Event Debrief Template: Agenda, Questions, and a Decision Log
An event debrief template is a one-page agenda for the internal review your team holds after an event: registrations, attendance, and spend against plan, a short list of questions about the lines that moved, and a log of decisions with an owner and a date. The version below runs in 60 minutes and ends with five decisions at most. A conference debrief or an event post-mortem is the same meeting under another name.
Download the event debrief template (CSV). No email required.
What is an event debrief, and how does it differ from a post-event report?
An event debrief is a meeting. The team that planned and delivered the event reads the results against the plan, agrees on causes, and decides what to keep, change, and stop.
A post-event report is a document for readers outside the team: finance, leadership, sponsors, clients. The post-event report template covers it. The debrief can be blunter, because the notes stay in the room. A line owner can say that a guarantee was set on a guess.
Hold the debrief first. The report then carries causes the team has already agreed.
When should you hold the debrief, and who should be in the room?
Hold it in two parts.
Hot debrief, within 48 hours. Fifteen minutes, no numbers. Each person names one thing to keep and one to change while memory is fresh. Someone writes them down, and nothing is decided.
Full debrief, within ten business days. Sixty minutes, with the numbers. By then the registration and check-in counts are frozen, and spend is either invoiced or can be estimated from contracts.
Invite the six to eight people who own a line of the plan: registration, marketing, operations, production, sponsorship, and the event owner. Give the chair to someone who did not run the event. An event owner chairing a review of their own decisions gets polite answers.
What numbers should be on the table before the meeting?
Three sets of records, each with its plan figure beside it: registrations, attendance, and spend. Send them as a pre-read 24 hours ahead, so the hour goes to causes.
Variance % = (Actual − Plan) ÷ Plan × 100
Attendance rate = Attendees ÷ Registrations × 100
Cost per attendee = Total spend ÷ Attendees
Marketing spend per paid registration = Marketing spend ÷ Paid registrations
Split registrations by ticket type, at minimum into paid and complimentary. A total can hold while the mix underneath it moves. The attendance rate page covers what counts as attended, and the plan column comes from the approved event budget.
The agenda rule. Before the event, write down the threshold that puts a line on the debrief agenda. A workable default: any count or cost line 10% or more away from plan, in either direction, and any rate 5 points or more away. Favorable variances count, since a cost line 15% under plan may be something that was never delivered. Without a rule, the hour goes to whichever line the most senior person finds interesting.
A survey covers only the people who answer it, while registration, check-in, and spend records cover every registrant. Treat the post-event survey as a supplement. Pipeline and revenue attributed to the event stay out of this meeting: they arrive over months and belong in the report.
What does the event debrief template look like?
Six blocks in a fixed order. The CSV holds the same blocks as 25 lines you fill in.
| Block | Minutes | What happens | What comes out |
|---|---|---|---|
| 1. Last debrief's decisions | 5 | Read each one: done by its due date, or not | Decision close rate |
| 2. Numbers against plan | 10 | Read plan, actual, and variance for every line. No explanations yet | Flagged lines |
| 3. Flagged lines | 20 | The owner of each flagged line says what happened and which record shows it | One agreed cause per line |
| 4. What the numbers missed | 10 | One keep and one change per function, from the hot debrief notes | Observations list |
| 5. Decisions | 10 | Sort proposals into keep, change, or stop. Assign owner, due date, and check | Decision log |
| 6. Close | 5 | Read the log back. Name the figures nobody trusted and what goes into the report | Data repair list |
Which event debrief questions get useful answers?
Questions that point at a record. "How do we think it went?" produces opinions in order of seniority. "Which cost line moved more than 10%, and who approved the change?" produces a name and a document. Use the ones that match your flagged lines.
| Area | Questions |
|---|---|
| Registrations | Which registration lines finished outside the threshold, and in which weeks did the gap open? Did the mix of paid and complimentary places match the plan? Which channel or list did we count on that did not deliver? |
| Attendance | What share of registrants attended, by ticket type? Were the no-shows concentrated in one ticket type, segment, or region? Did the headcount we gave the venue and the caterer match the room? |
| Spend | Which cost lines moved more than the threshold, and who approved each change? What did we pay for that nobody used? How much of the change in cost per attendee is spend, and how much is headcount? |
| Delivery | Which on-site decision was made without the person who owned the budget? What did the team do well that is not written down anywhere? What would we stop doing if we ran this event again next month? |
| Data | Which number in the pre-read did someone not trust, and why? Which figure could we not produce at all? Which definition differed between two people in the room? |
How do you turn the answers into decisions?
Sort every proposal into keep, change, or stop. Write it into the log only if it has five fields: the decision, the evidence, one named owner, a due date, and the number that will show at the next event whether it worked. A proposal with no owner in the room is dropped, and the log says so.
Cap the log at five lines, and open the next debrief with the figure that shows whether your debriefs are working:
A low rate usually means the log was too long or the owners were never told.
Example: a debrief for a two-day conference
Take a two-day regional conference planned at 650 registrations, 520 attendees, and $230,000 of spend. The agenda rule is the default above: 10% on counts and costs, 5 points on rates. All figures are hypothetical.
| Line | Plan | Actual | Variance | Variance % | On the agenda |
|---|---|---|---|---|---|
| Registrations, total | 650 | 612 | −38 | −5.8% | No |
| Paid registrations | 580 | 480 | −100 | −17.2% | Yes |
| Complimentary registrations | 70 | 132 | +62 | +88.6% | Yes |
| Attendees (checked in) | 520 | 453 | −67 | −12.9% | Yes |
| Attendance rate | 80.0% | 74.0% | −6.0 points | Yes | |
| Venue | $84,000 | $84,000 | $0 | 0.0% | No |
| Catering | $62,400 | $66,000 | +$3,600 | +5.8% | No |
| AV and production | $38,000 | $44,650 | +$6,650 | +17.5% | Yes |
| Marketing | $29,000 | $26,400 | −$2,600 | −9.0% | No |
| Other | $16,600 | $16,950 | +$350 | +2.1% | No |
| Total spend | $230,000 | $238,000 | +$8,000 | +3.5% | No |
| Cost per attendee | $442.31 | $525.39 | +$83.08 | +18.8% | Yes |
| Marketing spend per paid registration | $50.00 | $55.00 | +$5.00 | +10.0% | Yes |
The meeting opens with last year's log: 6 decisions, 2 completed by their due dates, a close rate of 33%. Total registrations and total spend both sit inside the threshold. Seven of the 13 lines are flagged anyway, and in block 3 they reduce to three causes.
Complimentary places replaced paid ones. Paid registrations finished 100 short while complimentary places ran 62 over, and the total hid the shift. Marketing spent 9.0% less than planned, and each paid registration cost $55.00 against a plan of $50.00.
Complimentary registrants did not come. Of 480 paid registrants, 398 checked in, a rate of 82.9%. Of 132 complimentary registrants, 55 checked in, a rate of 41.7%. The plan assumed 80% for everyone. Complimentary places were 22% of registrations and 77 of the 159 no-shows.
That shortfall is most of the rise in cost per attendee. At planned attendance, the actual spend would have come to $457.69 a head ($238,000 ÷ 520), which is $15.38 over plan. The other $67.70 came from 67 fewer people in the room.
It also exposes a cost the agenda rule did not flag. The catering guarantee was set at 550 covers six weeks out, at $120 a head, and never revised. With 453 in the room, 97 covers went unused: $11,640. At the venue's final deadline 14 days out, registrations stood at 590. The planned 80% rate on 590 gives 472 covers and a bill of $56,640, which is $9,360 less than the $66,000 paid.
AV overtime had no approver. The $6,650 overrun was overtime after the opening session ran long, plus a breakout room added on the first morning. Both were agreed verbally on site.
| Decision | Evidence | Owner | Due | Check at the next event |
|---|---|---|---|---|
| Change: plan attendance with separate rates for paid and complimentary registrants | 82.9% paid, 41.7% complimentary, 80% planned for both | Registration lead | Before registration opens | Planned attendees within 5% of actual |
| Change: set the catering guarantee at the venue's final deadline, from registrations to date | 97 unused covers, $11,640 | Operations lead | Next venue contract | Unused covers under 5% of the guarantee |
| Change: confirm every complimentary place 7 days out, or release it | 77 of 159 no-shows | Event owner | Before the first sponsor contract | Complimentary attendance rate on its own line |
| Stop: on-site changes above $1,000 without the event owner's written approval | $6,650 of AV overtime, no change order | Production lead | Next AV contract | No overrun without a signed change |
| Keep: the reminder sequence for paid registrants | 82.9% paid attendance against 80% planned | Marketing lead | Next cycle's plan | Paid attendance rate at or above 80% |
The log stops at five. A proposal to rework the session schedule had no owner in the room and was dropped.
What mistakes make a debrief useless?
Holding it without the numbers. A debrief run on memory rewards the most confident voice. If the pre-read is not ready, move the meeting.
Writing actions with no owner or date. "Improve onsite communication" is a wish. A decision names a person, a date, and the number that will be checked.
Reviewing only what went wrong. A practice that worked and is not written down gets dropped by accident next cycle. Every log needs at least one keep.
Reading totals only. In the example, total registrations sat inside the threshold while both lines underneath were far outside it.
What to do this quarter
- Write the agenda rule for your next event: one threshold for counts and costs, one for rates.
- Add plan figures for registrations by ticket type, attendees, and each cost line to the event file.
- Book both debriefs before the event, and name a chair who is not the event owner.
- Find the log from your last debrief and calculate the decision close rate.
Common questions
Is an event post-mortem the same as a debrief?
In practice, yes. Post-mortem, retrospective, and after-action review all describe an internal review against the plan, and an event post-mortem template needs the same six blocks.
Does a conference debrief need a different template?
The blocks stay the same and the numbers section grows. For a multi-day conference, add attendance by day and by session or track, with room capacity beside each count. If it has exhibitors or sponsors, add one line per commitment made to them and whether it was delivered.
What if the cost figures are not final?
Use the contracted amount plus approved changes, mark the line as an estimate, and hold the meeting. If a late invoice moves a line across the agenda threshold, the owner reports it at the next debrief.
Where EventIQ fits
EventIQ replaces nothing. It connects on top of the platforms you already run: event platforms (Cvent, Zoom, Swapcard), 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.
For the pre-read, it holds registration and attendance records, and the spend and budget figures your team enters. Registrations from Cvent carry their date, ticket type, and price where the platform provides them, and the check-in is kept as a separate record, so registered and attended stay two counts. Zoom adds attendance by participant with time in session, and Swapcard adds session views per participant. Marketing spend that your team enters or imports from CSV sits by event and channel with the author of each change, beside the budget fields you enter for the event: total, marketing, venue, catering, and other.
The ad platform connections bring campaign name and status only, so every spend figure is one your team entered. Zoom does not give registrants per webinar, so a registered-against-attended comparison for a webinar needs your own Zoom registration report. EventIQ does not calculate the attendance rate or the cost per attendee: those are divisions you run in the template. It does not merge one person across sources, so check a combined attendee count for double counting. It makes no recommendations, and it has no place to record a decision, its owner, or a review date, so the decision log stays in your own document.
Book a demo to see registration, check-in, and entered spend records on a sample event, in a 20-minute demo.