AI Assistant
The AI Assistant is a coaching support tool built into Good Coach App. It analyses athlete data, generates structured training plans, and provides post-session insights — drawing on each athlete's training history, health state, schedule preferences, and your coaching methodology to produce contextually relevant output. The goal is to reduce the time coaches spend on routine planning and analysis, so more of that time goes toward the decisions that genuinely require human judgment.
The AI Assistant is not a replacement for a coach. Every output it produces is a starting point, not a final answer. It does not know things you have not told it, cannot feel how an athlete is responding, and has no stake in the outcome. A plan reviewed, adjusted, and approved by an experienced coach is always better than one that was not. The tool is designed to handle the mechanical work — drafting, calculating progressions, surfacing patterns in data — while leaving the judgment calls where they belong.
How the generators connect
The generators are most powerful when used in sequence. Each one feeds context into the next — primarily through AI Assistant memory, which accumulates knowledge about the athlete and your coaching approach over time and is read automatically before every generation.
1. Start with training cycles
Training cycles generation is the natural starting point for a new athlete or a new training season. You do not need a full-year plan — a single mesocycle or a short macrocycle covering the next 10–12 weeks is enough to give week generation the phase context it needs. Once set up in Good Coach App, this structure is read automatically by all other generators: week generation knows what phase the athlete is in, and analyses can assess whether training actually matched the phase intent.
The cycles generator also improves over time. After months of training and analyses, athlete memory contains a detailed picture of how this athlete actually responds — their effort calibration, injury patterns, how they handle loading. When you generate cycles again for the next season, that accumulated knowledge feeds into the plan, producing periodization that is shaped by what you have learned about this specific athlete rather than general principles alone.
2. Analyse each completed session
Single workout analysis examines one completed session: how well it was executed, how effort matched the data, what the intervals showed. After each analysis, athlete memory is updated — pacing tendencies, effort calibration patterns, how the athlete responds to specific workout types. The more sessions you analyse, the richer the picture the AI Assistant has of this athlete when generating future weeks.
3. Review the broader pattern periodically
Period analysis steps back from individual sessions to identify what is visible only across multiple weeks: training load trends, recovery patterns, volume progression compliance, overtraining signals. It updates athlete memory with these medium-term observations, giving future week generations a more accurate read on where the athlete actually is — not just where the numbers say they should be.
Memory as the connective tissue
None of the generators above work in isolation. AI Assistant memory is what makes each generation better than the last. It maintains two profiles — one for your coaching methodology, one for each athlete — built automatically from your analyses and coaching decisions, and read silently before every generation. An athlete with several months of analyses and multiple training cycles behind them will receive plans and reviews shaped by everything the AI Assistant has learned about them, without you having to restate any of it.