What the AI found
“"Your most active clients consistently log 15% more non-exercise activity (NEAT) on days when they engage in structured strength training, suggesting a synergistic effect rather than displacement."”
Before
Disparate client movement logs, unclear patterns
After
Integrated movement insights for targeted coaching
The same system, three states — real screens, not a screenshot
| Client ID | Activity Type — Duration (min) — RPE — Notes |
| 001 | Walking — 60 — 4 — Felt good |
| 002 | Gym — 45 — 7 — Strength training |
| 003 | Yoga — 30 — 3 — Relaxed |
Prompt
Here is anonymised client movement data from a Google Sheet, including activity type, duration, perceived exertion (RPE), and daily non-exercise activity (NEAT) estimates. Identify non-obvious correlations between specific structured exercise types (e.g., strength training, cardio) and subsequent NEAT levels, over a 4-week period. Quantify any significant relationships.
Here is anonymised client movement data from a Google Sheet, including activity type, duration, perceived exertion (RPE), and daily non-exercise activity (NEAT) estimates. Identify non-obvious correlations between specific structured exercise types (e.g., strength training, cardio) and subsequent NEAT levels, over a 4-week period. Quantify any significant relationships.
AI
Analysis reveals a notable correlation: clients engaging in structured strength training sessions consistently showed a 15% increase in their average daily NEAT (Non-Exercise Activity Thermogenesis) on those training days, compared to days without structured exercise. This suggests that for this cohort, strength training appears to stimulate rather than detract from overall daily movement levels, which was a surprising finding given common assumptions about post-workout fatigue.+15%
Avg. NEAT increase post-strength
78%
Clients hitting daily step goal
1:1.7
Structured vs. Unstructured Activity Ratio
From Haphazard to Harmonised: Movement Coaching with AI
A practitioner refines client movement plans by integrating AI insights, shifting from guesswork to data-backed decisions.
A nutritionist running a small EU practice, advising clients on holistic well-being.
Tools used
The real tools used here — swap any for your own equivalent. Each links to how we’d set it up.
Starting state
Before anything was set up
Before the 10-Day Challenge, movement tracking for clients was a mixed bag. Some used fitness trackers, others jotted notes, and many simply reported how they "felt." Consolidating this into a coherent picture was time-consuming, often leading to broad, unspecific advice. The practitioner relied heavily on client self-reporting, making it difficult to discern actionable trends across their diverse clientele.
| Client ID | Activity Type — Duration (min) — RPE — Notes |
| 001 | Walking — 60 — 4 — Felt good |
| 002 | Gym — 45 — 7 — Strength training |
| 003 | Yoga — 30 — 3 — Relaxed |
| 001 | Desk Work — 480 — 2 — Sedentary day |
| 002 | Walking — 90 — 5 — Commute |
Working state
10-Day Challenge, doing its job
The practitioner began by uploading anonymised, aggregated client activity data (collected via a standard form) into a Google Sheet. They then used Gemini to analyse these patterns. The goal was to identify overlooked correlations between different types of movement and client well-being parameters, establishing a more evidence-based foundation for coaching recommendations. The prompt was meticulously crafted to guide the AI toward uncovering non-obvious relationships in the data.
Prompt
Here is anonymised client movement data from a Google Sheet, including activity type, duration, perceived exertion (RPE), and daily non-exercise activity (NEAT) estimates. Identify non-obvious correlations between specific structured exercise types (e.g., strength training, cardio) and subsequent NEAT levels, over a 4-week period. Quantify any significant relationships.
Here is anonymised client movement data from a Google Sheet, including activity type, duration, perceived exertion (RPE), and daily non-exercise activity (NEAT) estimates. Identify non-obvious correlations between specific structured exercise types (e.g., strength training, cardio) and subsequent NEAT levels, over a 4-week period. Quantify any significant relationships.
AI
Analysis reveals a notable correlation: clients engaging in structured strength training sessions consistently showed a 15% increase in their average daily NEAT (Non-Exercise Activity Thermogenesis) on those training days, compared to days without structured exercise. This suggests that for this cohort, strength training appears to stimulate rather than detract from overall daily movement levels, which was a surprising finding given common assumptions about post-workout fatigue.Use case implemented
The finished system, running on its own
With the system established, the practitioner now has a clear dashboard offering weekly insights into client movement patterns. This allows for proactive adjustments to exercise recommendations and educational materials, tailoring advice to individual needs rather than relying on generic guidelines. The AI now serves as a quiet, ever-vigilant assistant, highlighting potential areas for optimisation and ensuring the coaching remains precise and effective. The new workflow saves valuable time, freeing up more moments for direct client engagement.
+15%
Avg. NEAT increase post-strength
78%
Clients hitting daily step goal
1:1.7
Structured vs. Unstructured Activity Ratio
What an outside observer would notice
2 hours
Time saved on data analysis per week
Up 20%
Client engagement with movement plans
Significantly increased
Specificity of coaching recommendations
The stack — build it yourself
Ubiquitous, flexible for rapid data input from various client sources, and straightforward for initial data cleaning.
Its natural language interface allowed for complex data queries without needing to write code, making sophisticated analysis accessible for a busy practitioner.
These are the tools used in this story. Any can be swapped for an equivalent you already trust.
Go deeper
Do this yourself
Discover how AI can refine your practice
This story runs on 10-Day Challenge. The tools and prompts above are the real build — swap any tool for your own equivalent and follow the same steps.