Results · a national Christian university college · twelve-month window

The 66 per cent came from five specific changes. Here they are.

The national Christian university college at the centre of this case study is a Christian multi-campus higher education provider offering CSP and full-fee places across theology, education, business and social science. The account was already running paid search and social when the engagement began. Nothing about the budget changed. The structure changed.

01 · The situation

A capable account, quietly under-performing.

The account had been running for years under a competent generalist agency. Reporting looked reasonable. Enrolment enquiries were arriving. But the number that matters — enrolled students against target — had stopped moving, and the cost of each enquiry was drifting upward every quarter. The board was asking the marketing team to explain why more budget was producing the same result.

The pre-engagement audit found three structural problems. Two agencies had been asked to look at it before. Neither had named all three.

02 · The five changes

Order matters. Do these in a different sequence and half of them don't hold.

Restructured a six-campaign account into intake-window SKAG structure

Domestic semester one, domestic semester two, and international rolling each got their own campaign, budget, creative and negative-keyword list. The pre-existing account had all three timelines competing inside the same campaigns, and the auction was reliably won by the highest-bidding audience — usually not the one the semester was actually recruiting from. Structural change · Google Ads account architecture

Killed budget on domestic-only search terms auctioning against international-only audiences

Campaign cannibalisation was consuming 27 per cent of monthly spend — the same keywords were being bid on by three internal campaigns simultaneously. This one change alone freed $6,400 per month, without touching creative or landing pages. Financial change · direct media saving

Rebuilt landing pages against sub-sector intent

A prospective Master of Education applicant and a prospective undergraduate Bible college applicant no longer land on the same page and see the same three CTAs. Sub-sector intent split the landing page architecture into six templates driven by course family and intake type. Conversion rate on the same paid traffic more than tripled within four months of the rebuild. Structural change · site architecture and CTA logic

Rewrote Performance Max country and language allowlists

Audited lead source data showed 19 per cent of leads arriving from countries the sales team could not enrol from — the ML expansion feature inside Performance Max was ignoring the account's stated geo restrictions. Documented country and language allowlists were built and enforced, and Performance Max was demoted to a supporting role rather than a lead engine. Structural change · Performance Max governance

Instrumented offline conversion imports

Paid channels started optimising against enrolled-student value, not form fills. This is the shift that made the 360 per cent conversion figure durable at month twelve rather than a month-three spike. Without this change, everything above still works — but the machine-learning bidding gradually drifts back toward whatever produces form-fill volume, which is not what the institution actually needs. Measurement change · conversion API and CRM integration

Engagement with a national Christian university college · Q3 2025 – ongoing · results audited internally and confirmed via Google Ads Support · client name available under NDA

03 · What the numbers looked like month-to-month

The 66 per cent showed up at month twelve, not month one.

Blended CPC · domestic + international · Google Ads

	Month 1 · pre-engagement	Month 4	Month 12
Blended cost per click	$4.82	$3.14	$1.64
Enquiry-to-application conversion	2.1%	5.4%	9.7%
Monthly media spend	$14,200	$14,200	$14,200
Wrong-geography leads	19%	6%	1%
Cannibalisation loss	$3,834	$920	$0

Where the biggest saving actually came from

Cannibalisation loss · monthly spend consumed by internal auction competition

Month 1

$3,834

Month 4

$920

Month 12

Restructuring the account into intake-window SKAG structure with tight negative-keyword hygiene was the cheapest change made in the engagement and accounted for roughly two-fifths of the twelve-month CPC saving. See change 02 in the breakdown above.

The spend line does not move. That is the point. This is not a "spend more, get more" story — it is a "spend the same, get the right traffic" story, and the difference between the two is the whole engagement.

04 · What has not happened

The honest disclosures.

The 360 per cent is on enrolment enquiry conversion, not enrolled-student conversion.

Enrolled-student data lags by a full intake cycle. Directional signal from month-15 reporting is positive and consistent with the enquiry data, but calling it a 360 per cent enrolment lift right now would be dishonest.

Not every campaign was a win.

A test on TikTok organic content aimed at international-agent audiences ran for three months and produced negligible qualified traffic. It was killed and the budget moved to LinkedIn agent-network ads, which have since performed.

Attribution is imperfect.

Some of the enquiry lift is attributable to organic and brand-search improvements the internal marketing team drove independently over the same period. The 66 per cent CPC figure is clean; the 360 per cent enquiry conversion figure is directionally correct but is not solely a paid-channel result.

The 66 per cent came from five specific changes. Here they are.

Cannibalisation loss · monthly spend consumed by internal auction competition

Bring your account. Same audit conversation.