Where the money is working, where it's leaking, and the moves that convert spend into revenue — grounded in the last 90 days of account data.
| Campaign | Platform | 90-Day Spend | ROAS | Verdict |
|---|---|---|---|---|
| Hardwall Cleanroom (AI Max) | $5,418 | 20× | Scale up | |
| PMax: Cleanrooms | Bing | $479 | 21× | Starved — fund it |
| PMax: Cleanrooms | $7,860 | 14× | Scale up | |
| Sales-Shopping | $6,092 | 15× | Hold / grow | |
| PMax: Gloves & Wipes | Bing | $7,566 | 8–10× | Grow |
| PMax: Furniture / Gloves | $10,703 | 3× | Fix or trim | |
| PMax: Apparel | $4,206 | 2× | Cut 50% | |
| PMax: Apparel · Sales-Shopping | Bing | $2,605 | <1× | Pause |
Cleanrooms + consumables return 8–21×. Furniture & apparel return 2–3× and are subsidised by the winners. ROAS uses platform-reported value — trust the ranking, treat absolute multiples as generous.
Raise daily budgets on Hardwall AI Max, PMax Cleanrooms (both platforms), and Sales-Shopping. Fund the Bing Cleanrooms campaign — 21× on $479 is severely under-invested. Cut Google Apparel by half; pause Bing Apparel and Sales-Shopping.
Capture GCLID on quote submission → join to closed order in Insightly → upload to Google & Bing. Your best campaigns are cleanroom quotes that close by phone — invisible to Smart Bidding today. This feeds the 14–20× campaigns the signal they're missing. Ties directly to the packaging-quote join-key fix already scoped.
AOV spans $70 gloves to $16K cleanrooms — optimising for conversion count chases cheap orders. Switch Cleanrooms / Hardwall / Softwall to Max Conversion Value with a target ROAS (needs 30+ clean conversions). Keep consumables on Max Conversions where volume is the goal.
Give each winning asset group the full set: 5+ headlines, long headlines, multiple descriptions, real product + lifestyle images, and supplied video. Segment by product family — cleanrooms shouldn't share creative with gloves. Add customer-list & high-intent audience signals ("ISO cleanroom," "ESD workstation").
Monthly review of PMax & DSA search terms. Build a shared negative list: consumer/retail terms, job-seekers, out-of-catalog categories. Exclude thin/out-of-stock pages from DSA so paid clicks don't hit dead ends. Bing Sales-Shopping CTR of 0.33% signals wasted impressions.
Bing is 16% of spend but converts consumables at ~half Google's cost-per-conversion ($50 vs $200) — the right audience for MPS. Import winning Google campaigns, then let them bid independently. Grow Bing consumables & cleanrooms only; don't expand the unprofitable Bing apparel/shopping.
Site guardrail: aggressive paid traffic drives uncached PHP requests against a limited-worker P1 plan. Confirm landing pages cache for ad traffic (watch gclid query-string cache-busting) before scaling spend materially — slow pages lower Quality Score and conversion rate at once.