The science
Every number in your Revelation is grounded in established exercise science and real research. Here's what gets measured, why it matters, and how it turns into the insight that changes what you do next.
What your wearable is actually measuring
Your wearable records several signals that, taken together, paint a detailed picture of how your body is doing. Most health apps show you these numbers one at a time โ separate screens, separate graphs, no connection between them. A Revelation reads them together. The real value is in how they interact.
Heart rate variability (HRV)
HRV measures the tiny variation in time between each heartbeat. It sounds like a small thing, but it's one of the most powerful markers of overall health and recovery that exists.
Higher HRV means your nervous system is in a balanced, recovered state โ ready to handle stress, training, and daily life. Lower HRV means your body is under load: fighting off illness, recovering from a hard session, dealing with poor sleep, or carrying built-up stress.
HRV is sensitive to almost everything โ training intensity, alcohol, sleep quality, emotional stress, illness. That sensitivity is what makes it so useful. It's your body's broadest recovery signal.
What matters most isn't a single reading, but the trend over time. A gradual rise in your average HRV over weeks means your body is adapting well. A sustained drop means something is off โ even if you feel fine.
Resting heart rate
Your resting heart rate is how fast your heart beats when you're completely at rest. A lower resting heart rate means your heart is stronger โ it can pump the same amount of blood with fewer beats.
Over time, as fitness improves, resting heart rate drops. A sudden spike โ even by just a few beats per minute โ can signal illness, dehydration, overtraining, or built-up stress.
The Signal tracks this daily and looks at the trend. A resting heart rate that's been slowly climbing for 2 weeks tells a different story than one that spiked for a single day after a bad night's sleep.
Sleep
Sleep is where your body does most of its repair and adaptation. Without enough sleep, HRV drops, resting heart rate rises, and the benefits of training shrink.
Duration matters, but it's not the whole picture. Consistency โ going to bed and waking up at roughly the same time every day โ has a measurable effect on recovery. Someone sleeping 7 hours at the same time every night will typically show better HRV than someone sleeping 8 hours at random times.
The analysis tracks bedtime, wake time, total duration, and consistency. When your sleep timing drifts โ say, bedtime moves from 22:30 to midnight over a few months โ your Revelation will show you exactly what that drift is costing you in recovery.
Training load and heart rate zones
Not all exercise is equal, and the difference often comes down to intensity โ specifically, which heart rate zone you're training in.
Zone 2 (roughly 60โ70% of your maximum heart rate) is where your body burns fat for fuel and builds endurance. It feels easy โ conversational pace. The science recommends spending about 80% of your training time here.
Zone 4 and above (80โ90%+ of max heart rate) is high-intensity work โ intervals, sprints, hard efforts. Valuable in small doses, but it's taxing on your nervous system and needs serious recovery time.
The problem most people have โ and this is one of the most common Revelations in the data โ is spending too much time in Zone 3. Zone 3 is the dead zone: too hard to recover from easily, too easy to drive real improvement. It feels like a solid workout, but it's often the least productive zone to train in.
Every workout gets broken down by heart rate zone, so you can see exactly where your time is going. Often, the gap between what you think you're doing and what the data shows is the entire Revelation.
Weight direction
A single weigh-in tells you almost nothing. Weight fluctuates by a kilogram or more day to day based on hydration, food timing, and other factors.
What matters is the trend over weeks and months. The analysis tracks the direction โ is weight gradually moving the way you want, stuck, or going the wrong way? โ and looks at it alongside your training load, sleep patterns, and recovery.
A weight plateau despite heavy training and calorie restriction, for example, is a classic sign that your body is adapting to stress. It's holding on to everything it can because you're asking too much of it. That's exactly the kind of pattern that shows up in a Revelation.
One number. 5 inputs. Calibrated to you.
The Signal is a daily score from 0 to 100 that answers a simple question: is your body recovering or declining? It combines 5 metrics into a single number that anyone can understand without knowing what HRV stands for.
The single best marker of overall recovery. Sensitive to training, sleep, alcohol, and illness.
Duration, consistency, and timing. Not just hours โ regularity and how well your patterns hold.
How hard your heart works at rest. Tracks long-term fitness gains and flags stress when it spikes.
Intensity and volume relative to your capacity. Flags overtraining or poor recovery.
The direction all other components are moving over the past 1 to 2 weeks. Catches slow declines before they become problems.
Why these weights
Heart rate variability carries the highest starting weight because it's the most responsive recovery marker your wearable records. It reacts to training load, sleep quality, alcohol, stress, and illness โ often before you notice anything yourself. Sleep quality sits second because adaptation happens during sleep: suppress it and everything downstream suffers, regardless of how well you train. Resting heart rate and training load provide the structural picture โ long-term fitness trajectory and whether you're asking more of your body than it can currently absorb. Trend momentum is weighted lowest because it's a derivative: it watches the other four and catches slow slides before they become problems.
These aren't fixed. The engine adjusts the balance as it learns which signals are most predictive for you specifically. Someone whose HRV barely moves but whose sleep patterns drive everything will see that reflected in their weighting over time. The starting point comes from the research. The personalisation comes from your data.
Personalised, not generic
The Signal doesn't use population averages. It calibrates to your own baseline over your first 2 to 4 weeks of data. An HRV of 45 milliseconds might be excellent for a 55-year-old and concerning for a 25-year-old athlete. The Signal knows the difference because it's read your history.
It also looks at how metrics interact. If your HRV is fine but your resting heart rate is raised and your sleep timing has shifted later, those 3 signals together tell a different story than any one of them alone. The weighting adjusts based on which signals matter most for you, specifically.
Find. Rank. Frame.
The Revelation Engine is the system that turns your data into one powerful, specific Revelation. It works in 3 stages.
Find
Your full dataset gets analysed and a long list of candidate insights is generated. The analysis looks for patterns across 6 categories:
Effort versus outcome โ Training 5 times a week but fitness markers are flat or declining
Hidden correlations โ HRV drops 18% in the 48 hours after any workout with average heart rate above 165 bpm
Timing patterns โ Workouts after 7pm correlate with 35 minutes less sleep and lower next-day HRV
Inflection points โ Resting heart rate began climbing in September, coinciding with a shift from 3 to 5 sessions a week
Sleep architecture โ Bedtime has drifted from 22:30 to 23:45 over 18 months. HRV is 22% higher on earlier nights
Training zone misalignment โ A "Zone 2 ride" with only 30% of time actually in Zone 2 โ the rest is Zone 3 and Zone 4
Rank
Not all insights are equal. Each candidate is scored on 3 dimensions:
Surprise
How counterintuitive is this? An insight that confirms what you already know scores low. An insight that contradicts your self-image โ "your hardest workouts are your least effective" โ scores high. Surprise is what makes you stop scrolling.
Impact
How much would acting on this change your results? An insight about Tuesday versus Thursday timing is interesting but low-impact. An insight that your entire training approach is above your fat-burning threshold could change everything.
Actionability
Can you do something about this tomorrow? "Your genetics predispose you to slow recovery" is interesting but useless. "Moving your bedtime 45 minutes earlier would improve your HRV by an estimated 15%" is something you can act on tonight.
The engine multiplies these 3 scores and surfaces the highest-ranking insight as your Revelation.
Frame
The same insight can be framed as a statistic or as a story. The statistic is forgettable. The story changes behaviour. Every Revelation follows a consistent editorial structure:
The headline โ One sentence. Surprising. Personal. The thing you'd screenshot and send to a friend.
The evidence โ Specific numbers, specific timeframes, from your data. Not generic statistics.
The contrast โ What you're doing versus what your data says you should be doing.
The fix โ One clear, specific recommendation. The single biggest change you can make this week.
The promise โ What the data suggests will happen if you follow the fix, based on your own patterns.
Every number is checked
AI can hallucinate โ invent plausible-sounding numbers that aren't real. There's a validation layer built specifically to prevent this.
Before your Revelation is generated, every fact it might need to reference is pre-computed: your zone distributions, your personal bests, your recent averages, your weight direction. This becomes a facts sheet โ a single source of truth.
After the Revelation is written, every specific number it cited is automatically cross-checked against that facts sheet. If a number is wrong โ even slightly โ the output is rejected and regenerated. If it fails 3 times, the system falls back to a simpler, pre-computed insight that's guaranteed to be accurate.
A compelling headline with a wrong number is worse than a boring headline with a right number. Accuracy over flair, every time.
What wearables get wrong
No wrist-worn sensor is a medical device, and we don't pretend otherwise.
Wrist-based heart rate variability is noisier than a chest strap reading. Sleep staging from an accelerometer is approximate โ it infers sleep phases from movement, not brainwaves. Different wearables measure differently: an HRV reading from an Apple Watch and an OURA Ring won't match exactly, even on the same wrist on the same night.
The Revelation Engine is designed around these limitations, not despite them. Single-day readings are unreliable, so insights aren't built from single-day readings. The engine uses trend patterns across weeks and months โ that's where the signal lives and where the noise washes out. If your data has gaps, device switches, or inconsistencies between sensors, it gets flagged rather than papered over.
The claims in your Revelation are about the direction and pattern of your data over time, not the absolute precision of any individual reading. That distinction matters, and we'd rather you understood it upfront.
The research behind the engine
The principles that power the Revelation Engine aren't proprietary. They're published science, applied personally.
Heart rate variability and recovery monitoring
Plews, Laursen, Stanley, Kilding & Buchheit. Training adaptation and heart rate variability in elite endurance athletes: opening the door to effective monitoring. Sports Medicine, 43(9):773โ781, 2013.
Polarised training intensity distribution
Seiler. What is best practice for training intensity and duration distribution in endurance athletes? International Journal of Sports Physiology and Performance, 5(3):276โ291, 2010.
Sleep regularity and metabolic health
Huang & Redline. Cross-sectional and prospective associations of actigraphy-assessed sleep regularity with metabolic abnormalities: the Multi-Ethnic Study of Atherosclerosis. Diabetes Care, 42(8):1422โ1429, 2019.
Cortisol and chronic high-intensity training
Hackney. Stress and the neuroendocrine system: the role of exercise as a stressor and modifier of stress. Expert Review of Endocrinology & Metabolism, 1(6):783โ792, 2006.
Zone 2 training and fat oxidation
Achten & Jeukendrup. Optimizing fat oxidation through exercise and diet. Nutrition, 20(7โ8):716โ727, 2004.
Wearable accuracy and limitations
Bent, Goldstein, Kibbe & Dunn. Investigating sources of inaccuracy in wearable optical heart rate sensors. npj Digital Medicine, 3:18, 2020.
We didn't invent the science. We made it personal.
The data is already there
Your wearable has been collecting the signals. You just haven't been shown what they mean.
Get your Revelation โFree ยท Any wearable ยท Your data stays on your device