Every center that takes training seriously runs mock calls. A training officer reads from a script, the trainee answers, and somebody scribbles notes about what went well and what didn't. It works, sort of. But the math behind hand-run mock calls quietly works against you, and in a year when retention sits at 3-5 years and centers run 30% or more below authorized strength, you can't afford training that burns your best people's hours while missing the gaps that wash recruits out.
Here are five mistakes centers make when training on mock calls by hand, and how to close each one.
1. You spend your scarcest resource running the script
Your training officers are also your most experienced telecommunicators. Every hour a CTO spends playing a hysterical caller from a paper script is an hour they are not on the floor, not mentoring, and not recovering from their own shift. When the floor is already underwater, that trade is brutal.
The fix is to stop spending a senior dispatcher to generate a stimulus a machine can generate better. Dispatch360 drives the caller with an AI voice that carries real emotion, hesitation, and escalation, and it reacts to how the trainee handles the call. The trainee gets a live, unpredictable caller. Your CTO gets their hours back.
2. The caller is too predictable
A human reading a script knows the answers, so they leak them. They soften when the trainee struggles. They answer the question the trainee meant to ask instead of the one they actually asked. Real callers don't do any of that. They give a wrong address, talk over you, and go quiet at the worst moment.
A trainee who only ever practices against a cooperative script is not ready for the headset. An AI caller that withholds, escalates, and responds to call control teaches interrogation and control under genuine uncertainty, which is the whole point.
3. Scoring is inconsistent and unrepeatable
Two CTOs grade the same call two different ways. The same CTO grades differently at hour two of a double than at hour fourteen. Notes get lost. "Good job, work on your callback" is not evidence, and it is not something a struggling trainee can act on.
Dispatch360 scores every session against an immutable ground-truth snapshot on a consistent rubric: call control, caller interrogation, location and callback, protocol adherence, CAD accuracy, and time to dispatch. Every session writes to an append-only audit trail. When a trainee asks why they lost points, you have the call, not a memory.
4. You train the conversation but not the console
Hand-run mock calls test whether a trainee can talk. They rarely test whether the trainee can talk, type, run an NCIC lookup, and watch a map at the same time. That gap matters more every year. Incoming applicants are smartphone-native and arrive with measurably weaker keyboard and multi-window desktop skills, and NG911 is piling text, images, and richer location data onto the call-taker's workload.
Build console fluency on purpose. Typing360 sets the floor: timed WPM and accuracy drills, dictation, and CAD-narrative composition. Speed is the floor, not the goal. Then Dispatch360 puts the whole surface in front of them at once, incident entry, person and vehicle lookups, radio traffic, and a live geocoded map, so the trainee learns to do the job, not just describe it.
5. The reps don't scale, and they don't repeat
A hand-run mock call happens once, when a CTO is free. If a trainee blows the location protocol, you can't replay that exact call on demand at 9 p.m. when the trainee has time to drill it. So the rep that would close the gap never happens, and the gap shows up in week six as a washout you could have caught in week two.
Simulation makes the reps repeatable and safe to fail at. Dispatch360 ships 24 versioned scenarios across medical, crimes in progress, traffic collisions, fire and rescue, domestic disturbances, mental-health crises, alarms, and multi-agency incidents. Trainees retry to close the gaps. Instructors assign required attempts with due dates, watch a review queue, and export progress as program evidence. The reps happen whether or not a CTO is in the room.
The point
You can't out-hire a retention problem, but you can out-train your washout rate. Hand-run mock calls aren't wrong, they just don't scale, don't repeat, and cost you the hours of the people you can least afford to lose. Simulation builds call control, console fluency, and stress tolerance before a trainee ever touches a live headset, and it does it without burning your bench.
If you want to see how Dispatch360 and the rest of the PSS360 platform run through STACC for your program or center, request a demo. We'll walk your training the way a CTO would.