Hold the Lobby. Do Not Fire Question One at an Empty Room.
The Count Is the Only Honest Census
The quiz lobby waiting room is where a live quiz starts. Not the title slide. Not the joke. The lobby lists who has joined. The host sees the count. Question one stays unfired until that count matches the people the score is supposed to describe.
We keep losing the late half of corporate rooms. Hosts advance because the script says minute two. Early phones get the item. People still typing a name do not, and the rank pretends that never happened. The file looks precise. It is not.
This wait is inside the session. It is not a hype mail from two weeks ago. Mail does not put a name on the wall. The lobby does.
Classroom tools often say start whenever and let stragglers in. Fine when a missed item is a shrug. Bad when the company will quote the score. We ship timers and a speed bonus. We also ship a host who advances by hand. Use that control. Do not let a clock in the script override an empty list.
Read the Count Before Question One
People join in a browser. They enter a name. They land in the lobby. They do not see the stem yet. The host sees names and a count. That count is the census before any points exist.
RiLiFi's free tier allows 100 participants per session. If the lobby stops at 100 and the room is larger, that is the plan cap. It is not a mystery stall. Know it before the lobby. A waiting room that refuses the next person is a bad surprise. Say the limit in the briefing when the room is close.
Names on the wall
The lobby is also a name check. People type jokes. People join twice. The host has a moment, while no points are live, to ask for a name a manager will recognize. An attendee can change the display name after joining. Better in the lobby than on the final rank.
If a name is obscene, deal with it before question one. Once the timer runs, the host is reading a stem. They cannot also police nicknames.
The late half
Look at the chairs. If the lobby says 40 and the chairs say 80, half the score is already fiction. Check in this order.
- The code left the screen. Put it back. Leave it. A lobby that hides the code to look clean starves itself.
- People are on the wrong Wi-Fi. That is a portal problem. Starting the quiz does not complete a splash page.
- They think an app is required. It is not. Say the browser path once, out loud.
- They are waiting to be told it is time. The lobby is that announcement. If a name is not on the wall, that person is not in. Then wait.
Roediger and Karpicke on test-enhanced learning is about retrieval, not lobbies. We use it for a narrower claim. A question only reaches the people who see it. A score only describes the people who were in the item. Firing question one at a half-empty lobby invents a knowledge gap that is really an attendance gap.
What the Host Does in the Lobby
The lobby is short. It is not a second keynote.
- Open the session. Confirm it is live, not a rehearsal screenshot.
- Put the code and QR on the screen. Say the join path once.
- Watch the count. Name the gap if it is obvious.
- Set the rules. Named quiz. First timer is 30 seconds unless the stem is long. Default points are 10. A speed bonus exists when several people are right. Lengthen the timer if the item is a policy paragraph. The host range is 5 to 999 seconds. Pick it now.
- Advance only when the count matches the plan. Then the first stem appears for the people who are in.
Questions cap at 250 characters. Options cap at 100. At least two options. If a stem does not fit, fix it in the folder before the lobby, or cut it. The live quiz is built for that host pace. A poll in the same hour uses the same join habit. That is why the live poll belongs in the same folder. One backup host. No second lobby invented on the spot.
We live-host only. The lobby is not an embed in the deck. If the deck advances, the lobby should still be the screen people see until the host chooses question one. A title slide with no code, covering a lobby nobody can see, is how the count stays at zero.
After the Host Advances
A late join is not a full attempt. Someone who enters during question three did not take the first two. Their rank is not comparable. Do not narrate it as if it were.
If the check is for the record, say so in the lobby. A late name will not be in the full score. If the check is a warm-up, say late joiners play from the current item and the podium is unofficial. Pick one sentence before the first stem. A surprise rule is how a waiting room feels like a trap.
The speed bonus rewards a faster correct answer when more than one person is right. That assumes the same start. A room that began with half its names still in the hall did not have the same start. Holding the lobby is how the bonus stays a bonus.
When the session ends, the archive shows people, average score, and the spread. Those numbers inherit the lobby decision. A clean export of a quiz that started at a third joined is a record of the early phones. We would rather have a shorter wait and a file we can defend.
Harvard Business Review on meeting load is about time that does not serve the meeting. A two-minute lobby serves it. A twelve-minute lobby while the host tells stories does not. Hold for the count. Do not hold for atmosphere. When the names match the chairs, advance.
Questions Hosts Ask in the First Four Minutes
Does the host wait for everyone before question one?
The host waits for the people the score should describe. If four people are still in the hall and the company will treat their absence as ignorance, wait. If the room is in and two people will not leave a call, say that and advance. The lobby count is the instrument. The script minute is not.
Can people still join after the quiz has started?
The code can still work. Their attempt is partial. For a scored record, do not pretend a late join is a full try. Say the rule before the first stem.
What should the waiting room show?
The code, the QR, the names, and the count. Not question one. Not a rank for a quiz that has not begun. Not a video that covers the code. This screen is about whether the room is in.
What if the count never reaches the chairs?
Do not burn the items and call it data. Check the code, the network name, and the browser instruction. If the network is the block, the pre-doors test is failing in public. Pause. Fix the path. Return to the lobby. A quiz that starts on a broken join produces a confident, wrong story.
Hold the lobby. Read the count. Then fire question one at the people who are actually there.