Real-Time Feedback Framework for VR Training
Established a feedback framework for Axon’s VR training platform — partnering with Engineering and Art to define how real-time feedback and after-action reporting are categorized, prioritized, and delivered across headset, tablet, and web.
IMPACT
A reusable feedback framework — replacing ad hoc, per-exercise decisions with a system, shortening the path to shipped feedback for PMs, Designers, and Engineers.
Built one callout at a time
Real-time feedback is the signal a trainee gets in the seconds between an action and the system’s response, while the dummy is still in range. Each new drill added another tag. Nobody owned the list as a system.
PMs, designers, and engineers were making the same callout decisions per exercise. That was slow, and it did not compose. Two surfaces sat on the same mess: live feedback during the drill, and the After-Action Report after it.
Eight things talking at once, none of them ranked
By the time the pattern was audited, a single drill could surface up to eight overlapping callouts. Three failure modes sat under the clutter: too much at once, the wrong thing at the wrong moment, and every message delivered the same way.
Realtime feedback and After-Action Report looked like different products. Underneath, they shared one ungoverned system. Nothing decided what got shown, when, or how urgently.
Three axes, in that order, before any visual pass
The fix started as a category exercise, not a UI exercise.
DECISION 1 · SORT BY FUNCTION
Every existing tag landed in one of five functions: Instructional, Shot-Specific, System Event, Result, Mixed-Use. If you do not know what a piece of feedback is, you cannot decide when or how to show it.
DECISION 2 · PLACE IT ON THE DRILL TIMELINE
Instructional feedback belongs in preparation, before the dummy appears. Shot-specific and system-event feedback belong while the dummy is active. Results belong after the dummy disappears, when nothing else should compete for attention.
DECISION 3 · RANK HOW LOUDLY IT SPEAKS
Almost everything had been world-locked UI. Five modalities got ranked, then paired: informational and system-event with voice-over and optional light; shot-specific with HUD or attached UI, never both; results with standalone UI.
Three tenets, named only after the pattern was real
Mapping every item against every modality it actually used produced the same three questions. Those became the tenets, not the other way around.
MINIMIZE COGNITIVE LOAD
At most two pieces of information at once. If two modalities fire on the same event, they say the same thing.
DESIGN FOR THE TRAINEE’S PERFORMANCE
If it helps in this drill, say it now. If it helps in the next drill, put it in the mini-AAR. If it helps in the next exercise, wait for the full AAR.
DRIVE CONSISTENT MENTAL MODELS
Instructional always gets the same pairing. Users should know where to look before the callout arrives.
