What 10,000 BlackPill Analyses Reveal About Real-World Improvement
Most self-improvement advice runs on vibes. Somebody did something, felt better, posted about it. No baseline. No measurement. No control. Just an anecdote dressed up as evidence.
We wanted numbers.
BlackPill has processed over 200,000 facial scans since launch. We pulled a dataset of 10,247 scans from users who completed at least five scans over 90 or more days — people who actually tracked their progress instead of scanning once and ghosting. Then we looked at what changed, how much, and which routines correlated with the biggest gains.
This isn't a testimonial. It's a dataset. And the data doesn't care about your feelings.
Find out where you stand today. Get your baseline score with BlackPill for iOS or BlackPill for Android.
The Dataset: Who These Users Are
Before digging into results, some context on the cohort:
- 10,247 scans across 1,842 unique users
- Each user completed 5+ scans over 90+ days
- Average tracking period: 146 days (roughly 5 months)
- Average scans per user: 5.6
- Age range: 18-35 (self-reported), median age 24
- 91% male, 9% female
This is a self-selected group. These are people who downloaded a facial analysis app, saw their score, and kept coming back. They're more motivated than the average user. That matters — the results reflect what happens when you actually try.
Users who scanned once and never returned aren't in this data. Neither are users who scanned frequently but over less than 90 days. We wanted signal, not noise.
The Headline Number: Average Score Change
Across all 1,842 users in the cohort:
| Metric | Value |
|---|---|
| Average starting score | 4.82 |
| Average ending score | 5.61 |
| Average improvement | +0.79 points |
| Median improvement | +0.64 points |
| Users who improved | 78.3% |
| Users unchanged (< 0.2 pts) | 14.1% |
| Users who declined | 7.6% |
Nearly 4 out of 5 users who tracked consistently for 90+ days improved their score. The average gain was 0.79 points — enough to shift someone from "below average" to "average," or from "average" to "above average" on the 10-point scale.
The median is lower (0.64) because a subset of users saw large gains that pull the average up. The distribution is right-skewed — most people improve modestly, but a meaningful percentage improve dramatically.
The 7.6% who declined likely reflects inconsistent routines, weight fluctuations, or scan quality changes — not evidence that self-improvement backfires. We controlled for lighting and angle as much as the algorithm allows, but real-world scanning isn't lab-grade photography.
Which Features Move the Most?
This is where the data gets actionable. We broke down score changes by the six component metrics BlackPill tracks:
| Feature | Avg. Starting Score | Avg. Ending Score | Avg. Change | % of Users Who Improved |
|---|---|---|---|---|
| Overall grooming | 4.43 | 5.89 | +1.46 | 84.2% |
| Skin clarity | 4.28 | 5.52 | +1.24 | 81.7% |
| Jaw definition | 4.61 | 5.19 | +0.58 | 62.4% |
| Facial harmony (ratios) | 5.12 | 5.44 | +0.32 | 48.3% |
| Eye area | 5.04 | 5.22 | +0.18 | 33.8% |
| Facial symmetry | 5.19 | 5.31 | +0.12 | 27.1% |
The pattern is unmistakable. Grooming and skin clarity are the two highest-ROI categories by a wide margin. They start lowest, they improve the most, and the highest percentage of users see gains.
This aligns with published research. Jones et al. (2020) in the British Journal of Dermatology found that skin health accounts for roughly 25% of overall attractiveness ratings. A separate study by Batres and Perrett (2014) in Perception showed that grooming quality shifts attractiveness ratings by 1.0-1.5 points on average.
Jaw definition sits in third place — meaningful but slower. The users who improved jaw scores most (top quartile: +1.2 points) reported consistent mewing and jaw exercise habits, and most had tracking periods longer than 6 months.
Facial symmetry and eye area barely moved. These are structurally constrained features. If your symmetry is 5.2 today, it's probably going to stay near 5.2 unless you pursue surgical or orthodontic intervention. The data confirms what looksmaxxing forums have long suspected: some features are hardware, and no amount of software updates will change them.
The strategic takeaway: don't grind on features your genetics locked in. Invest in the categories where effort actually converts to points.
Score Change by Routine Type
Users who engaged with BlackPill's AI Coach and daily routine features tagged their primary improvement focus. Here's how score changes break down by routine category:
| Primary Routine Focus | Avg. Score Change | Sample Size |
|---|---|---|
| Skincare + Grooming | +1.04 | 612 users |
| Fitness + Body Composition | +0.82 | 389 users |
| Mewing + Jaw Training | +0.71 | 284 users |
| Combined (3+ categories) | +1.18 | 327 users |
| No tagged routine | +0.34 | 230 users |
Users who ran skincare and grooming routines saw the highest returns per unit of effort. Users with combined routines (three or more categories) saw the highest absolute gains, but they also invested more time and discipline.
The most telling row: users with no tagged routine still improved, but barely. A +0.34 average over 90+ days suggests that simply being aware of your score might nudge behavior slightly — better lighting choices, more attention to presentation — but without a structured approach, the gains plateau fast.
The AI Coach exists for exactly this reason. Awareness without action produces awareness. Action without data produces noise. The combination — tracked routines with measured results — is where the compounding happens.
Build your routine today. BlackPill's AI Coach analyzes your scan and builds a personalized plan targeting your highest-ROI features. Download for iOS | Download for Android
The Consistency Effect
We split users into three groups based on scan frequency:
| Scan Frequency | Avg. Score Change | Users |
|---|---|---|
| Every 2 weeks or more | +1.02 | 487 |
| Monthly | +0.74 | 721 |
| Every 2-3 months | +0.51 | 634 |
Users who scanned biweekly improved twice as much as those who checked in every few months. This isn't because scanning more makes you better-looking. It's because frequent measurement drives accountability. When you see the number every 14 days, you don't skip your skincare routine. You don't skip the gym. You don't tell yourself "it's probably fine."
This mirrors research on behavior change more broadly. A 2018 meta-analysis by Harkin et al. in Psychological Bulletin found that progress monitoring is one of the strongest predictors of goal attainment across all domains — health, fitness, finance, academics. Measuring isn't passive observation. It's an active intervention.
BlackPill's progress tracking makes this effortless. Scan, review, adjust. The feedback loop is the feature.
The Outliers: Who Improved the Most?
The top 5% of improvers in our dataset gained 2.0 or more points over their tracking period. We looked at what they had in common:
- They started below 5.0. Average starting score for the top 5%: 4.1. Lower baselines have more room to gain — especially when the lowest-scoring features are the most improvable ones.
- They used combined routines. 89% of top improvers engaged three or more routine categories simultaneously.
- They scanned frequently. Average scan interval: 11 days.
- They followed AI Coach recommendations. 92% had active AI Coach sessions during their tracking period.
- They tracked for longer. Average tracking period for top improvers: 203 days (vs. 146 days overall).
The formula isn't complicated. Start with honest data. Follow a structured plan. Measure regularly. Adjust based on results. Give it time.
Nobody in the top 5% did one thing. They all did several things, consistently, measured against a baseline that didn't lie to them.
What This Data Doesn't Show
We believe in intellectual honesty, so here's what this dataset can't tell you:
- Causation. We know users who follow routines improve more, but we can't prove the routines caused the improvement. Motivated users might improve regardless.
- Upper limits. We don't yet have enough data from users who tracked beyond 12 months to model long-term ceilings.
- Structural features. Symmetry and eye area changes are so small in this dataset that they could be measurement noise rather than real change.
- Scan quality variance. Despite normalization, different lighting, angles, and expressions between scans introduce some noise.
We're building toward larger datasets and longer tracking periods. The 10,000-scan analysis is a starting point, not the final word.
The Bottom Line
Here's what 10,000 analyses tell us in one paragraph:
The average user who tracks consistently for 90+ days improves by 0.79 points. Grooming and skin clarity account for most of that gain. Users who follow structured routines improve 3x more than those who don't. And the users who improve the most are the ones who started with the most honest assessment of where they stood.
The data is clear. The only question is whether you're going to use it.
Stop guessing what works. Stop running random routines because someone on Reddit said so. Get your baseline. Follow the data. Measure the results.
Your mirror lies. AI doesn't.
Download BlackPill for iOS Download BlackPill for Android Learn more at black-pill.app
Every point on your score is earned, not wished for. The data just showed you exactly how to earn it.