January is kickoff season. Somewhere this month, a commercial excellence team will fly several hundred people to a resort hotel, run two days of plenaries and breakouts, and send a survey on the plane home. A few weeks later, the report will say that satisfaction was 4.4 out of 5, attendance was 97%, and the gala dinner was the highlight.
All of that may be true. None of it answers the question the CFO will eventually ask: what did we get for it?
This piece is about how to answer that question honestly. We think a pharma or MedTech event should be measured first on what people know afterwards, not on how they felt on the day, and that this is now straightforward to measure if you design for it from the start. Below is the framework we use, the metrics that matter, and a practical plan for the next kickoff, launch meeting or cycle meeting on your calendar.
What most event reports measure, and why it misleads
The standard event report has three kinds of number.
Logistics: registrations, attendance, sessions delivered, cost per head. These tell you the event happened as planned. They are necessary, and they say nothing about impact.
Sentiment: satisfaction scores, net promoter style questions, favourite session, comments. These tell you whether people enjoyed it. Useful for the next agenda, but weakly connected to what happens in the field. A charismatic keynote speaker scores well whether or not anyone remembers the content.
Activity: app downloads, content views, social posts, time spent in the virtual platform. These tell you people were busy. They are easy to inflate and easy to misread.
Put together, these numbers can make a mediocre event look excellent. High attendance, happy attendees and lots of clicks are exactly what you would expect from a well organised two days in a good hotel, whether or not anyone learned anything.
What almost no event report contains is a measure of knowledge or capability, before and after, per person and per topic. And that is the thing the event was supposed to change.
A satisfaction score measures the evening. A knowledge gap measures the next six weeks of calls.
Why the gap matters more than it used to
A kickoff or launch meeting is one of the most expensive things a commercial organisation does. The venue and travel are the visible part. The larger cost is usually invisible: the entire field force, for two or three days, not seeing customers.
That cost is easier to justify when the stakes are high, and they are. IQVIA's Launch Excellence IX report (June 2025) estimates that around $200 billion of revenue is at risk from loss of exclusivity by 2030, and that companies need an unprecedented number of launches with consistently strong performance to replace it. McKinsey's analysis of 210 launches between 2003 and 2009, published in 2014, found that about two thirds of new drugs miss pre-launch consensus expectations in their first year, and that 78% of year one underperformers are still behind in year two.
If the first year of a launch sets the trajectory, the launch meeting is where the field's first year starts. It is worth knowing whether it worked.
Four pillars, and how to measure each
A common way to frame event success uses four pillars: engagement, knowledge sharing, retention and instrumentation. We like the structure. What matters is how each one is measured, on whom, when and at what cost in field time.
1. Engagement: measure participation in retrieval, not presence
Being in the room is not engagement. Answering questions is. If every block of the agenda ends with a short round of questions, answered on phones by everyone present, you get a real engagement measure: the share of people who answered, per session, per team. It is hard to fake, and in our experience it tracks attention far better than a head count.
Sample, timing, effort: everyone present, at the end of every agenda block; about a minute per block for each attendee, and no extra work for the event team if the questions are written in advance.
2. Knowledge sharing: set a baseline before the event
You cannot measure what an event taught unless you know what people knew before it. Two or three weeks before the event, send a short set of questions on the topics it will cover. That baseline shows where the real gaps are, lets you adjust the agenda, and gives you the "before" half of any before and after comparison.
Sample, timing, effort: everyone invited, not only those who say yes early, two to three weeks before the event; ten to fifteen questions, taking a few minutes in total per participant. Report the response rate alongside the results, because a baseline answered by only the keenest half of the field will flatter the starting point.
3. Retention: measure again weeks later, per person
Retention is the pillar that most events skip, and the one that matters most. Memory research gives a general warning here. When Murre and Dros replicated Ebbinghaus' forgetting curve in 2015, in a relearning experiment with a single participant and nonsense syllables, savings after 31 days were down to about 4%. That does not tell us how much of a kickoff your field will forget. It does show that unrehearsed material fades fast and that spaced practice helps, which is reason enough to measure rather than assume.
So the measure that counts is not a quiz at the end of day two. It is how people answer the same questions two, four and six weeks later, and whether the people who were wrong the first time are right now.
Sample, timing, effort: every attendee, with follow-up on initial errors and checks on questions previously answered correctly, at 48 hours and then two or three short challenges a week for six weeks; roughly five minutes a week per person. Track who drops out, because closure measured only on the people still playing in week six overstates the result.
4. Instrumentation: plan the data before you plan the venue
Measurement has to be designed in from the start. If you only think about data after the event, all you will have is the survey. Decide before the agenda is finished which topics you will measure, how you will define a gap, who will receive the reports and what they are expected to do with them.
Sample, timing, effort: a half-day workshop with the brand, training and medical owners before the agenda is locked, and one named owner for the data through week six.
The metric we would put at the centre: gaps closed
The single most useful event metric we know is simple to define.
A knowledge gap is one person, wrong on one question, at their first attempt. It counts as closed only when that same person later answers that same question correctly. One later right answer does not rule out a lucky guess, so for critical topics we use a stricter rule: a critical gap counts as closed only at 5 of 5 later check questions on the same concept. In both cases, closure is a verified correction within the assessment, not proof of what happens in a call. Report cumulative closure and current status separately: a closed gap answered wrong later is a relapse and counts as open again.
This measure has a few properties that averages lack:
- It is about individuals. A team average can rise because the strong got stronger while the weak stayed weak. Gap closure shows who actually moved.
- It ignores what people already knew. Someone who was right at the start does not inflate the result.
- It points to action. An open gap has a name, a topic and a question attached. A manager can do something with that.
- It separates two kinds of problem. Most gaps close with repetition and feedback. A minority do not, and those are usually real misconceptions that need a coaching conversation rather than more practice.
Report gap closure by topic, team, country and role, and you have an event report that tells the business where the field is ready and where it is not.
From knowledge to behaviour: the harder link
A fair objection: knowing is not doing. A rep can answer every question about the new subgroup data and still not use it in a call. Shouldn't we measure behaviour and business results instead?
Yes, and knowledge measurement does not replace that. But it is the missing link between the event and the field. Business results arrive months later and depend on access, competition, supply and a dozen other factors. Behaviour is hard to observe at scale. Knowledge can be measured within days, per person, and it is a precondition for the behaviour you want. A rep who cannot state the primary endpoint correctly is not going to use it well in front of a specialist.
The practical answer is to connect the layers. Use gap data to decide which reps managers should ride along with first. Check whether the topics with the most open gaps are the ones HCPs raise in CRM call notes. Compare readiness by territory with early uptake signals by territory in the first months. None of this proves causation. Uptake depends on access, targeting, competition and supply, and a knowledge metric on its own cannot explain a commercial change. But together these layers give a far better picture than a satisfaction score.
What about social media and the virtual platform?
Social media used to get a lot of space in event measurement guides. For external congresses and public-facing events, that still has a role, within your company's policies and with Medical and Legal involved. For internal kickoffs and cycle meetings, it is mostly beside the point, and often not permitted for confidential content.
The same goes for many virtual event platform metrics. Minutes logged in and sessions clicked are activity, not impact. They are worth tracking for operations. They should not be in the headline of the report.
What a good event report looks like
Put it together and the report changes shape. Instead of a slide of satisfaction scores and a photo of the gala dinner, the first page says something like this: the baseline showed the field was weak on the mechanism and on two label points; the event and the six weeks after it closed most of those gaps; three countries are behind on one topic; and a small group of misconceptions about competitor positioning has resisted repetition and has gone to managers for coaching.
Notice what is missing from that page: the claim that the event was a success. The report does not need it. It shows where the field started, where it is now and what still needs work, and it leaves the reader to judge. In our experience that honesty earns more trust with finance and senior leadership than any satisfaction score, because it admits the parts that did not move.
That is a report a brand lead can act on and a CFO can read. It also changes the next event, because the agenda can start from what the field still does not know rather than from what the brand team would like to present.
Start with the next event on the calendar
You do not need to rebuild your event strategy to do this. Take the next kickoff, launch meeting or cycle meeting and add three things: a baseline before, questions during, and six weeks of short reinforcement after, with gaps tracked to closure.
That is the structure Atrivity is built around: one-minute challenges between real colleagues, battles in a synchronised window during the event, Gap Fixing that sends each person the fix for their own mistakes afterwards, and readiness reported by team, country and topic, with persistent misconceptions flagged for coaching. But the principle holds with any tool. The question to bring to your next event planning meeting is this: six weeks after this event, what will we be able to prove our people know that they did not know before?