What is cost per enrolled student?
Your total marketing spend for a period divided by the number of students from that spend who enrolled and started a course. Not the people who filled a form, not the people who booked a call, and not the people who paid a deposit and vanished: the ones who showed up to learn.
Cost per lead measures a different event entirely. A lead is an expression of interest; an enrolled student is revenue. Between the two sit funding checks, eligibility, timing against your intake dates, and every drop-off in your admissions process. Judging marketing on the first event while your business runs on the second is how a provider ends up with a full inbox and a half-empty classroom.
Cost per lead is still useful as a diagnostic inside a campaign. It is the wrong number to buy on.
Why does buying on lead volume punish cohort businesses?
Because a cohort business sells a capped number of seats inside a fixed enrolment window. Lead volume is unlimited and always-on; your revenue is neither. The mismatch punishes you in both directions.
Too few of the right leads and a cohort runs below capacity, which is pure margin lost: as a benchmark example, the course costs the same to teach at 60% full as at 100%. Too many of the wrong leads and your admissions team drowns in unqualified inquiries exactly when it should be converting the qualified ones before the window closes. A lead that arrives after the intake date passed is not a cheap lead, it is a rejected applicant and a refunded expectation.
A lead-volume buyer also pays for the platform's laziest behavior. Ask an ad platform for maximum leads at minimum cost and it will find you the people most likely to fill a form, who are reliably not the people most likely to fund, enrol, and start. The mechanics of matching spend to intake windows are covered in how to fill training cohorts in advance.
How do you calculate cost per enrolled student?
Divide marketing spend for the period by the number of students from that spend who enrolled and started. The arithmetic is trivial. The tracking behind it is the actual work, and it is worth being honest about that.
What the tracking honestly requires: a CRM or student management system with an enrolment stage that someone actually maintains, source and campaign captured on every inquiry and carried through to that stage, and patience, because an enrolment can close weeks or months after the click, so the number is only readable once a cohort's window has closed. None of this is exotic. All of it is work that most providers and most agencies have simply never wired up.
Whether the resulting number is acceptable is a margin question: set it against your course price and your seat capacity, not against anyone else's benchmark. What a sensible program costs to run is covered in what vocational training marketing costs.
Why does almost nobody publish cost per enrolled student?
Because almost nobody tracks enrolment back to the ad. The agency's data ends at the form fill, the provider's data starts at the application, and the join between the two is nobody's job.
There is also an incentive problem. Cost per lead nearly always looks better than cost per enrolled student, sometimes by two orders of magnitude, so the flattering number is the one that gets reported and the honest number is the one that would invite hard questions. An agency graded on leads has no reason to build the enrolment join, and a provider who never sees the join has no way to grade the agency on anything else. The loop closes itself.
This is worth using as a filter. Ask any agency pitching you one question: "How will you know, and show me, which enrolled students came from your campaigns?" A real answer describes CRM stages, source capture, and reporting cadence. A vague answer about pixels and platform conversions means they intend to be graded on leads. If you are weighing whether to build this capability yourself instead, read training marketing agency vs in-house.
What do the verified numbers actually look like?
Here is our published proof from the training and membership space, labelled precisely, because a page arguing for honest metrics should not blur its own. Elec Training's cost per lead fell from £257 to £5.28, per our client data. myTTConline held a 2.99x return across 21 consecutive months, per our client data. Note which metric each number is.
A lead-cost collapse that mattered because the cohorts behind it filled.
The pattern across all three: the number that got optimized was the one closest to revenue that the tracking could reach, and the lead cost improved as a side effect. The same discipline holds across verticals: in our anonymized immigration engagement, cost per signed case fell from $2,372 to $1,064 across a dataset of n=1,391 signed cases. That is the order of operations to insist on.