What the AI found
“Your post-lunch energy dips are strongly correlated with lunches containing processed grains and refined sugar, showing a 35-40% drop in perceived energy levels within 90 minutes, compared to only a 10-15% drop after protein and vegetable-rich meals.”
Before
Disparate Logs, Persistent Fatigue
After
Clear Patterns, Sustained Vitality
The same system, three states — real screens, not a screenshot
| Date | Food Items — Energy Score (1-10) |
| 2024-03-01 | Pasta, bread, soda — 4 |
| 2024-03-02 | Chicken salad, water — 8 |
| 2024-03-03 | Sandwich, crisps, juice — 5 |
Prompt
I've been tracking my food intake and subjective energy levels (1-10) for the past month. I suspect my afternoon energy dips are food-related. Can you analyse the patterns in my attached Google Sheet data (columns A=Date, B=Lunch Items, C=Energy Score 90min post-lunch) and tell me which specific food types or meal compositions most consistently predict a significant drop in my energy score? Quantify the average energy score difference associated with these patterns.
I've been tracking my food intake and subjective energy levels (1-10) for the past month. I suspect my afternoon energy dips are food-related. Can you analyse the patterns in my attached Google Sheet data (columns A=Date, B=Lunch Items, C=Energy Score 90min post-lunch) and tell me which specific food types or meal compositions most consistently predict a significant drop in my energy score? Quantify the average energy score difference associated with these patterns.
AI
Based on your data, lunches containing processed grains (e.g., pasta, white bread) and refined sugars (e.g., soda, pastries) are strongly correlated with the most significant energy drops. These meals show an average energy score of 4.2 within 90 minutes post-lunch, representing a 35-40% decrease from your pre-lunch baseline. In contrast, lunches rich in lean protein and vegetables consistently result in an average post-lunch energy score of 7.8, with only a 10-15% perceived decrease.7.5/10
Avg. Post-Lunch Energy
1-2/week
Days with 4pm Slump
+45 min
Avg. Focus Duration
From Hazy Notions to Concrete Energy Insights
A single AI prompt transforms scattered self-tracking data into a focused plan for consistent energy.
A 38-year-old marketing consultant in Northern Europe, managing a demanding schedule and two young children.
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
For months, Sarah meticulously logged everything she thought affected her energy: sleep, coffee intake, exercise, even mood. The data lived across several apps and a notebook, a sprawling collection of entries that felt more like a chore than a source of insight. Despite the effort, her afternoon energy dips persisted, leaving her frustrated and without a clear direction.
| Date | Food Items — Energy Score (1-10) |
| 2024-03-01 | Pasta, bread, soda — 4 |
| 2024-03-02 | Chicken salad, water — 8 |
| 2024-03-03 | Sandwich, crisps, juice — 5 |
| 2024-03-04 | Lentil soup, rye bread — 7 |
Working state
All-Access, doing its job
Convinced the solution lay buried in her own data, Sarah compiled her recent logs into a Google Sheet. She then turned to Gemini, feeding it a concise prompt designed to cut through the noise and identify specific dietary correlations with her afternoon energy levels. The AI’s output offered an unexpected, quantifiable pattern.
Prompt
I've been tracking my food intake and subjective energy levels (1-10) for the past month. I suspect my afternoon energy dips are food-related. Can you analyse the patterns in my attached Google Sheet data (columns A=Date, B=Lunch Items, C=Energy Score 90min post-lunch) and tell me which specific food types or meal compositions most consistently predict a significant drop in my energy score? Quantify the average energy score difference associated with these patterns.
I've been tracking my food intake and subjective energy levels (1-10) for the past month. I suspect my afternoon energy dips are food-related. Can you analyse the patterns in my attached Google Sheet data (columns A=Date, B=Lunch Items, C=Energy Score 90min post-lunch) and tell me which specific food types or meal compositions most consistently predict a significant drop in my energy score? Quantify the average energy score difference associated with these patterns.
AI
Based on your data, lunches containing processed grains (e.g., pasta, white bread) and refined sugars (e.g., soda, pastries) are strongly correlated with the most significant energy drops. These meals show an average energy score of 4.2 within 90 minutes post-lunch, representing a 35-40% decrease from your pre-lunch baseline. In contrast, lunches rich in lean protein and vegetables consistently result in an average post-lunch energy score of 7.8, with only a 10-15% perceived decrease.Use case implemented
The finished system, running on its own
With this specific insight, Sarah implemented a small, targeted change: lunches now consistently prioritize lean protein and a substantial portion of vegetables, minimizing processed grains. Her afternoon slump is noticeably reduced, and she feels more consistently alert, allowing her to tackle her late-afternoon tasks with renewed focus. The system now provides clear, actionable feedback.
7.5/10
Avg. Post-Lunch Energy
1-2/week
Days with 4pm Slump
+45 min
Avg. Focus Duration
What an outside observer would notice
From 5 to 7.5
Avg. Post-Lunch Energy Score
Increased 25%
Afternoon Productivity
Reduced 50%
Reliance on Afternoon Caffeine
The stack — build it yourself
Familiar, flexible for mixed data types, easy to export.
Excellent for natural language queries and identifying non-obvious correlations in textual data.
Integrates seamlessly with iPhone, convenient for quick daily inputs and viewing trends.
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.