Notes on method.

Attendance is not an outcome

Most training reports lead with two numbers: how many people attended, and how satisfied they were. Both are worth recording. Neither tells you whether anything changed.

Attendance measures logistics. Satisfaction measures the experience of the day — the pacing, the facilitator, the lunch. An organization can score highly on both and return on Monday to exactly the practice it had on Friday. When those numbers are presented as results, they quietly redefine success downward: the program becomes accountable for being pleasant rather than for working.

At Ztnexus, satisfaction and attendance are recorded as operational data and never presented as outcomes. Outcomes are defined in the Design step, before delivery begins, in three layers: can participants demonstrate the capability in realistic conditions; does it hold weeks later; and did the indicators the partner cares about move. If a measure can be maximized by a program that changes nothing, it is not an outcome measure.

The discipline is inconvenient, which is rather the point. Defining outcomes first constrains the design — and that constraint is what makes the close-out report worth reading.

Fast without cutting corners

There's a common assumption that rigor and speed trade off — that a program launched in weeks must be shallower than one incubated for a year. Look closely at how long projects actually fail, though, and the assumption falls apart. Long lead times don't usually buy rigor; they buy drift. Sponsors change, budgets move, and the discovery findings age until the program launches into an organization that no longer matches them.

Rapid deployment works when the sequence is respected, not skipped. Discovery and synthesis are done properly and quickly because they are scoped to a decision, not to a literature. Design produces the outcome framework alongside the curriculum, so evaluation never becomes a bolt-on. And because measures are agreed before delivery, a fast launch carries no risk of "we'll figure out how to judge this later."

Speed, in other words, is a property of a well-ordered method — the five steps exist precisely so that moving quickly through them removes nothing.

What a retention check actually tests

Weeks after delivery, we return. Not with a quiz — recall is the least interesting thing to test — but with the conditions of the work: a case, a scenario, a live decision with incomplete information and a clock. The question is never "do you remember the framework?" It is "does the framework show up in how you perform when it matters?"

Retention checks are scheduled into every engagement during Design, so they are expected rather than sprung. Their findings go into the close-out report either way: capability that held, capability that faded, and what that implies for the partner's next step. A faded result is not a failure to hide — it is exactly the information an evidence-led institution should want, because it distinguishes what the organization has absorbed from what it has merely encountered.

If a provider never checks back, ask what they think would happen if they did.