How do I quantify shedding well enough to know whether it is getting better or worse?

Accepted answer

Yes, and the two useful methods are a standardised shed count and standardised serial photography. Both are free and both are far better than perception, which is unreliable for a specific reason worth understanding: you cannot perceive density, you perceive contrast and parting width, and both of those change with lighting, humidity, hair length and how recently you washed it.

Method 1: the standardised shed count

Total daily shedding in a normal adult is commonly cited as 50-100 hairs, though the range is wide and depends on hair length, wash frequency and grooming. The absolute number matters less than the trend in a consistently measured count.

How to make it mean something:

• Fix the wash interval. Hairs shed between washes accumulate and come out during washing, so a two-day interval and a four-day interval produce very different counts. Pick one interval and never vary it during the measurement period. This is the single largest source of spurious variation.
• Count everything from one wash cycle: hairs collected during washing (a plughole strainer), during towelling, during combing, and off the pillow. A shower-only count misses roughly half.
• Count weekly, not daily. Day-to-day variation is large and meaningless. A weekly count on the same day of the week, at the same point in your dosing cycle, gives you a usable series.
• Record the number, not an impression of the number. Counting a hundred wet hairs is tedious; laying them on a dark towel in rows of ten makes it about two minutes.
• Expect a decline of the order of tens of percent over two to three months if the episode is resolving. Effluvium resolution is gradual, not abrupt.

The refinement worth adding: note the proportion of short hairs. Regrowth appears as fine, short, tapered hairs. Once your shed sample starts containing hairs of markedly varying length, you are seeing turnover of a recovering cohort rather than a continuing insult. That is the earliest positive sign available and it precedes any visible density change by months.

Method 2: standardised photography

This is the one that gives you the answer you actually care about, which is density rather than shedding. Almost everyone does it badly, and badly done photographs are worse than none because they generate false trends.

The variables that must be held constant:

• Lighting. Same room, same time of day, same lights, no window light, because daylight varies. Overhead light exaggerates scalp show; side light hides it. Pick one and keep it.
• Hair state. Dry, unstyled, no product, same number of days since washing. Wet hair always looks thinner and this catches people out badly.
• Parting. Same parting, made the same way, in the same place. A parting made 1 cm to the side of the previous one changes apparent width substantially.
• Camera position. Same distance, same angle, same phone. Mark a spot on the floor if you need to.
• Four views: top of head, parting from directly above, each temple, and the crown. Change is often visible in one view and not others.
• Monthly, not weekly. Hair grows about 1 cm per month; weekly photographs show only noise.

A useful trick for the parting view: place a ruler or a coin flat against the scalp alongside the parting. It gives you a scale reference, which turns an impression of "wider" into a measurement.

What a clinician adds

• The pull test. Grasping roughly 50-60 hairs and applying gentle traction; more than about 10% coming away suggests active effluvium. Crude, quick, and it distinguishes active shedding from a stable state, which is the single most useful binary here. It is affected by recent washing, so it is done a day or more after a wash.
• Trichoscopy. A dermatoscope shows hair calibre variability, which is the discriminating feature for androgenetic alopecia, plus follicular openings, perifollicular inflammation and scarring. This is the examination that changes the diagnosis rather than the monitoring, and it is the reason to see someone once even if you then monitor at home.
• A trichogram or biopsy, rarely needed, occasionally decisive, particularly where a scarring alopecia is suspected — those are permanent and time-sensitive, which is the one scenario where delay genuinely costs you.

How to use the numbers for your actual decision

Your decision is whether to slow your weight loss further. Two things make that decision tractable:

1. Remember the lag. Anything you change today affects shedding two to four months from now. So the shed count over the next eight weeks tells you about the insult two to four months ago, not about the change you are making. This is the single most common reason people conclude an intervention did not work, and it makes a rapid feedback loop impossible. Set expectations accordingly or you will churn through interventions.
2. Because of that lag, decide on the input rather than the output. Rate of loss as a percentage of body weight per week and protein in grams per kilogram are both measurable today, both known to be causally upstream, and both give you an immediate read. Manage those, and use the shed count and photographs to confirm months later that you were right.

Managing a delayed-feedback system by watching the output is how people end up oscillating. Manage the input, verify the output on its own timescale.

edited 11 Jun 2026 by per_haugen — clarified the distinction between purity and content