What the AI found
““Clients consuming over 35g of fibre daily consistently exhibit a 15% lower average fasting glucose compared to those below 25g, a stronger correlation than any specific macronutrient ratio.””
Before
Client notes scattered across apps
After
Actionable, data-backed dietary insights
The same system, three states — real screens, not a screenshot
| Client ID | date — fasting glucose — post-meal glucose — total carbs — total fat — total protein — total fibre |
| A001 | 2023-11-01 — 5.8 — 7.2 — 180 — 60 — 90 — 22 |
| A002 | 2023-11-02 — 6.1 — 8.1 — 210 — 70 — 100 — 28 |
| A001 | 2023-11-03 — 5.7 — 6.9 — 170 — 55 — 85 — 38 |
Prompt
Analyze the provided Google Sheet data. Identify correlations between daily nutritional intake (total carbs, fat, protein, fibre) and fasting glucose levels across all clients. Specifically, quantify any surprising relationships. Focus on fibre intake vs. fasting glucose.
Analyze the provided Google Sheet data. Identify correlations between daily nutritional intake (total carbs, fat, protein, fibre) and fasting glucose levels across all clients. Specifically, quantify any surprising relationships. Focus on fibre intake vs. fasting glucose.
AI
Okay, after analysing the data, a notable pattern emerges regarding fibre intake and fasting glucose. Clients consuming over 35g of fibre daily consistently exhibit a 15% lower average fasting glucose (mean of 5.6 mmol/L) compared to those below 25g (mean of 6.6 mmol/L). This correlation is stronger than that observed with variations in specific macronutrient ratios within the provided ranges.5.6 mmol/L
Average Fasting Glucose (High Fibre)
6.6 mmol/L
Average Fasting Glucose (Low Fibre)
15% Lower Glucose
Fibre Intake Impact
From Scattered Notes to Targeted Blood Sugar Insights
A nutritionist moves from disparate client observations to precise, AI-driven dietary recommendations in under 10 minutes.
A nutritionist running a small EU practice, focusing on metabolic health.
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 implementing the new system, client data for this busy nutritionist was a mosaic of handwritten notes, spreadsheet entries, and various food diary screenshots. Extracting actionable patterns across multiple clients or even a single client's long-term data was a time-consuming, often manual, process. This made identifying subtle but significant dietary impacts on metabolic markers challenging, leading to generalised advice rather than precise recommendations.
| Client ID | date — fasting glucose — post-meal glucose — total carbs — total fat — total protein — total fibre |
| A001 | 2023-11-01 — 5.8 — 7.2 — 180 — 60 — 90 — 22 |
| A002 | 2023-11-02 — 6.1 — 8.1 — 210 — 70 — 100 — 28 |
| A001 | 2023-11-03 — 5.7 — 6.9 — 170 — 55 — 85 — 38 |
| A003 | 2023-11-04 — 5.5 — 7.0 — 190 — 65 — 95 — 31 |
Working state
All-Access, doing its job
The nutritionist consolidated anonymised client dietary and glucose data into a Google Sheet. They then used Google's Gemini to analyse this structured data. The prompt clearly outlined the data fields and requested correlations between dietary components and metabolic markers, specifically blood glucose levels. The AI processed weeks of data across multiple clients in moments, identifying a non-obvious relationship.
Prompt
Analyze the provided Google Sheet data. Identify correlations between daily nutritional intake (total carbs, fat, protein, fibre) and fasting glucose levels across all clients. Specifically, quantify any surprising relationships. Focus on fibre intake vs. fasting glucose.
Analyze the provided Google Sheet data. Identify correlations between daily nutritional intake (total carbs, fat, protein, fibre) and fasting glucose levels across all clients. Specifically, quantify any surprising relationships. Focus on fibre intake vs. fasting glucose.
AI
Okay, after analysing the data, a notable pattern emerges regarding fibre intake and fasting glucose. Clients consuming over 35g of fibre daily consistently exhibit a 15% lower average fasting glucose (mean of 5.6 mmol/L) compared to those below 25g (mean of 6.6 mmol/L). This correlation is stronger than that observed with variations in specific macronutrient ratios within the provided ranges.Use case implemented
The finished system, running on its own
With the insights from Gemini, the nutritionist now had a streamlined, evidence-based approach to client consultations. Each week, new data is appended to the Google Sheet. A quick re-run of the prompt in Gemini provides updated correlations, allowing for continuously refined, data-driven dietary adjustments. This system supports more precise, individualised client plans, enhancing the clinic's efficacy and reputation.
5.6 mmol/L
Average Fasting Glucose (High Fibre)
6.6 mmol/L
Average Fasting Glucose (Low Fibre)
15% Lower Glucose
Fibre Intake Impact
What an outside observer would notice
Reduced by 80%
Time spent identifying dietary patterns
Increased by 25%
Client dietary plan precision
Up 30%
Client engagement with fibre goals
The stack — build it yourself
Familiar, accessible, and easily structured for quantitative analysis.
Excels at nuanced data interpretation and correlating disparate variables beyond simple averages.
These are the tools used in this story. Any can be swapped for an equivalent you already trust.
Go deeper
Do this yourself
See All-Access
This story runs on All-Access. The tools and prompts above are the real build — swap any tool for your own equivalent and follow the same steps.