entrance, and everyone counts on their own phone with the same categories.

4. 4

   Turn on “Keep screen on” and start counting

   If the screen dims while you’re counting, you’ll miss presses, so turn on “Keep screen on.” On supported phones you can also have it vibrate when you press. On a computer, the keys 1–9, 0, -, and ^ (from left) press the counters; hold Shift for −1, and Backspace undoes one.

5. 5

   Fix mis-presses with “Undo”

   “Undo” deletes the last press recorded. When you press the wrong category, undoing immediately is the surest fix. “−1” subtracts one from that counter.

6. 6

   Check the 15-minute table and peak hour

   Choose a summary interval of 5, 10, 15, 30, or 60 minutes to see a clock-aligned table by time slot. The peak hour is highlighted as the hour with the largest total across all counters. To see the peak for entries alone, have separate people count the entrance and exit so each counts only one direction, or add four consecutive slots in the Entries column of the CSV and compare.

7. 7

   Save as CSV and calculate occupancy

   “Save the 15-minute table as CSV” gives a file that opens in Excel or Google Sheets without garbled characters. With an Entries column and an Exits column, just fill the formula “previous occupancy + entries − exits” down the rows to get the number of people inside at each time. For reports or chat, “Copy summary as text” lets you paste the totals, shares, peak hour, and per-slot breakdown in one go. Why CSV files garble is explained in why CSV files look garbled in Excel.

Day-of checklist

• Is someone stationed at every entrance (and exit), and are uncounted entrances closed?
• Does everyone know the count line and who to count and not count (staff, exhibitors, infants, etc.)?
• Is the re-entry rule (wristbands, stamps, etc.) decided?
• Are everyone’s phones on automatic time, and is everyone using the same counters (the same URL)?
• Is “Keep screen on” switched on, and does everyone have a power bank?
• Are rotation times and the way to report numbers to the control desk every 15 minutes decided?
• Have you decided the occupancy at which entry will be paused?
• After closing, did you save the CSV and send the text summary to the control desk?

FAQ

What’s the usual way to count event attendance?

At venues with limited entrances, staff usually stand at the entrances and count each person with a tally counter. For paid or pre-registered events, the number of torn tickets or scanned QR codes is used. At open-air events without gates, organizers count for part of the time and scale up, or estimate from the occupied area × density.

What’s the difference between total visits and unique visitors?

Total visits is the number of entries, so someone who comes twice counts as two. Unique visitors is the number of people who came, so the same person counts once however many times they visit. At a two-day event where some people come both days, total visits will be higher. When reporting, always state which number you mean.

How do I calculate how many people are inside right now?

Subtract cumulative exits from cumulative entries. If 577 people have entered since opening and 248 have left, there are 577 − 248 = 329 people inside. If you count entries and exits every 15 minutes, “previous occupancy + entries − exits” gives the number at each time. Add staff and exhibitors who were inside before opening separately.

Should re-entries be counted?

It depends on your purpose. To see crowding and how much people come and go, count every entry. To get closer to the number of people who came (unique visitors), use a rule that people showing a wristband or stamp on the way back in aren’t counted. Whichever you choose, note it in the report so you can compare correctly with last time.

What if it’s too crowded to count?

The basic step is to put two people on the door and split entries and exits, or adults and children. If you still can’t keep up, switch to sampling (count for 10 minutes each hour and multiply by 6) or an estimate that multiplies the occupied area by density (about 1–4 people per m²), and record that the figure is an estimate and how it was made.

How should children and groups be counted?

If you want to report children separately, use separate “Adults” and “Children” counters, and decide in advance whether infants are included. For groups, ask the leader for the head count and add that many; if you also want the number of groups, press the “Groups” counter once per group so both the head count and the group count are recorded.

How do I count customers coming into a shop?

Count only people coming in. In a shop with a single door, counting people leaving as well doubles your foot traffic. Combined with the number of register transactions, you can calculate the conversion rate (transactions ÷ foot traffic × 100). With 480 visitors and 168 transactions, the conversion rate is 35.0%.

How do I find the peak time for attendance?

Build a table of entries every 15 minutes, then compare one-hour windows made of four consecutive slots, shifting 15 minutes at a time; the largest window is the peak hour. If you only split by clock hours such as 10 o’clock and 11 o’clock, you can miss a peak like 10:30–11:30 and underestimate the number of people at peak.

Can I count visitors on my phone and put the data in Excel?

With the Tally Counter (Multiple) tool, every press is recorded with its time, and you can save 5–60 minute tables or the full list of presses as CSV. The format (UTF-8 with a BOM) opens in Excel without garbled characters, so you can use it directly for occupancy calculations and charts. Records are saved only in your device’s browser and never sent to a server.