The brief: a struggling account burning budget

Krikacy had been running Google Ads for over a year. The spend was consistent. The results weren't. ROAS was hovering just above breakeven, the account structure was a mess of old campaigns that had accumulated over time, and nobody had done a proper audit since the account launched.

They came to us with a simple ask: fix it. Make the ads work. They had a good product with proven demand. The issue wasn't the market, it was the execution. We had 60 days before the end of their fiscal quarter, and they wanted to finish strong.

The account audit: understanding what was actually happening

The first two weeks were pure analysis. We pulled every campaign, every ad group, every keyword into a spreadsheet and looked at the data without assumptions. What we found was a common story: years of campaign bloat, keywords cannibalising each other, broad match terms burning budget on irrelevant queries, and bids that hadn't been touched in months.

The conversion tracking was also partially broken. Three of the seven tracked events were firing double, inflating the reported ROAS and making the account look better than it was. That explained why the team had been relatively comfortable with performance when the real numbers were significantly worse.

The rebuild: structure before scale

We paused all existing campaigns and started from a clean structure. Brand campaigns isolated from non-brand. Competitor terms in their own campaigns. Shopping separated from search. Each campaign with a single, clear objective and a bid strategy matched to that objective.

Keyword strategy was rebuilt around intent tiers, with high-intent commercial terms getting aggressive bids, informational terms held back until we had data, negative keyword lists built out exhaustively from the search term reports. No broad match without tightly managed negatives. Every new campaign started with exact and phrase, earning the right to broaden over time.

"We went from burning budget to scaling profitably in under two months."

Creative testing: finding what actually converts

For search, ad copy was rebuilt with three variants per ad group: one benefit-led, one problem-led, and one competitor-aware. We ran them in rotation for two weeks, killed the underperformers, and doubled down on what was working. For Performance Max, we audited the asset groups and rebuilt them with tighter creative themes tied to specific audience signals.

The winning creative across formats was consistently the problem-led copy, with headlines that named the specific pain the customer was feeling before offering the solution. Not surprising in hindsight, but the data confirmed it clearly.

The results: 5.4x ROAS and a record quarter

By day 45, ROAS had climbed to 5.4x. CPA had dropped 62% from where it was when we took over. The account was profitable at every significant keyword cluster. In the final two weeks, we scaled spend by 40% without losing efficiency, which is the real test of whether an account is structurally sound.

Krikacy closed their best revenue quarter on record, driven almost entirely by paid search. The foundation is now set to scale further. The structure supports it, the data is clean, and the learnings from 60 days of testing are baked into every campaign.

Takeaways: what this project reinforced

Broken conversion tracking is more common than most advertisers realise, and it's always the first thing to fix. You cannot optimise what you're not measuring correctly. Every hour spent validating tracking before touching campaigns is an hour of misdirected optimisation avoided.

The other lesson: structure before scale. It's tempting to start adjusting bids and adding keywords immediately. The better move is to get the architecture right first, then optimise within it. A well-structured account compounds. A messy one just gets messier under pressure.

Related reading

The Google Ads playbook for small and mid-size brands
How to build a high-converting landing page in 2026