How the platform counts

A client becomes regular on the third visit — here’s how we count it.

The platform doesn’t guess or label clients by hand. It works out a client’s stage from facts: visits, bookings and their usual rhythm. Below are the simple rules behind it and the research they rest on.

45% → 70% → 79%of clients come back for the next visit after the 1st, 2nd and 3rd — based on 11+ million salon appointments
1 · 2.5 · 4usual intervals — the thresholds for “time to come back”, “sleeping” and “gone”
12 monthsbefore this, we never count a client as gone

The label next to each rule is its status in the work plan: “Live” means the platform already counts it this way. LiveIn progressPlanned

Two axes

Who the client is and what’s happening now are two different questions

Many systems use a single ladder: “new → regular → gone”. But a regular client who disappears doesn’t stop being a regular — only their rhythm has slipped. That’s why the platform, like the major CRMs worldwide, keeps the two axes separate.

In progress
Experience

Experience: how many times the person has been

Only moves forward. A client who comes back after a break stays regular — their history isn’t erased.

  1. Inquirywrote, called or left a request, no booking yet
  2. Bookedhas a booking, no visit yet
  3. Newone visit
  4. Repeattwo visits
  5. Regularthree visits or more
Rhythm

Rhythm: how long ago they came — in usual intervals

Recalculated every day and moves both ways: the client comes back — they’re in rhythm again.

  1. In rhythmless than 1 interval, or has a future booking
  2. Time to come back1 to 2.5 intervals
  3. Sleeping2.5 to 4 intervals
  4. Goneover 4 intervals and no sooner than 12 months

Try it yourself

demo
5
40
StageRegularTime to come back

40 days have passed — that’s 1.4 of the usual interval (28 days).

What the platform doesSends a reminder itself; where a conversation is needed, the administrator gets a task.
A stage comes as a pair: “Regular · time to come back”. This client goes to the top of the administrator’s list: they already trust the salon, and it’s worth inviting them back at the right time.
This is how the major CRMs work: in HubSpot, the lifecycle stage moves forward automatically only, while everything about “right now” lives in separate fields. [5]
Why the third visit

Most people are lost between the first and second visit

Boulevard analysed over 11 million appointments across 2,500+ salons. After the first visit, 45% of clients return on average; of those who come a second time, 70% book a third; after the third, 79% come for a fourth. After that, the curve levels off — which is where the “regular” threshold comes from. [1]

Share of clients who come for the next visit
All clients, on averageTop salons
1st → 2nd45%70%
2nd → 3rd70%81%
3rd → 4th79%85%
Between the 1st and 2ndmore than half of new clients don’t come back. This is where attention matters most: book the next visit during the first one.
Between the 2nd and 3rdalmost one in three is still lost. That’s why “repeat” is a step of its own: two visits aren’t a habit yet.
After the 3rdlosses drop to one in five and barely change after that.

The threshold is the same across the whole platform — so numbers from different salons can be compared.

Rhythm

“A long time” means something different for every service

A manicure is done every few weeks, while an adult’s dental check-up is scheduled individually — anywhere from 3 to 24 months. A single number of days for everyone would mark people as “gone” when nothing is wrong. That’s why the platform measures the gap in usual intervals. [4]

Where the usual interval comes from — in order, the first one available

  1. 01

    Recommended date

    The specialist’s or doctor’s word: “come back in four weeks”. The most accurate source — we take it first.

    Live
  2. 02

    Direction median

    The typical gap between visits in this direction over the last 12 months: first in the branch; if there’s too little data — in the city, then in the country. Recalculated every night; the card shows where the number came from.

    Planned
  3. 03

    Default value

    The interval the salon set for the direction — until there’s enough data of its own.

    Live

The platform doesn’t use a client’s personal “average gap”: life often throws schedules off, and the specialist’s recommendation is more accurate.

Rhythm thresholds

In rhythm
Time to come back
Sleeping
Gone
01×2,5×4×

“Gone” — no sooner than 12 months after the last visit, even if the interval is short.

Has a future booking — the client is always in rhythm. You can’t write “we haven’t seen you in a while” to someone who’s coming on Thursday.

Why a simple rule instead of a complex model

A peer-reviewed study in the Journal of Marketing showed that the rule “hasn’t bought in a long time — inactive” identifies active clients just as well as complex probabilistic models. For regular services, tracking rhythm makes the forecast even more precise. And most importantly, the rule can be explained: the administrator sees why it’s time for the client and can say so in a human way. [2] [3]

Neither science nor the industry has universal multipliers, so 1 · 2.5 · 4 intervals and 12 months are defaults: you can change them in your account settings.

Network and directions

Stopped coming for laser — but hasn’t left the salon

The platform counts the stage twice: across the whole network and separately for each direction. But the words about the person — “gone”, “sleeping”, “regular” — are decided by the network alone. A direction only gives a reason to remind them about a service.

In progress

Client card

demo
ManicureRegularIn rhythm
LaserRepeatSleeping
Whole networkRegularIn rhythm

Partial defection: the person still comes, but has dropped one service.

The platform will send “time for laser” — and never “we haven’t seen you in a while”. People who use several directions are usually the most loyal clients.

  • Each service belongs to exactly one direction.
  • Any future booking — no direction reminder is sent.
  • A visit at any branch in the network is the same client, not a new one.

A study of a retail chain showed that when loyal customers partially defect — stop making some of their purchases — the best predictors are recency, frequency and amount spent. Simple facts, in other words, not guesswork. [6]

Booking events

What happened to a booking — in plain words

Salons fill in Altegio in different ways: some delete the booking, others mark it “no-show”. The platform looks not at the current status but at the change log: what changed and when — before the visit started or after. So the same events are counted the same way in any salon.

before the startLive

Cancellation

The booking was deleted or marked “no-show” before the visit started — in any way: by the client in the bot, by the administrator or by an integration. Who cancelled is recorded in the event.

Why it matters: The administrator has time to save the visit or fill the slot.

less than N hours beforeLive

Late cancellation

The same cancellation, but late: less than N hours before the visit (48 by default, changeable in settings).

Why it matters: The owner sees the specialist’s lost hours, the administrator sees which slots to fill urgently.

after the startLive

No-show

“No-show” or deletion after the visit has started, if it wasn’t marked “arrived” before.

Why it matters: Shows how many visits are lost without warning.

after the visitPlanned

Not marked

The visit time has passed, but the status is still “pending” or “confirmed”. It’s neither a visit nor a no-show, but a gap in the records.

Why it matters: We don’t move the stage or guess; the owner sees how many bookings aren’t marked and can put things in order.

same bookingLive

Reschedule

The day was changed in the same booking: one booking, new date.

Why it matters: This isn’t a lost visit.

cancellation + new bookingLive

Saved visit

The booking was cancelled, but within N days (7 by default) the client booked again in the same direction.

Why it matters: An administrator who persuaded the client to reschedule rather than cancel sees it in their numbers.

new bookingIn progress

Booked

The person has booked, but the visit is still ahead. We wait to see whether they come.

Why it matters: We don’t count a return on a promise; a booking after a reminder is a separate metric.

visit after a breakIn progress

Returned

A visit took place after “time to come back”, “sleeping” or “gone”. Rhythm is back to “in rhythm”, experience isn’t reset.

Why it matters: We measure a return by the visit — that shows whether reminders and tasks worked.

booking aheadLive

Future booking

A booking for a time later than now — not deleted, not marked “no-show”, at any branch in the network.

Why it matters: The platform doesn’t send comeback reminders to someone who’s already booked.

Reschedule or saved visit?

RescheduleOne booking — only the date changed.
Saved visitThe booking was cancelled, and a few days later a new one appeared. These are two events; the credit goes to whoever created the new booking.

And a visit is client × day × branch where at least one booking is marked “arrived”. Three services in one trip — one visit.

Markets and currency

Different countries, different markets

Prices, habits and visit rhythm differ from country to country. That’s why the platform counts each market on its own terms, and brings them together by accounting rules.

Planned

We don’t mix countries

The direction median is calculated within a country. We compare branches within one country, and across countries — with a “different markets” label.

Local currency within a country

Branch revenue in each country is in its own currency, with no conversion — just like at the till.

Consolidated report — at the period’s average rate

When the network looks at everything together, we convert the period’s revenue into the report currency at the average official rate (NBU, ECB). The international standard IAS 21 allows this. [7]

Growth — in constant currency

We convert both periods at the same rate. Then a devaluation doesn’t look like a drop in revenue, and a stronger currency doesn’t look like the salon’s success.

Percentages, days and counts — anything without money — we compare directly, also with a “different markets” label.

Your numbers

Your own ladder next to the global one

Every account will see what share of its clients reach the 2nd, 3rd and 4th visit — by network, branch and direction — next to a global benchmark. And exactly where the most people are lost.

Planned

Share of clients who come for the next visit

demo
Share of clients who come for the next visit
Your salonGlobal benchmark — on average
1st → 2nd38%45%
2nd → 3rd66%70%
3rd → 4th80%79%

In this demo, the biggest loss is between the 1st and 2nd visit: 7 points below the benchmark. That’s where to focus: book the next visit during the first one and send a reminder before the recommended date.

The benchmark is data from another market, not a norm: compare the shape of the curve, not every percentage.

Settings

What you can adjust for your salon

  • Rhythm thresholds1 · 2.5 · 4 usual intervalsIn progress
  • Earliest “gone”12 monthsIn progress
  • Direction median window12 monthsPlanned
  • How much data is enough for the medianthreshold in settingsPlanned
  • Late cancellationN hours, 48 by defaultLive
  • Saved visit7 daysLive
  • Default direction intervalfor each directionLive

The same for everyone

  • New, repeat, regular means 1, 2 and 3+ visits: this way reports from different salons can be compared.
  • A visit is client × day × branch.
  • Booking events come from the change log, not from the current status.
Sources

The research this page relies on

Every source is linked — check for yourself.

  1. Boulevard — Boulevard Report Reveals Top-Performing Salons Retain 56% More First-Time Visitors (2023) — share of clients who come for a 2nd, 3rd and 4th visit; over 11 million appointments, 2,500+ salons, January 2022 — March 2023.
  2. Wübben M., von Wangenheim F. — Instant Customer Base Analysis: Managerial Heuristics Often “Get It Right”. Journal of Marketing, 2008 — simple recency rules identify active customers as well as Pareto/NBD and BG/NBD models.
  3. Platzer M., Reutterer T. — Ticking Away the Moments: Timing Regularity Helps to Better Predict Customer Activity. Marketing Science, 2016 — the more regular a client’s rhythm, the more tracking it improves the forecast.
  4. NICE — Dental checks: intervals between oral health reviews (CG19) — clinical guidelines: the check-up interval for adults is individual, from 3 to 24 months.
  5. HubSpot Knowledge Base — Use lifecycle stages — automatic stage updates only move it forward.
  6. Buckinx W., Van den Poel D. — Customer base analysis: partial defection of behaviourally loyal clients in a non-contractual FMCG retail setting. European Journal of Operational Research, 2005 — partial defection of loyal customers; the best predictors are recency, frequency and amount spent.
  7. IFRS Foundation — IAS 21 The Effects of Changes in Foreign Exchange Rates — para. 40: income and expenses are often translated at the period’s average rate as an approximation of the rates at the transaction dates.

We first test these approaches in practice with the salons we work with. We don’t name them or share their numbers — it is their data, and we protect their confidentiality. That is why we cite open sources: you can open and check each one yourself.

Want to see your clients the same way?

Tell us about your salon or clinic — we’ll show you how the platform will count your clients.

Or write to us directly:
founder@alteg-tech.com

What are you interested in (pick any)

By submitting the form you agree to the Privacy Policy.