Cost per new patient is climbing — fertility networks

Cost per consult is the only figure anyone has built

PRESENTATION — WHAT THE OPERATOR SEES

Cost per consult is rising and nobody can produce a cost per cycle start. A consult is not revenue in this business; a started cycle is. Every diagnosis below turns on which of the two the group is actually measuring.

DIFFERENTIAL — LIKELY CAUSES, MOST LIKELY FIRST

01
The denominator is a consult, not a cycle start

Between the first consultation and a started cycle sit diagnostics, a treatment plan, a cost conversation, a benefits check, financing, a partner who may not be ready, and a wait for the next cycle window. Any of those can end it, and none is visible in a report that stops at the consult. Cost per consult can improve for a year while cost per cycle start worsens.

No published cost-per-cycle-start distribution exists for this category. What circulates is either vendor-produced or clinic-selected, and the register at /sources/ sets out why neither is used as a denominator.

No published distribution — ADMEN source register, /sources/

NOT THIS IF — Cost per cycle start is already reported per clinic and moved in the same direction as cost per consult.
02
Every clinic defines an inquiry differently, so the numerator and denominator do not match

Networks assembled from established clinics inherit a different stack, a different call-handling arrangement and a different definition of an inquiry at each one. A platform cost figure built by summing those is arithmetic performed on incompatible units.

One definition set applied identically under every legacy brand is the precondition for any cost figure at platform level. Where it does not exist, the monthly platform view is a manual rebuild that arrives late and gets argued with.

No published distribution — ADMEN source register, /sources/

NOT THIS IF — One instrumentation layer and one definition set already run under every clinic brand.
03
Benefit eligibility is ending the process and is recorded nowhere

The patient wants to know what is covered, what is not and what the remainder costs, before she commits to a plan. If that answer takes eleven days and arrives by voicemail, a share of patients quietly stop moving — and the marketing report records a successful consult.

Only fifteen states mandate any infertility coverage, and most large employers offering fertility benefits do not cover IVF treatment itself. That is why the benefits answer is a decision point rather than a formality in most of the country.

KFF, coverage of infertility services: fifteen states mandate any infertility coverage, and most large employers offering fertility benefits do not cover IVF treatment itself

NOT THIS IF — Time from benefits request to a clear answer is measured, short, and the drop-off after it is small.
04
The cost was built from ad dashboards rather than from the ledger

A platform running fifteen brands has agency fees, per-clinic vendors, call handling and content production that never appear in a media dashboard. The cost per consult built from the platform understates the real one, and the understatement grows with the number of brands.

The ledger already carries the full number. Rebuilding the cost line from it is arithmetic, not research.

No published distribution — ADMEN source register, /sources/

NOT THIS IF — The cost line is reconciled to the general ledger monthly across every clinic.

HOW TO TELL THEM APART

How to tell these apart in your own numbers

Each of these is a measurement you can run yourself, without us.

01 · The denominator is a consult, not a cycle start

Rebuild the cost line three ways for the same twelve months, per clinic: spend over inquiries, spend over kept consults, spend over cycle starts.

CONFIRMS IF

Cost per consult fell or held while cost per cycle start rose.

EXCLUDES IF

All three moved together.

02 · Every clinic defines an inquiry differently, so the numerator and denominator do not match

Write down each clinic's current definition of an inquiry, a consult and a cycle start, then re-count one month at every clinic against a single definition set.

CONFIRMS IF

The re-counted month differs materially from the reported one at any clinic.

EXCLUDES IF

Every clinic reproduces its reported figures under the common definitions.

03 · Benefit eligibility is ending the process and is recorded nowhere

Time from benefits request to a clear written answer, per clinic, and the proportion of patients who receive one before their next appointment, plus the drop-off rate for those who wait longest.

CONFIRMS IF

The answer routinely arrives after the next appointment, and the long-wait cohort drops off more.

EXCLUDES IF

The answer is fast, written and near-universal before the next appointment.

04 · The cost was built from ad dashboards rather than from the ledger

Rebuild twelve months of cost from the general ledger and vendor invoices, then compare against the dashboard figure, per clinic.

CONFIRMS IF

The ledger figure is materially higher or the gap is widening.

EXCLUDES IF

The two reconcile.

WHAT RESOLVES EACH

What resolves this, and how you will know it resolved

The denominator is a consult, not a cycle start	Rx 04 · marketing attribution →	The cost line rebuilt on a cycle-start denominator, per clinic, from the ledger. What does not work: optimising a consult number. An agency measured on consults will optimise for consults, which is the lead problem one stage further along.
Every clinic defines an inquiry differently, so the numerator and denominator do not match	Rx 04 · marketing attribution →	One instrumentation layer and one definition set under every legacy brand, with the local names left exactly where they are.
Benefit eligibility is ending the process and is recorded nowhere	Rx 03 · patient conversion →	Benefit eligibility treated as a conversion step with a measured turnaround, and financing options presented alongside the answer rather than after it.
The cost was built from ad dashboards rather than from the ledger	Rx 04 · marketing attribution →	Cost reconciled to the general ledger monthly across every clinic, not assembled from platform dashboards.

WHAT "RESOLVED" LOOKS LIKE — Cost per cycle start, per clinic, built from the ledger

MEDIAN

Not published. This vertical carries a specific caution: the figures that circulate are frequently clinic-selected and are not comparable across programmes.

TOP DECILE

Not published, and for the same reason.

TARGET

The network's own instrumented baseline on one definition set. Resolved means cost per cycle start is being reported at all, per clinic, and the spread between clinics is narrowing.

No published distribution — ADMEN source register, /sources/

HOW THIS DIFFERS BY SCALE

How this differs by scale

Single site	One clinic, one definition set already. The consult-to-cycle-start gap is the whole diagnosis.
Group	Two or three clinics is where definition drift begins and where a platform figure first becomes untrustworthy.
Platform	A portfolio of legacy brands cannot produce a credible cost figure until one definition set runs underneath all of them, which is why fertility networks built by acquisition carry this problem and a single clinic does not.

OTHER PRESENTATIONS — FERTILITY NETWORKS

A differential narrows the field. It does not replace the examination — that is what the six weeks are for. Every figure above is an industry reference range, not a client's numbers; those stay sealed. Sources are set out at /sources.

APPLY — 6 / QUARTER →