Contents
- Dashboards fill up faster than decisions do
- A number needs an agreed reaction before it counts
- Samples handed out at a mall corner measure effort
- Where did the target guests drop off?
- Half empty halls after lunch are data too
- Queue times and reprinted badges belong in the client report
- Three to five indicators, each with a name and a date beside it
There is a comfortable belief in our business that the more you measure, the better you understand your event. Registration platforms, apps, badge scanners and the rest of today’s event technology now produce data almost by themselves, so it feels natural to put all of it into the report and call each line a KPI.
I think the opposite is closer to the truth. Past a certain point, every extra number makes the event harder to understand, because the few figures that matter get buried among the ones that are merely available. Key performance indicators were borrowed from the business world to mean the handful of measures that tell you whether you are on track, and good event KPIs are exactly that: three to five figures about who came, what happened inside, how the operation ran and what changed in the months after. In events, the term has grown so loose that almost anything countable now carries the label, and that is exactly why so many event dashboards look busy and say very little.
Dashboards fill up faster than decisions do
Almost every event I have worked on had numbers. Far fewer had indicators. The difference shows up very practically in the week after the event.
A team with numbers sends a long report full of registrations, app downloads, impressions and badge scans. A team with indicators sends a short note that says what went as planned, what did not, and what that means for the next decision. The first one is usually more impressive to look at. The second one is the one that gets used when next year’s budget is discussed, because it was written with a decision in mind rather than a presentation.
I have sat on the receiving end of both kinds of report, on the client side and later as an agency partner, and the long ones are never the problem in themselves. The problem is that length lets everyone avoid choosing. When forty figures are on the table, nobody has to say which three would make them change the program, the venue or the supplier next year, and so nobody does. Better software only makes that choice harder. A shorter list forces that conversation before the event, which is the only time it is cheap to have.
A number needs an agreed reaction before it counts
The simplest test I know for an event KPI is to ask what happens if it comes out low. If the honest answer is that nothing would happen and it would simply be reported, then it is a statistic, not an indicator.
Take the show up rate for a free industry conference. If the organizer has decided that a rate below a certain level means changing the reminder sequence, charging a small registration fee next year or moving the date away from a busy week, the show up rate is a real KPI, because it is attached to a decision. If it is only there to fill a slide, it will be read, nodded at and forgotten.
This is also why indicators differ so much from event to event. A dealer meeting, a medical congress, a shopping mall activation and a city festival do not share a purpose, so they cannot share indicators, and copying a list from another event or from a template is one of the quickest ways to measure the wrong thing very carefully.
Samples handed out at a mall corner measure effort
A lot of what ends up in event reports is activity: registrations, emails opened, app downloads, social media posts, photos at the branded wall, samples handed out by the hostess team over a weekend. These numbers are not meaningless. They tell you that things happened, and they can point to problems in the process.
But activity is not a result. Communication professionals have long made this point, and the Barcelona Principles 4.0 from AMEC, the measurement and evaluation association, still begin with clear objectives and keep outputs, outcomes and impact apart, a separation event reports would do well to borrow. A thousand samples handed out in an experiential marketing corner of a mall tell you how hard the team worked. They do not tell you whether anyone bought the product the following week. Strong reach on social media tells you the content travelled; whether it reached the people the brand needed is another matter.
If you ask me, the healthiest reports keep activity in a short appendix and put results on the first page. The temptation is the opposite, because activity numbers are big and results are often small. A launch that produced twelve serious distributor conversations sounds modest next to a reach figure, yet those twelve conversations are what the event was for.
Where did the target guests drop off?
For most B2B events, congresses and corporate meetings, the most important indicator is settled before anyone arrives. It is the quality of the guest list, followed step by step.
How many of the people you actually needed were invited? How many of them registered? How many of the registered ones came? Free events in particular have a familiar pattern in every market I know: event registration numbers look healthy, and then a large part of the list simply does not show up, because nothing was at stake for them and something else came up that morning. If you only look at the final headcount, you miss where the loss happened.
A number becomes a KPI only when everyone has agreed, before the event, what they will do if it disappoints.
The guest list shows why. If the target guests were never invited, the fix is in the list. If they were invited but did not register, it is in the invitation or the program. If they registered and did not come, it is in the reminders, the date or the location. Each of those has a different owner and a different remedy, and none of them is visible in the single number on the last line of the registration export.
Half empty halls after lunch are data too
Once guests are inside, the questions change. Did they stay? Where did they spend their time? Which sessions held the room, and which emptied it?
Anyone who has run a congress, whether in a city congress center or a resort hotel on the coast, knows the afternoon problem. The opening is full, the morning plenary is full, lunch runs long, and the first session after lunch speaks to a half empty hall while the corridors and the hotel lobby are busy. That pattern is an indicator in itself, if someone bothers to record it. A simple count of people in each session at a fixed time, done by the floor team, tells the program committee more about next year’s agenda than any satisfaction score.
Some committees push back on this, and I understand why. A thin room after lunch can be blamed on the speaker, the topic or the catering, and nobody wants a figure that points at a colleague who chaired that session. But the count does not have to be read as a verdict on anyone. Kept over two or three editions, it usually shows that the slot itself is weak, whoever is on stage, and that is a scheduling decision the committee can actually make: shorter lunch, a livelier format at two o’clock, or the most awaited speaker placed where the hotel lobby would otherwise win.
For fairs and hosted buyer programs, the inside measure is meetings: those scheduled, those that actually took place, and those that produced a next step. Accepted and completed meetings belong in separate lines, and every rate needs its starting group written next to it, because a figure such as thirty percent means one thing against all badge scans and something quite different against the conversations the team itself judged qualified. For a virtual event, it is how long people actually stayed in each session, which the platform records whether anyone asks or not. In a hybrid event the two audiences need separate counts, because a full hall can hide an online audience that left after ten minutes. For workshops, it may be how many groups finished the task. For a gala dinner, honestly, it may be nothing more than whether the program ended on time and people stayed past the main course.
Queue times and reprinted badges belong in the client report
Some of the most useful indicators never reach the client, because they are considered internal. How long was the longest queue at registration? How many airport transfers left late? How many guests were told at check in that their room was not ready? How many badges had to be reprinted because of spelling mistakes? How many questions did the information desk fail to answer?
Nobody enjoys collecting these, because they are a record of things that went wrong. Yet they are exactly what guests remember and repeat. A participant rarely goes home saying the keynote was average; they often go home saying they waited half an hour for a badge. An organization company that tracks these figures event after event gets steadily better at the parts of the job clients notice only when they fail.
In our business, the people at the door are often temporary: young people booked by the day through an agency, a few university students asked to help, a team whose members met that morning. Whether they had a short briefing before doors opened, and whether someone was there to answer their questions during the day, can be recorded too, and it predicts how the entrance felt better than almost anything else.
Three to five indicators, each with a name and a date beside it
The last group of indicators cannot be measured on the day at all, and this is where KPIs start to overlap with an event impact assessment. Did the leads from the fair get a call? Did the dealers who attended order more in the following quarter than those who did not? Did the sponsor renew? Did participants come back the next year and bring colleagues? Did the association that came on a site inspection put the city on its shortlist?
A word of caution belongs here, though. If the dealers who attended ordered more, the meeting may deserve the credit, or it may simply be that the most engaged dealers are the ones who come to meetings in the first place. The indicator shows the difference without proving its cause, and a report that says so is easier to trust.
These are often the indicators closest to the purpose of the event, and the ones most often dropped, because the team that ran it has already moved on. Agencies come and go, and their contract usually ends with the final invoice, so the follow up works best when a person on the client side is named to look at those figures again on a specific date, ideally at a second debrief a few months on.
Put together, I find that three to five indicators per event is plenty: one for the guest list, one for what happened inside, one operational, and one or two for the months after, each with a written target and a name next to it. For a congress with a hosted buyer program, the short version of that list might look like this:
| Indicator | Counted against | If it comes out low | Owner |
|---|---|---|---|
| Target guests who attended | Target guests invited | Rework the list or the reminders | Client marketing lead |
| People in the first session after lunch | Opening plenary headcount | Change the slot or the format | Program committee |
| Longest wait at registration | The wait agreed before the event | More desks or staggered arrivals | Agency operations lead |
| Buyer meetings that produced a next step | Meetings that took place | Revisit matching and follow up | Client sales lead |
The rows will change from event to event, and they should. What should not change is the length, because a longer list dilutes attention, and the report becomes a place where good numbers hide bad ones. A figure that came out below target is not an embarrassment in such a report. It is usually the most valuable line in it.
Who on your team decides what happens when one of these numbers comes out low?