Four nets and one clipboard
The instructor stands behind the operations tent with a headset split four ways and a clipboard braced on his forearm. Two squads are working a lane on the far side of the ridge. A third is running the medevac drill. A fourth is passing spot reports up to the command net. He is one man grading four conversations by ear.
He catches some of it. A call sign dropped on the fires net. A report given out of order. But while he writes down one error the other three nets keep talking, and whatever happened on them is gone. Nothing was recorded. The clipboard holds maybe a third of the truth, and the debrief gets built from that third.
Every skill the military cares about gets a number. Marksmanship has a scored target. Land nav has a punched card. The physical test has a stopwatch. Radio discipline has a man with a clipboard hoping he was listening to the right net at the right moment. And radio discipline is the skill the whole formation runs on.
The line that costs the hoist
The 9-line MEDEVAC request exists because a pilot needs nine specific facts before the aircraft is worth launching. Line 1 is the pickup location. Line 2 is the frequency and call sign at the site. Line 4 is special equipment. Line 4 is the one that tells the crew to bring a hoist.
Under stress line 4 is the one that vanishes. The casualty is at the bottom of a ravine. The RTO is breathing hard and jumps from line 3 straight to line 5. Nobody on the net catches it. The aircraft launches without a hoist, arrives over a pickup site it cannot use, and flies back to get one. In training that is a note in a debrief. On a real casualty that is minutes, and a femoral bleed does not give you minutes.
That failure is completely trainable. But it only gets trained if somebody catches it every single time it happens, and a clipboard stretched across four nets does not catch it every time. Most reps the skipped line dies in the air and the RTO walks away believing he sent a clean 9-line.
Score the call, keep the audio
EdgeLance Coach listens to the net so the instructor does not have to. It transcribes live radio traffic on the device in the room, on whatever hardware the schoolhouse already owns. It attributes every transmission to a speaker by voice fingerprint, so the transcript reads like a script with names on it instead of a wall of anonymous audio. And it scores each call against doctrine. Brevity codes. Call sign usage. Report formats like the 9-line and SALUTE, checked line by line, so a skipped line 4 becomes a finding the moment the transmission ends.
The instructor surface shows per-operator scoring with the audio anchored to every finding. A reviewer taps the finding that says line 4 was skipped and hears the actual transmission where it was skipped. Nobody argues about what was said. The receipt is the recording.
That changes the debrief from memory against memory into evidence. An instructor who tells a team leader he dropped his call sign four times on the fires net now says it with four recordings attached. The team leader listens to himself do it. Adults fix what they can hear.
Nobody measures the decay
Comms discipline is a perishable skill. It is trained by ear and sustained by habit, and it degrades the moment real stress arrives. Everyone who has run a lane knows this. Almost nobody can put a number on it, because the errors evaporate as fast as they are made. A unit can go years believing its comms are tight because no measurement ever said otherwise.
So the first honest measurement is a shock. The first time a squad sees its scored transcript from a stress lane, the score is worse than anyone in the squad expected. That shock is the product working. The Modern War Institute argues that the fastest path to drone dominance is to let the squad fail in training, because squad-level failure is where learning is cheap. Cheap failure only teaches if the failure is visible. Coach makes the comms failure visible down to the transmission.
Chatham House's study of Ukraine's wartime innovation reaches the same place from the other direction. The units that adapt fastest are the ones that train and iterate at their own level instead of waiting for the institution. A comms scoring tool that a single training cadre can install and run this week fits that pattern. It does not wait for a program of record to decide radio discipline matters.
Stands alone in the schoolhouse, feeds the AAR in the field
Coach deploys as a signed standalone mission pack. A training command installs it on hardware it already owns and runs it in a room with no network connection at all. Everything runs local. The recordings stay in the room, which means the squad can fail hard in front of the software without that failure living on a server somewhere waiting to be misread.
And the same scoring engine runs inside full EdgeLance missions. Every net a mission carries gets the same doctrine scoring, and the results feed the after action review next to everything else the mission recorded. The skill measured in the schoolhouse keeps being measured on operations, by the same yardstick, with the same audio receipts.
The instructor behind the tent puts down the clipboard and picks up a surface that heard all four nets, scored every call, and kept the tape. The next time that squad runs the medevac lane, line 4 is in the transmission.