Americans run about 754,000 searches a month across 85 facial feature and insecurity phrasings. The eye region draws the most search interest, and the fastest growing terms are subculture metrics like long philtrum and negative canthal tilt.
What America Searches About Its Own Face
The feature Americans search about most is the eye region, at 144,400 monthly searches across ten tracked terms. The nose is second at 118,630 across twelve. Every volume below is a 12-month US average from Google Ads data, and phrasings Google tracks as one query group are counted once.
| Feature area | Monthly searches | Terms tracked | Biggest term |
|---|---|---|---|
| Eyes and eyelids | 144,400 | 10 | hooded eyes (74,000) |
| Nose | 118,630 | 12 | hooked nose (49,500) |
| Midface and subculture metrics | 97,600 | 4 | philtrum (60,500) |
| Under-eye area | 81,000 | 2 | eye bags (40,500) |
| Chin | 66,080 | 9 | double chin (33,100) |
| Forehead | 57,770 | 6 | big forehead (49,500) |
| Teeth | 51,000 | 3 | crooked teeth (18,100) |
| Direct self-questions | 36,400 | 5 | am i pretty (12,100) |
| Symmetry | 27,590 | 7 | asymmetrical face (12,100) |
| Skin texture and tone | 14,700 | 4 | textured skin (5,400) |
| Ears | 14,520 | 4 | big ears (12,100) |
| Cheeks and cheekbones | 13,760 | 5 | chubby cheeks (6,600) |
| Lips and mouth | 10,830 | 5 | thin lips (6,600) |
| Jawline | 10,080 | 6 | no jawline (2,900) |
| Face shape and width | 9,720 | 3 | flat face (8,100) |
One reading discipline matters here. A search for "hooded eyes" can be worry, but it can also be a makeup tutorial or plain curiosity, and volume data cannot separate the three. That is why this index reports feature interest and keeps a stricter second layer, below, for phrasings that cannot be anything except a person asking about their own face.
The Questions People Ask Directly
Thirteen tracked phrasings are unambiguous: questions and "too big" or "fix" constructions where the searcher is the subject. Together they run at 43,520 searches a month, and this is the layer that grew. The direct self-question group rose 41 percent between August 2025 and July 2026 while most feature groups held flat or fell.
| Phrasing | Monthly searches | Aug 2025 | Jul 2026 |
|---|---|---|---|
| am i pretty | 12,100 | 12,100 | 18,100 |
| am i ugly | 9,900 | 8,100 | 14,800 |
| am i attractive | 6,600 | 6,600 | 8,100 |
| why am i so ugly | 5,400 | 5,400 | 5,400 |
| how to fix asymmetrical face | 3,600 | 3,600 | 3,600 |
| why am i ugly | 2,400 | 2,400 | 2,400 |
| one eye smaller than the other | 1,900 | 2,400 | 1,600 |
| why is my face so asymmetrical | 720 | 590 | 720 |
| is my face symmetrical | 590 | 480 | 1,300 |
| is my nose too big | 170 | 140 | 260 |
| nose too big | 70 | 50 | 90 |
| forehead too big | 50 | 50 | 40 |
| face too long | 20 | 40 | 20 |
These numbers are people, mostly typing alone. The register this site holds for every one of those searches is the same: a score is geometry, not worth. A face can be measured, and the measurement can be interesting, and neither fact says anything about what a person deserves. The support resources above exist because the data in this table keeps growing.
The Fastest Growing Facial Insecurities
The sharpest growth is not in classic worries. It is in measurement vocabulary that spread from looksmaxxing communities into the mainstream. "Long philtrum" grew 236 percent in twelve months, from 4,400 to 14,800 monthly searches, with a spike to 27,100 in May 2026. "Negative canthal tilt" grew 23 percent to 22,200 and peaked at 40,500. Terms like these are defined one by one in the glossary of facial measurement terms, and the community they came from is covered in What Is Looksmaxxing? Terms, Claims, and Evidence.
| Term | Aug 2025 | Jul 2026 | Change |
|---|---|---|---|
| long philtrum | 4,400 | 14,800 | +236% |
| recessed chin | 8,100 | 14,800 | +83% |
| am i ugly | 8,100 | 14,800 | +83% |
| am i pretty | 12,100 | 18,100 | +50% |
| midface ratio | 590 | 880 | +49% |
| negative canthal tilt | 18,100 | 22,200 | +23% |
| chubby cheeks | 6,600 | 8,100 | +23% |
The mirror image is just as clear. The forehead group fell 42 percent across the same window, the jawline group fell 31 percent, and the ears group fell 34 percent. Attention is migrating from single features toward verdict questions and subculture metrics, which judge the whole face against a framework rather than one part against a mirror.
How We Measured
- Source: Google Ads search volume via the DataForSEO API, United States, English, pulled on 2 September 2026.
- Volumes are 12-month averages. Trend columns compare the August 2025 and July 2026 monthly values from the same pull.
- Phrasings Google tracks as one query group (identical volume and identical monthly history, such as "asymmetrical face" and "facial asymmetry") stay in the dataset but count once in every total.
- One seeded term, "acne scars", returned no volume data from the source and is excluded rather than estimated.
- Google Ads reports volumes in rounded buckets, so every figure is an estimate of scale, not a count.
What This Data Cannot Say
Search volume measures attention, not distress. A descriptor search can be curiosity, a tutorial, or a worry, and only the direct-question layer separates them cleanly. The data covers one country and one year, the single-month trend comparison is not seasonally adjusted, and none of it diagnoses anything about any searcher. What it does show, on the record, is where attention sits and where it is moving. If you want to know what your own face actually measures, the free on-device face analysis returns numbers instead of a verdict, and the formula behind it is public on the methodology page.
Download the Dataset
The full dataset is 87 rows and 9 columns: feature group, keyword, 12-month average volume, the two monthly trend points, variant flags, distress-phrasing flags, source and pull date. Download it at /data/facial-insecurity-index.csv and check every number on this page against it. Our earlier study of the tools people land on after these searches is We Audited 25 AI Face Raters. Here's What They Actually Measure.