Getting better fitness planning results from AI requires moving beyond generic requests — "give me a workout plan" — to specific, constrained prompts that provide context about your goals, starting point, constraints, and preferences. The quality of the plan is directly proportional to the quality of the prompt. This concept covers the prompting approach that produces AI fitness plans worth following rather than plans worth ignoring.
Prompt engineering is the art of asking AI questions in a way that gets you useful answers. It sounds simple—just ask better—but there's real skill involved. In health and fitness contexts, the difference between a vague prompt and a detailed one is often the difference between generic advice and something genuinely tailored to your life.
A vague prompt: "Create a workout plan." A better prompt: "I have 45 minutes, 4 days a week, access to a home gym with dumbbells up to 50 lbs and a pull-up bar. I'm training for hiking endurance but also want upper-body strength. I have a history of lower back sensitivity. Create a 4-week plan."
That second version gives the AI context. It knows your constraints (time, equipment), your goals (hiking endurance + strength), and your limitations (back sensitivity). The output will be vastly more useful.
Constraints: Time available, equipment access, location, injuries or limitations, dietary restrictions, sleep consistency.
Goals: Be specific. "Get fit" is vague. "Train for a 10K in 12 weeks," "lose 15 lbs while maintaining muscle," or "improve flexibility to touch my toes" are specific.
Context: Your current fitness level, how long you've been training, what you've tried before and how you responded, your schedule quirks (shift work? unpredictable schedule?).
Format preference: Do you want a written plan, a spreadsheet, week-by-week breakdown, or daily instructions?
All of this information—constraints, goals, context, format—helps the AI tailor its response instead of defaulting to a generic template.
AI systems, especially language models like ChatGPT, are literal. They respond to what you ask. If you ask for "a workout," you'll get a workout—but it might be for someone very different from you. The AI doesn't have access to your fitness history, your specific injuries, or your equipment. It has to guess or default to something safe and generic.
When you provide context, you're essentially having a conversation where you're teaching the AI about you. Each detail narrows the solution space, making the output more relevant.
Good prompt engineering also includes follow-up. After the AI generates a plan, ask clarifying questions: "Can you adjust week three to include more lower-body focus?" or "Can you explain why you chose this exercise order?" Each follow-up refines the output. You're not asking the same question twice; you're iteratively building a better answer.
This is especially useful for nutrition, recovery, and injury prevention where your specific situation matters enormously. A plan for injury prevention needs to know which injuries you're at risk for.
Try this: Write two fitness prompts for ChatGPT or Claude: one vague ("Give me a workout routine") and one detailed (include your time, equipment, goals, and constraints). Compare the outputs. You'll immediately see why specificity matters. Then save the detailed format and reuse it—it becomes your template for future fitness requests.
Peri can explain this concept, give practical examples, help you decide whether it applies to your situation, or recommend a journey if appropriate.
Explore related journeys or tell Peri what you're working through.