7.3x ROAS.
From clean attribution.

01 — The Problem

Real spend. No idea what it was actually buying.

This client wasn't starting from zero. They were already spending about $25k a month on Google Ads — $75,594 across the 90 days we later analysed — and it was generating real enquiries: 437 Free Skills Assessment form submissions in that window, at a blended cost of $173 per lead.

The problem was underneath the numbers. The account had 30 separate conversion actions live, with multiple set as primary, meaning Google's bidder was optimising toward all of them equally. Maps direction requests and business-listing clicks were being counted the same as a real assessment enquiry. GA4-imported events were duplicating conversions that were already being tracked elsewhere.

Sales knew leads were coming in. Marketing knew spend was going out. Nobody in the business could put those two facts on the same page and say which enrolments actually started with a Google ad.

02 — Our Approach

Strip the signal back to one truth. Then trace it all the way to the sale.

Collapse the conversion stack. We audited all 30 conversion actions and found exactly one that represented a real, qualified lead: the Free Skills Assessment form. Everything else — Maps engagements, direction requests, duplicate phone-click triggers — got demoted out of the primary goal. Google's bidder started optimising toward the one action that actually meant a prospective student.

Rebuild the campaign structure. With a clean signal to bid against, we restructured the account around what was actually converting, instead of one blended set of campaigns trying to average very different buyer intents together.

Stand up lead-source sales attribution. A clean ad account only tells you what Google thinks is working. To know what actually closes, we built a weekly sales process where every single enrolment got tagged with the source it came from at the point of sale — not modelled, not estimated, tagged by the person who closed it.

Put it on a live dashboard. Spend, conversions, cost per lead, click-through rate and week-on-week deltas, all in one place the college's own team could check without opening Google Ads or waiting on a report from us.

03 — The Results

Google Ads went from a line item nobody could account for to the #1 source of revenue in the business.

Once source-tagged attribution had been running for a few months, one thing became obvious: Google Ads wasn't just working, it was the single biggest revenue driver in the business. Across 313 tracked sales worth $2.6M+ in course fees booked over seven months, 150 enrolments — about half of all enrolment revenue, $1.4M — were tagged to Google Ads at the point of sale.

In the matched 90-day window, where we have both spend and attributed sales data lined up, Google Ads produced 70 enrolments worth $554K in course fees against $75,594 in spend — a 7.3x return. Mobile accounted for 75% of conversions in that window, and the best-performing search term in the account was a recognition of prior learning (RPL) query, converting at $14 — well under the $173 blended average, and proof of exactly how efficient the account could be once the signal was clean.

None of this was multi-touch modelling or a guess dressed up as a dashboard. It was sales tagging every enrolment to its source, every week, and the numbers holding up once we started counting properly.

04 — See It For Your Business

If your Google Ads account has been running for years and nobody can say for certain which sales it's actually producing, this is the fix: strip the conversion signal back to one true action, then tag every sale to its source so attribution stops being a guess. If that sounds familiar, get in touch and we'll show you what it looks like for your business.