There is a clinical tendency, particularly in environments that feel the need to justify their cost through volume, to measure everything. A full battery of indicators, printed and laminated, handed to the client at the end of a session as proof that the hour was spent productively. We have seen sheets with twenty-two metrics. We have seen software that tracks thirty-seven. We have also seen the faces of people given those sheets, and what they register is not clarity.
The problem with over-collection
When you measure too many things, you introduce noise. Some indicators move in concert, hydration and barrier function, for example, are so closely coupled that tracking both as independent metrics produces redundant data without improving your ability to act. Others move in directions that contradict each other within the same session, creating the appearance of ambiguity where there is in fact a clear clinical narrative. The temptation to add metrics is always justified at the collection stage. It is never justified at the decision stage.
A metric without a corresponding action is not a measurement. It is decoration.
Over three years of single-studio practice, we tracked thirty-one indicators across our full cohort. Not all at once, most members encountered ten to fourteen, depending on their presenting concerns. What we were looking for was not a universal protocol. We were looking for the minimum viable signal: the smallest set of measurements that reliably predicts treatment response and allows us to adjust cadence meaningfully between sessions.
What stayed, and the logic behind each
Hydration, measured via corneometer reading at three skin zones, survived because it is the single fastest-responding indicator. A well-hydrated barrier responds to treatment differently than a dehydrated one, and it changes measurably within fourteen days of protocol adjustment. It is not a long-term outcome metric; it is a proximate indicator of how the skin is responding right now.
Elasticity, measured via cutometer at the lateral cheek and forehead, is the long-form metric. It changes slowly, over twelve to twenty-four weeks. It is the number that tells you whether the foundational work of the treatment is accumulating. On its own it is too slow to guide session-by-session decisions, but paired with a longer cohort window it is the most honest measure of structural skin improvement we have.
Luminosity and pore clarity are assessed visually under standardised lighting rather than via instrument. This was a deliberate choice. Instrument-measured diffuse reflectance and follicle density tools exist, and we trialled them. What we found is that visual assessment under consistent conditions, same room temperature, same lamp colour temperature, same distance, produces inter-session reliability equivalent to automated tools for these two indicators, at a fraction of the session time. The gain from instrumentation was smaller than the loss from the two additional minutes it required.
Tone evenness, the fifth indicator, was the last to be confirmed. We cycled through melanin index measurement, sebum distribution, erythema scoring. What we settled on was a composite: a clinician-scored tone evenness scale that integrates post-inflammatory hyperpigmentation, diffuse redness, and surface texture into a single 0–100 value. It is the most subjective of the five. It also correlates most strongly, across our cohort, with member satisfaction at the quarterly review. Which is, ultimately, what a clinic is for.
What was cut
pH, sebum rate, transepidermal water loss, melanin index as a standalone, surface roughness via profilometry, skin temperature differential, all were tracked for at least twelve months. All were cut. The reason in each case was the same: the measurement either duplicated a signal already captured by one of the five, or it produced actionable information only in clinical edge cases that do not represent the membership population. TEWL, for example, is a useful diagnostic tool. As a recurring session metric, it tells you what you already know when you look at the barrier indicators.
Five is not a round number chosen for elegance. It is the number that remained after everything without a clear decision tree was removed.
We publish this not as a prescription for other practices, but as an account of how we arrived at the protocol we use. The number may change. If a new indicator is validated that outperforms one of the five, it replaces rather than joins. The principle is not five, the principle is minimum viable signal. Everything else is noise that would eventually undermine the data layer we are building for every member over the course of their residency.