Methodology
The Kirby-Desai Scale
The session-count estimate is based on the Kirby-Desai Scale, published by dermatologist William Kirby and colleagues in the Journal of Clinical and Aesthetic Dermatology in 2009. It assigns point values for six factors (skin type, body location, ink amount, ink colors, scarring, and layering) and produces a session-count range rather than a point estimate. The scale was originally validated with r = 0.757 on a cohort of 100 patients. The scale has known limits, including a single 2009 cohort, limited skin-type diversity, and a pre-picosecond device baseline, which is why our calculator outputs a range and links to the methodology article for the full evidence.
Correction for picosecond lasers
The original Kirby-Desai study used Q-switched nanosecond lasers. Since 2013, picosecond systems (PicoSure, PicoWay, PicoPlus, and others) have become common, and validation studies have consistently found the original scale over-predicts session counts against modern devices. Our calculator applies a picosecond correction framed as a range rather than a single multiplier, because no published study supports a universal factor. The range is built from two editorial multipliers applied to the Kirby-Desai score:
- Lower bound: score × 0.60 (a 40% reduction, reflecting best-case picosecond performance on compatible black and dark-blue inks).
- Upper bound: score × 0.95 (a 5% reduction, reflecting cases where picosecond barely improves on the original scale, e.g. Fitzpatrick V-VI skin, white or yellow ink, or non-standard inks not characterized in 2024-25 work).
These are editorial bounds chosen to reflect the direction and rough magnitude of the 2024-25 literature (Egozi 2024, Menozzi-Smarrito 2025), not factors derived from a regression on primary data; the per-patient prediction error of Kirby-Desai itself in the cleanest modern head-to-head is ±3.7 sessions, which is the honest precision floor for any single case. See our session-count calculator methodology article for the full evidence, the studies behind each bound, and where the correction is most and least supported.
How clinic laser types are classified
The directory labels each clinic as Picosecond, Q-switched, or Unknown. Picosecond means the clinic or its parent chain advertises picosecond laser use in its public materials; this is not verified for individual locations, and a chain location may use a backup Q-switched device on any given day. Q-switched means only nanosecond devices were recorded. Unknown means we have no laser inventory data on file. None of these labels imply a recommendation; they are reproductions of what the clinic or chain publishes. The label reflects what the clinic publishes about its equipment, not operator skill, settings selection, or treatment outcome, all of which matter more than device class on most tattoos.
Where our pricing data comes from
- Clinic public pricing pages. We pull published pricing from clinic websites where the clinic has made it public, and record the date each page was read.
- National fallback. For city-and-size combinations where fewer than three distinct chain corporate parents have published prices, we fall back to a national distribution drawn from the same chain-deduped sample, label the row as a national fallback, and note the absence of sufficient in-city data. We do not pool from nearest cities or otherwise impute from neighboring metros.
Every price we display is labeled with its source and recorded date. We do not present national fallback ranges as if they were live local quotes. All prices are reported, not quoted; always confirm current pricing with the clinic directly before scheduling treatment.
What we don’t include
- Before-and-after photos of individual patients.
- Claims of guaranteed outcomes.
- Paid placements presented as editorial.
Where this methodology lives in practice
The methodology drives the session-count and cost calculator, the cost-by-city ranges, and every clinic listing in the directory. The full bibliography behind the figures and citations on this page is on the sources page; the longer-form methodology article is in the guide.