The One-Week Follow-Up: A 5-Minute Survey That Tells You If Training Transferred

Seven days after an Operation Aetherfall session ends, every player gets an email. Inside is a survey – thirteen items, about five minutes – and the first question is the only one anyone remembers: have you run into the situation you wrote on your card at the end of the workshop?

A training follow-up survey sent one week after the session is the earliest honest check on whether training transferred. It goes out by email on day seven and asks four things: whether participants have used the if-then plan they wrote during the debrief, whether the trained skills have shown up in real work incidents, whether the experience is still on their minds, and whether they've talked with anyone from the session since. Five minutes of each participant's time, and it tells you more about transfer than everything you collected in the room.

That's a strong claim, so here's the reasoning. Everything you measure on the day of the training – reactions, confidence ratings, pre/post shifts – measures a person still inside the experience. Transfer happens outside it. By default, only 10-15% of trained skills make it back to the job, and you can't see which side of that line your program landed on until people have been back at work long enough for work to push back. The follow-up instrument we built for Conflict Campaign's pilot came out of the same research synthesis as the rest of our measurement design – 200+ studies on instructional design – and it's the only piece pointed at behavior rather than intention. (The structured debrief is still the highest-leverage part of the whole wrapper. The survey just tells you whether it worked.)

Why is one week the right window?

One week hits the overlap of three clocks: decay, opportunity, and memory.

The decay clock starts the moment the session ends. Implementation intentions formed in isolation, without follow-up support, produce diminishing effects over time, and the research on spaced reminders recommends the heaviest cadence in the first week – daily nudges, tapering to weekly – because that first week is when a new response either starts wiring to its cue or doesn't. Wait a month to check in and you're not measuring the decay curve. You're standing at the bottom of it.

The opportunity clock runs slower. An if-then plan names a specific work situation, and one workweek is usually enough for that situation to occur – a week contains a status meeting, a slipped estimate, and at least one disagreement about scope, which is all most plans need. Ask at day seven and "it hasn't come up yet" is still a perfectly plausible answer, while the other answers already mean something.

The memory clock is the one people forget. At seven days, participants can still describe specific incidents – the actual conversation where they tried the new question, the actual moment they defaulted to their old pattern. Those open-ended descriptions are the richest data a small pilot produces, and they blur fast.

There's a fourth reason that isn't a clock: position. The research recommends a booster session 2-4 weeks after training – a 30-minute virtual check-in to troubleshoot application. The one-week survey sits exactly between session and booster, and its answers decide what the booster needs to cover.

One honest limit. A week is far too soon to claim durable change. This is an early indicator of what Kirkpatrick's model calls Level 3 – behavior on the job – and early is the point. You're not proving transfer at day seven. You're catching the plans that are already in trouble while there's still time to help.

What should a training follow-up survey ask?

Four things: implementation-intention follow-through, skill-application incidents, salience, and peer connection. Our pilot instrument covers them in four short sections plus one open question, and every item maps to a transfer mechanism rather than to curiosity.

The survey opens with the implementation intention. Have you encountered the situation described in your plan since the workshop? Three answers, all legitimate: yes and I tried my planned response; yes, but I responded the way I usually do; no, it hasn't come up yet. If they tried it, an open field asks how it went. If they didn't, a checklist asks what got in the way – more on that checklist below, because it's the most diagnostic part of the whole instrument.

Then skill application: four yes/no/not-sure items, each matched to a skill the session practiced. In the past week, have you asked someone "What's driving that?" during a disagreement? Shared an uncomfortable concern with someone in authority? Made a deliberate effort to gather everyone's input before a decision? Noticed a disagreement starting to feel personal? Each "yes" gets an optional one-line description. These items catch transfer the written plan can't – a participant whose if-then targeted managing up may still have started asking better questions in peer conflicts.

Then salience, three 1-5 ratings: how much have you thought about the workshop this week, how relevant does it feel now that you're back in the work, and how likely are you to keep trying the approach you committed to. Salience isn't transfer, but a cohort that's stopped thinking about the training by day seven has told you something worth hearing.

Last, peer connection: have you talked with anyone from the workshop about what you learned, and would a brief 30-minute virtual follow-up in 3-4 weeks be valuable? These two earn their place because peer support is the single strongest environmental predictor of sustained transfer, and because the second question lets participants opt into the booster instead of having it imposed on them.

A final open question – anything else about how the workshop has or hasn't shown up in your work – catches whatever the structure missed. Thirteen items total. Nobody has ever abandoned a survey for being too short.

How do you send it so people actually answer?

Digital, by email, seven days out, sized to five minutes, and framed before anyone leaves the room. Most post-training follow-up fails by being either too late or too long; this one is neither.

The framing starts at the session close, not in the email. The script we use is two sentences: "You'll get a brief survey in one week. It's not a test – it's a chance for you to reflect on whether the plan is working and what's getting in the way." Adam's a licensed professional counselor; "it's not a test" is more or less his native register. The line does two jobs. It sets the expectation so the email doesn't land as a cold audit, and it pre-frames imperfection as normal, so "I responded the way I usually do" feels like an answer rather than a confession.

The answer options carry the same message structurally. No scoring, no red flags, no "why not?" branch. The default-response option sits right next to the success option, same weight. The barrier checklist normalizes obstacles by listing them – forgetting, stakes, uncertainty – as expected outcomes. People tell you the truth about not changing their behavior only when the instrument treats not changing as the statistically normal thing it is.

The follow-up is an intervention, not just a measurement

A follow-up survey re-activates the commitment it's checking on, and that's a feature.

The 2025 meta-analysis of implementation intentions – 642 independent tests – found that plans rehearsed after they're written show larger effects than plans written and filed. Reading question one means re-reading your own plan. That's a rehearsal. The same literature recommends brief digital nudges that restate a participant's if-then plan to keep it accessible while the habit forms; the day-seven survey is the heaviest nudge in that sequence, arriving right where the daily-to-weekly taper would put it. Even the skill-application items work this way – a participant who reads "have you asked someone what's driving that?" and answers no just got the cue re-planted for next week.

The peer-connection questions prime two more mechanisms with well-documented numbers behind them. People who wrote down goals and shared progress with a friend achieved 76% follow-through versus 43% for people who merely thought about their goals, and scheduled accountability check-ins pushed success to 95% (Matthews, 2015). Our debrief pairs accountability partners with a two-week check-in instruction; the day-seven survey lands in between, asking whether they've talked yet.

So yes, the data is slightly contaminated in the strictest sense – you're measuring people you've just nudged. For a pilot, take that trade every time. You're not running a controlled study. You're trying to get one more person across the transfer gap, and the measurement is doubling as support.

How do you read the barrier signals?

Each barrier answer points at a different fix. The checklist in question three is where the survey pays for itself:

What they checked	What it usually means	What to do
"Haven't had the opportunity yet"	Nothing's broken – the plan is idle, not dead	Keep it warm with a reminder; check again at the booster
"Forgot about the plan in the moment"	Cue problem – the if-clause isn't firing when the situation hits	Rewrite the cue as an internal state ("if I notice myself getting defensive...") and add spaced reminders
"The situation felt too different from what I practiced"	The plan was over-specified for a messy, variable skill	Broaden the cue from one scenario to a situation type; internal-state cues generalize where external ones don't
"The real situation felt higher-stakes than I expected"	Missing coping plan	Add the second layer – "if it goes badly, I will..." – and make it the booster's first topic
"I wasn't sure the new approach would work"	Confidence gap, not a knowledge gap	An accountability-partner conversation, or peer success stories at the booster

Read the column at cohort level too. If half your participants checked "forgot in the moment," that's not five personal failures – it's a template problem in how the session helped people write their cues, and it's fixable before the next cohort. The research is candid that repeated, context-variable interpersonal situations are the hardest case for if-then planning, because varied opportunities keep any single cue from staying distinctive. Barrier data matters more for conflict skills than it ever would for gym attendance. And since coping plans become more instrumental over time as people hit real-world obstacles, the barriers your cohort reports at day seven are the raw material for the coping plans they actually need.

The default transfer rate means roughly one in seven participants sustains a behavior change without support. Some people will return to old patterns no matter what you send them – the research allows nothing rosier. The survey's job is to find the ones who are still reachable, a week out, while the plan on their card is still warm.

The asking is part of the answer.