New: our own Hairmetry line. Every ingredient explained

Topic

How to work out your hair type at home

A step-by-step guide to describing your own curl, thickness, density, damage and scalp, using questions that research suggests people can answer reliably.

hair lossthinningsheddingdrynessfrizzbreakagepoor curl definitionlow volumeoily scalpdry scalpitchy scalpdandruff

Simplified diagram of the Norwood scale. Some temple recession after puberty is almost universal, and even dermatologists agree only moderately on stage. Treat a stage as a rough band, not a diagnosis. Norwood 1975 ↗ Gan 2005 ↗ Guarrera 2009 ↗
Simplified diagram of part-width scales used for women. Women usually lose hair as diffuse thinning with a widening part and the front hairline kept, rather than male-type recession. Ludwig 1977 ↗ Ramos 2015 ↗
The four broad curl families. Types are bands, not boxes: in one study no person's sampled hairs fell into a single class, and raters agreed better on fewer groups. Mkentane 2017 ↗

Key takeaways

  • Describe your hair with history and observation, not physical tests. The best-supported questions ask what you can recall: chemical treatments, heat use, trim dates, how fast roots turn greasy, whether your part has widened. The float test for porosity has no validation, so skip it (see hair porosity).
  • Self-description can work when the question is well designed. Women's self-described "fine hair" matched instrument data Bouabbache 2016, men's recall of earlier balding patterns was consistent over time (test-retest kappa 0.71-0.81) Littman and White 2005, and trained observers' curl ratings carried the same genetic signal as machine measures Ho 2020.
  • Even trained raters disagree. The best-measured curl system had only fair-to-moderate agreement between trained raters (kappa 0.38-0.46), and one head can contain 2-3 curl classes Mkentane 2017. Note your main type plus a range.
  • Check the red flags first: patchy loss, scarring signs, sudden heavy shedding, pain or intense itch. These matter more than any hair type and point to a dermatologist (see hair loss scales and the hair growth cycle).
  • Judge thickness and density against people with similar hair, not against an absolute scale (see hair by ancestry).
  • Photos can confirm some traits and mislead on others: curl family, length, gray and gross pattern loss show up; density, thickness, porosity and scalp state do not (see taking a good hair photo).
  • Your goals matter as much as your traits. In a UK mixed-methods study, hair goals (look, feel, practical, emotional) predicted hair esteem better than length or curl type Daniels and Heitmayer 2024.
  • Almost none of these questions have been tested against measurement, so treat the result as a good estimate, not a lab result.

Ground rules for every step

  1. For curl, volume and frizz, look at hair on a normal wash day, product-free and air-dried.
  2. For treatment history, think back 12-24 months. Hair grows a few centimeters a month, and ends older than 2-3 years carry cumulative wear Thibaut 2010.
  3. If you are not sure, leave the item blank rather than picking the middle.
  4. Think in bands (for example "fine to medium") and notice how confident you are. Confidence is higher when your answers agree with each other and with a photo.
  5. Redo it after puberty, pregnancy, menopause, new medicines, big cuts or chemical treatments.

Step 0. Red flags

Question If yes
Bald patches, smooth shiny scalp areas, a scarred-looking hairline, or lost eyebrows? See a dermatologist before changing your routine (hair loss scales)
Sudden heavy shedding in the last 3 months? Read about telogen effluvium; get a medical check if it lasts beyond 6 months (hair growth cycle)
Redness, burning, pain, pustules or intense itch on the scalp? See a dermatologist or read the condition pages
Tight styles (braids, extensions, ponytails) with soreness or a receding edge? Read about traction (traction and hairstyles)

Step 1. Hair loss and thinning

  • Ask: compared with two years ago, is my part wider, my ponytail thinner, or my hairline or crown more visible? (No change, slightly, clearly.) Change over time is the key signal.
  • Men can compare against Norwood-style drawings; women against part-width pictures such as the Sinclair scale. Raters disagree on fine stages, so choose a band (none or mature, early, moderate, advanced) rather than an exact stage. See hair loss scales.
  • Family history is supportive context only.

This step matters most: a positive answer changes everything else.

Step 2. Curl pattern

  • Look at washed, air-dried hair hanging free and choose straight, wavy, curly or coily. Add a subtype A-C if you like, but accept that subtypes are the least reliable part; observers agree at family level more than subtype Mkentane 2017, Callender 2026.
  • Check whether the pattern is the same all over: front, crown and nape often differ.

Curl matters a lot for detangling, shampoo and conditioner choice, and styling. See curl pattern and hair typing systems.

Step 3. Strand thickness

Ask whether your individual strands are finer, about average or coarser than other people of your background. See strand thickness. Thickness is a moderate factor for product weight and heat tolerance.

Step 4. Density

  • A ponytail test: roughly under 5 cm, 5-10 cm or over 10 cm circumference. This mixes density with thickness, so use it as one input.
  • Part your dry hair: can you barely, somewhat or easily see the scalp? Judging curly density from photos is unreliable (see hair density).
  • Compare yourself with people of similar background: sparser, average or fuller.

Strand thickness and density are separate. Plenty of people have thick strands and low density, or the reverse.

Step 5. Porosity and damage

Answer from history rather than a test: bleach or lightening, relaxers, perms, keratin straightening, how often you use hot tools, color fading, tangling when wet, length and when you last trimmed. Treatment history counts most. The result is low, normal or high, or mixed if roots and ends differ. See hair porosity.

Step 6. Breakage (optional)

Ask whether you see many short snapped pieces, or hairs snapping when gently stretched wet. Treat this as a damage clue only (see hair elasticity).

Step 7. Scalp type

Count how long after washing your roots look greasy: within 1 day suggests oily, 2-3 days balanced, rarely greasy even after 4 or more days dry. Treat any flaking or itch as its own category whatever your oiliness. This mapping is a sensible rule of thumb; no study has validated it. See sebum and scalp types.

Step 8. Goals

Pick up to three: fuller, longer, shinier, less frizz, more curl definition, less breakage, less oily, fewer flakes, cover gray, thicker hairline.

Which answer changes what

Advice area Most affected by Less affected by
Medical questions and treatments Hair loss answers, red flags Curl, thickness
Shampoo and cleansing Scalp type, flakes or itch, curl, wash frequency Thickness
Conditioning and repair Damage history, curl, thickness Density
Heat and styling Damage history, thickness, curl Scalp type
Volume and cuts Density, thickness, curl, hairline Porosity
Tools and detangling Curl, length, breakage Scalp type

What photos can add

Standard photos can show curl family, length, gray, hairline shape and gross crown or part thinning. They cannot show density, thickness, porosity or scalp condition. Use a photo to cross-check your answers, and trust your answers where they conflict. See taking a good hair photo.

Myths and facts

  • "The float test tells you your porosity." No validation exists; surface residue and trapped air confound it.
  • "Your curl type is one letter." One head can span 2-3 classes Mkentane 2017.
  • "Fine hair means low density." Thickness and density are separate.
  • "An oily scalp means hair loss." No; see sebum and scalp types.

Bottom line

Start with the red flags, then describe your hair in bands: change in thinning, curl range, damage history, scalp rhythm and goals. Accept some uncertainty, because the research behind these questions is thin. Self-description is a starting point for choosing a routine, not a diagnosis.

Studies 7

Bouabbache S, et al. What is a Caucasian 'fine' hair? Int J Cosmet Sci. 2016

Design
other
People
151

Result: Self-described fine hair in women tracked instrument measures of thin strands, low density, low volume and weakness.

Read the study ↗

Mkentane K et al. Geometric classification of scalp hair for valid drug testing, 6 more reliable than 8 hair curl groups. PLoS One. 2017

Design
other
People
48

Result: Trained raters agreed only modestly on 8 curl groups (kappa 0.38-0.46) and better on 6 groups (0.61-0.73); no person's 10 sampled hairs fell into one class.

Read the study ↗

Littman AJ, White E. Reliability and validity of self-reported male balding patterns for use in epidemiologic studies. Ann Epidemiol. 2005

Design
other
People
100

Result: Men's self-reported balding patterns were consistent on retesting (kappa 0.71-0.81) but agreed only moderately with an interviewer (kappa 0.47).

Read the study ↗

Ho YYW et al. Comparison of genome-wide association scans for quantitative and observational measures of human hair curvature. Twin Res Hum Genet. 2020

Design
other

Result: Trained observers' curl ratings and machine-measured curvature pointed to the same genetic signal.

Read the study ↗

Daniels G, Heitmayer M. Towards a taxonomy for assessing and classifying the needs of curly hair. Int J Cosmet Sci. 2024

Design
other
People
212

Result: Hair goals fell into four kinds (look, feel, practical, emotional) and predicted hair esteem better than length or curl type.

Read the study ↗

Callender V et al. Classification of high curl pattern hair: a systematic review and clinical perspective. J Cosmet Dermatol. 2026

Design
systematic review

Result: Four curl classification systems exist; the Walker chart was built to market products and none has strong validation.

Read the study ↗
Show 1 more studies

Thibaut S et al. Chronological ageing of human hair keratin fibres. Int J Cosmet Sci. 2010

Design
other

Result: In hair longer than 2.4 m the cuticle wore away over roughly the first metre; further out, scale damage lined up with falling ceramide and 18-MEA levels and mechanical properties decayed.

Read the study ↗