Ask a launch team whether the field is ready and you will almost always get a training answer. The kickoff happened. The modules were completed. The certification was passed. Everyone nods, the slide goes green, and the meeting moves on to access.
Those are signs of effort. They are not proof of knowledge. Our argument is that Commercial Launch Excellence needs a different definition of done: trained means the field attended; ready means the field knows, person by person, on the facts that matter. Most launch dashboards only measure the first. The gap between the two is one of the places a launch can quietly lose ground, and it is measurable, fixable and, in our experience across 100 programs, larger than most teams expect.
Trained and ready are different states, and most dashboards only see one
The distinction sounds semantic until you try to act on it.
A trained field is a field that has been exposed to the content. You can prove it with records: who attended the kickoff, who finished which e-learning, who scored above the pass mark on the day. It is a fact about the organisation's activity.
A ready field, in the sense we can measure, is a field where each person can produce the approved answer on the facts that matter, on demand, weeks after the kickoff, without the visual aid open. That is a fact about each person's knowledge, and it cannot be proven with attendance records. It is also not the whole of readiness: whether someone says it well in front of a specialist is a skill, and that needs ride-alongs, role play or call observation to see. Knowledge is the part a short assessment can verify, and the part the rest depends on.
Two reps, identical training records, very different calls
Picture two reps on the same launch, a scenario we see often. Both attended the kickoff, both completed every module, both passed the certification with a comfortable margin. In the training report they are identical. In week three, the first is asked about the subgroup data by a sceptical specialist and gives the approved answer, exactly, then moves on to the patient the specialist has in mind. The second gives a rounded version of the number, attaches it to the wrong comparator, and spends the rest of the call recovering. Nothing in the launch dashboard distinguished them. Nothing will, until the CRM call notes or the prescribing data start to diverge, and by then the specialist has already formed a view of the product through the rep who got it wrong.
Multiply those two reps across a territory map and you have the difference between a launch that holds its message and one that drifts, with identical training metrics.
The trouble is that most launch reporting is built for the first state. Many learning system set-ups are configured to record completion, and completion is what they report. Commercial teams then borrow those numbers as a proxy for readiness because nothing better is on the table. The proxy is convenient. It is also, often, wrong in exactly the places that matter.
The iceberg: reassuring numbers above the waterline, gaps below it
Above the waterline, training metrics usually look reassuring. Attendance is high. Completion is close behind. The satisfaction survey says people enjoyed the kickoff and found the content relevant. Nobody is lying; the numbers are real.
Below the waterline sits what nobody reports: the specific facts each person gets wrong, and how many people share the same wrong answer.
In our case library, which covers 100 programs across 39 organisations between 2016 and 2026, we identified 794,130 knowledge gaps at first attempt, out of about 9 million answers. A gap is deliberately narrow in our definition: one person, wrong on one question, the first time they see it.
Across programs with gap fixing, the average is about 16 critical gaps per rep per year. Not trivia. Critical gaps: claims, safety information, key differentiators, the facts a positioning depends on. None of them would appear in a typical completion report, which is set up to record activity, not answers.
The programs identified 794,130 first-attempt gaps across different training and business contexts.
What ready actually looks like, in four properties
If trained is a record of activity, ready has to be a measurement of knowledge. In practice, we have found it needs four properties to be useful to a launch team rather than just interesting to L&D.
- Measured question by question: not a module score, but the answer each person gives to each question, collected through short challenges (five questions in sixty seconds) that people play on their phones between calls.
- Weighted by risk: critical gaps on claims, safety and key differentiators count for more than nice-to-know details, because being wrong about them costs more.
- Closed individually: each person gets the fix for their own gaps, and a gap only counts as closed when that same person answers correctly later. For critical gaps the bar is higher, 5 of 5 later check questions on the same concept, because one later right answer does not rule out a lucky guess.
- Visible to leaders: a readiness index by country, team and topic, next to the launch KPIs, from the first week.
None of these is exotic. Together they change the question the launch team can answer. Instead of "did we train them?", it becomes "which of our people are not yet ready on which facts, and what are we doing about it this week?"
The last property is the one that makes the other three matter. Readiness data that stays inside a learning report is a curiosity. Readiness data on the same page as early uptake, reviewed in the same weekly meeting, becomes a management tool.
Why this matters more now
We have made the first-year case in detail elsewhere, so briefly: launch volume is up (IQVIA estimates 73 novel active substances launched globally in 2025, above the past decade's average), many are in specialised indications on expedited timelines, and the first year matters. The field is being asked to master new, precise stories more often and faster. A process that assumes readiness from attendance was risky when launches were rarer. It is harder to defend now.
A fair objection: we already test, and our pass rates are high
A fair objection is that most launch teams already test their field. There is a certification at the end of the launch training, the pass rates are high, and the reps who fail get remediated. Why is that not readiness?
It is a start, and it is better than nothing. But a one-off certification has three weaknesses that matter at launch.
First, it is a snapshot on the day. Murre and Dros's 2015 replication of Ebbinghaus, a relearning experiment with nonsense syllables, shows that unrehearsed material fades fast and that spacing helps. It is not a measure of product knowledge, but it is reason enough to doubt that a pass on Tuesday tells you what survives to the first cycle meeting.
Second, a pass mark is an average. A person can pass comfortably while holding the wrong mechanism or rounding the primary endpoint, because the other questions carried them over the line. The certificate says ready. The conversation says otherwise.
Third, a test at the end cannot shape the training that came before it. By the time you know where the gaps are, the kickoff is over and the budget has been spent. Readiness measured before and during the launch can change what gets taught; readiness measured after can only report on it.
So keep the certification if you need it. Just do not let it stand in for readiness.
What the move from trained to ready looks like in practice
Two programs from our library show the shift in different settings.
A global MedTech annual readiness initiative invited 787 people across 25 countries and three channels, and 93% took part. That was an organisation-wide readiness program, not voluntary play, and the final reinforcement wave was opt-in. The design was a four-day battle followed by nine weeks of reinforcement, and it closed 57% of 15,758 gaps, each verified person by person. Our reading is that the battle created the moment and the reinforcement weeks gave people the repeat attempts that verification needs.
A consumer health program in the pharmacy channel ran always on for 139 days across five brand cycles, with 1,421 staff joining. The interesting finding there was where closure happened: it concentrated in the competitions. Content on its own, however well produced, moved far less. Our interpretation is that people corrected more when there was a reason to retrieve the answer, a colleague to beat and a result that mattered.
Across the library, the 16 programs tied to a business moment (a launch, kickoff, certification, congress or cycle meeting) closed 55% of their gaps, against 41% for the other 84. That is an observed association, not proof that the moment caused the difference. Still, a launch is the strongest business moment most teams will ever get. It would be a waste to spend it only on training.
Launch Excellence is a decision to stop assuming
None of this replaces what launch teams already do well. The kickoff still matters. Good content still matters. Managers in the field still matter most of all. What changes is the definition of done, and with it the data that gets looked at in the room where launch decisions are made.
We built Atrivity around that definition. One-minute challenges against real colleagues produce answer-level data without taking people out of the field. Battles in a synchronised window turn a kickoff or cycle meeting into a measurable moment. Gap Fixing sends each person the fix for their own mistakes and counts a critical gap as closed only at five of five later check questions on the same concept. Readiness indexes and the knowledge map by team, country and topic show leaders where the gaps sit, by team, country and topic.
But the tool is the second step. The first is a question to take into your next launch meeting: when we say the field is ready, what exactly are we measuring? If the honest answer is attendance and completion, you know what you have. You have a trained field. Whether it is a ready one is still an open question, and it is one you can answer before your specialists do.