The problem
A regulated Australian bank needed to fund a national term-deposit growth campaign against a set of aggressive capital targets. The existing Google Ads account had decent budget but fundamental problems: campaigns structured around legacy keywords, no measurement beyond form submission, and a landing experience that worked for brand traffic but throttled cold prospects.
The gap between "lead" and "funded account" was a black box. Nobody could say which campaigns, keywords or audiences were producing actual deposits — only which were producing signups. And in a category where a lead costs $60+ and only a subset of signups fund within 30 days, that gap was the difference between profitable and unprofitable.
What we did
1. Rebuilt the account around funded-account value
We restructured campaigns around funnel stage and intent, not product line. High-intent branded and category terms were isolated from discovery-stage terms. Smart bidding was re-targeted at a custom conversion event — not "lead", but "funded account within 30 days" — imported back into Google Ads via the offline conversion API.
2. Engineered lead-to-funded measurement
This was the unlock. We built a server-side data pipeline connecting application events, core banking data and Google Ads offline conversions. Every funded account was traced back to the campaign, ad group, keyword and creative that produced the original click — then fed back to Google as a value-weighted conversion signal. Within six weeks, Smart Bidding was optimising against real deposits, not form fills.
3. Dominated the auction
With accurate signals, budget flowed to the highest-yielding queries. We pushed the brand to a 98.39% top impression share on non-brand commercial terms — meaning 98 of every 100 times a ready-to-deposit prospect searched, the bank was at the top of the page.
4. Built landing pages paired to intent
Our CRO team built dedicated landing pages for each campaign group — a stark departure from the homepage-as-landing-page default. Each page pre-sold the specific product variant that matched the ad's intent. Conversion rate from click to application roughly doubled.
The result
$1.2 billion in new term deposits over a 12-month cycle. 12,000 new funded accounts — a run-rate of roughly 1,037 per month. Top impression share climbed to 98.39%, and cost per funded account dropped well below target.
More importantly: the bank now has a measurement system that ties every marketing dollar to a funded deposit. Budget decisions are data-led, not opinion-led. The engine compounds.
Why it worked
This wasn't a story of better ads. The ads got better because we fixed what Google was learning on. Most accounts at this scale optimise against proxy conversions that correlate loosely with revenue. By rebuilding the signal all the way to funded deposits — and then pairing that with landing pages designed for the specific intent of each campaign — we turned Google's bidding system into a genuinely aligned partner rather than a blunt instrument.
The same operating system — DDNA — drives every DDX engagement. Deep dive, design, deploy, double down. Details change. The discipline doesn't.