Scaling eCommerce sales across Australia & New Zealand with an intent-based Shopping architecture, turning a controlled budget into meaningful revenue.
The campaign generated 507 purchases and roughly NZ$98K in revenue from NZ$27K in total ad spend, combining Google and Meta acquisition.
Dawson Watches operates in a competitive eCommerce space where product categories generate a wide range of search intent. A search for “sauna” carries completely different commercial intent from “Osaki massage chair”, and an even stronger buying signal comes from “buy Osaki massage chair.”
Treating all of these equally creates a problem: broad searches generate cheap traffic, while specific product searches carry much stronger purchase intent. The strategy needed to control where Shopping traffic was captured, how aggressively each query type was bid, and which searches were allowed further down the funnel.
Catches broad demand at a low CPC. Aggressive negatives push stronger-intent queries down to the right tier.
Captures brand-aware searches showing greater product awareness than generic queries.
The strongest buying signals get the highest bids, reserved for queries most likely to convert.
| Product Price | Bid |
|---|---|
| 0–499 | $0.02 |
| 500–999 | $0.05 |
| 1,000–1,999 | $0.10 |
| 2,000–3,999 | $0.15 |
| 4,000–5,999 | $0.20 |
| 6,000–7,999 | $0.25 |
| 8,000+ | $0.35 |
| Product Price | Bid |
|---|---|
| 0–499 | $0.15 |
| 500–999 | $0.25 |
| 1,000–1,999 | $0.50 |
| 2,000–3,999 | $1.00 |
| 4,000–5,999 | $1.50 |
| 6,000–7,999 | $2.00 |
| 8,000+ | $2.50 |
| Product Price | Bid |
|---|---|
| 0–499 | $0.50 |
| 500–999 | $1.00 |
| 1,000–1,999 | $1.50 |
| 2,000–3,999 | $2.00 |
| 4,000–5,999 | $3.00 |
| 6,000–7,999 | $4.00 |
| 8,000+ | $5.00 |
Applied across all campaigns to eliminate irrelevant or low-value searches.
Brand terms (Ford, Chrysler, Osaki) pushed out of High Priority toward the right tier.
Specific product names and purchase-oriented queries routed to the highest-intent tier, creating the waterfall effect.
Mobile and tablet bids reduced aggressively based on device-performance analysis, preventing weaker devices from consuming disproportionate budget.
A shared budget for each tier (High, Medium, Low) managed spend at the waterfall level while keeping campaign structures separate by intent.
Continuous checks that queries entered the right tier. When impressions were low, bids were raised gradually, roughly 5%/day on High and 20%/day on Medium and Low, to scale without runaway CPCs.
Captured high-intent demand through Shopping and the waterfall structure.
Contributed additional purchase volume through paid social, with a best cost per purchase of NZ$16.
Structured Shopping architecture and cross-platform acquisition turned a controlled budget into meaningful eCommerce revenue at ~3.6x ROAS.
The waterfall separated broad demand from brand/category demand from high-intent purchase demand. Price variables then added a second dimension, intent × product value, creating a far more controlled bidding environment than treating the entire catalogue as a single pool.
With a large product catalogue, the question isn't “how much should I bid?” It's “which searches deserve which bid?” Let's build a smarter Shopping strategy.