Redefine Logic
Redefine Logic Media
Case study 01  ·  Google Ad Grant  ·  Nonprofit

We bought 67% less traffic.
They got four times the conversions.

A multicultural employment and language agency had run a Google Ad Grant for almost a decade — most of it underperforming, some of it in‑house, all of it measured by the wrong number. Here is what changed when we stopped optimising for clicks.

improvement in conversion rate — from 0.43% to 5.33%
0 .5×
more conversions — on roughly a third of the traffic
+ 0 %
click‑through rate by mid‑2024 — nearly double the 6.77% we started from
0 .22%
The scenario

Ten years of grant money,
spent on the wrong number.

An agency had set the Google Ad Grant up years earlier and got them a moderate lift. After about a year the work came back in‑house, where it sat. Nobody there had the hours to run it properly, and nobody outside was watching.

By the time we were brought in, the account had sprawled into branches of misconfigured campaigns and ad groups. It was grossly underperforming — while still consuming the full grant every month.

The account was not broken in an obvious way. That is what made it survive a decade. On paper it looked fine.

  • Client

    Nonprofit multicultural employment and language agency, Ontario. Named client withheld at their preference.

  • Engagement

    June 2023 — present. Ongoing management, not a one‑off audit.

  • Program

    Google Ad Grant — up to $10,000 per month in free search advertising for qualified charities.

  • The brief

    Make the grant produce something. It had been running for the better part of a decade without doing so.

Before Redefine Log!c  ·  Jan – Dec 2022

“It performed well on paper.”

Sixteen thousand clicks a year. A click‑through rate above the nonprofit average. And sixty‑eight conversions to show for it — roughly six a month, from a grant worth $120,000 a year in advertising.

A conventional campaign, optimised for clicks and spend. It hit every number it was being measured on and delivered almost nothing. This is exactly how organizations lose faith in digital marketing.

What we actually did

We changed what the campaign
was trying to win.

  • Audited the sprawl, then cut it

    A decade of in‑house edits had left branches of misconfigured campaigns and ad groups running in parallel. We mapped what was actually live, eliminated underperforming keywords and entire ad groups, and stopped paying attention to the parts that were only generating activity.

  • Rebuilt the structure around intent

    Ad groups were realigned to what someone is actually trying to do when they search — find language classes, find employment support — rather than to keyword lists inherited from the original build.

  • Moved the target metric to conversions

    The campaign had been optimised for clicks because clicks are what the grant obviously buys. We changed the objective to conversion rate optimisation, so every click had to be worth something rather than merely exist.

  • Accepted less traffic on purpose

    This is the part that needs explaining to a board. Total clicks fell by two thirds. That was the plan, not a side effect — and it is why the results moved.

Three years, side by side

2022
Before us

CLICKS
16,000

CTR
6.77%

CONVERSIONS
68

CONVERSION RATE
0.43%

2023
First full year

CLICKS
5,200 ↓ 67.5%

CTR
8.03%

CONVERSIONS
277 ↑ 307%

CONVERSION RATE
5.33% 12.5×

2024
Jan – Jun 20 only

CLICKS
1,700

CTR
13.22% ↑ 95%

More than one in eight people who saw the ad clicked it — nearly double where we started, and well above the nonprofit search benchmark.

The part that needs explaining to a board

A grant that spends is not
a grant that works.

The Google Ad Grant hands a charity up to $10,000 a month in search advertising. Because it is free, almost everyone measures it by how much of it they managed to use — and clicks are the easiest proof of use.

That is how an account runs for ten years looking healthy while producing six outcomes a month. Spending the grant is not the achievement. The grant is not the prize.

When we cut the traffic by two thirds, the monthly report got worse on the metric the client had been taught to watch. It took a conversation to get agreement to do it anyway. That conversation is most of the job.

Running an Ad Grant that isn’t producing?

Bring the last twelve months of your account to a thirty‑minute review and we will tell you what it is actually doing — including, if it applies, that it is fine and you do not need us.