Goals
Targets come from CCC's own measured history — every recommendation below links back to one of these.
Zero money-losing ad spend
Break-even verified ROAS is 2.23x at CCC's 44.8% blended gross margin (QuickBooks via Windsor, verified 2026-07-08). Any campaign verified below floor is buying revenue at a loss.
Source: context_pack.py roas_floor; QuickBooks P&L 17-month verified
Google efficiency back to good-year median
Google spend ($14,921/mo last 6mo) is ABOVE the 2025 peak window but verified ROAS is roughly half (7.1x vs 16.2x). This is efficiency erosion, not underspend. Target = CCC's own 2025 full-year median.
Source: ThoughtMetric 2yr history (2025-01..2026-07), context_pack.py
Bing ROAS: Stay Above Full-Year Median
Target is Bing's own trailing 2-year median ROAS -- the bar it has already proven it can clear.
Source: ThoughtMetric 2yr history, context_pack.py
Meta above break-even
Meta went NEGATIVE in July 2026 (0.08x verified) and its platform-reported value is separately inflated ~9x vs ThoughtMetric (64.7% of claimed value is 1-day view-through). Until it clears break-even it is an awareness/retargeting channel only, not a volume channel.
Source: ThoughtMetric vs Meta Marketing API, attribution breakdown 2026-07-09
Physical inventory on hand
Chris's #1 stated lever (2026-07-09/11): free warehouse cash and space by selling down overstock. Actual = total physical units on hand (manual warehouse count, 2026-07-24 baseline, ccc_inventory_snapshots). Target intentionally null: the goal is the downward TREND in on-hand units, tracked in goal history from today. NOTE (2026-07-24): this used to track 'excess units past 26-week cover', but that required a velocity/SKU mapping with an unresolved double-counting ambiguity (same root cause as the 2026-07-23 '$7M' incident) -- simplified to a direct on-hand count until that mapping is confirmed.
Source: agentic_ad_ops/inventory_manual_tracking.py; ccc_inventory_snapshots (Supabase)
Ad spend by platform — by month
Spend refreshes live every day for the current month; orders and revenue are ThoughtMetric-measured (linearNonDirect, multi-touch -- "orders" is fractional because credit splits across touchpoints, not an estimate). "All platforms" is the blended sum, not an average.
| Aug 2025 | Jul 2026 | Aug 2026 (MTD) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | |
| All platforms | $15,654 | 120.0 | $89,771 | 5.73x | $18,845 | 89.7 | $85,240 | 4.52x | $11,945 | 47.5 | $42,581 | 3.56x |
| Bing | $1,003 | 11.3 | $7,164 | 7.14x | $1,935 | 10.3 | $5,618 | 2.90x | $1,232 | 6.3 | $5,463 | 4.43x |
| $3,105 | 6.9 | $3,480 | 1.12x | $1,963 | 4.7 | $2,079 | 1.06x | $1,111 | 1.8 | $884 | 0.80x | |
| $11,546 | 101.8 | $79,127 | 6.85x | $14,947 | 74.7 | $77,544 | 5.19x | $9,603 | 39.5 | $36,234 | 3.77x | |
Show full history (20 months)
| Jan 2025 | Feb 2025 | Mar 2025 | Apr 2025 | May 2025 | Jun 2025 | Jul 2025 | Aug 2025 | Sep 2025 | Oct 2025 | Nov 2025 | Dec 2025 | Jan 2026 | Feb 2026 | Mar 2026 | Apr 2026 | May 2026 | Jun 2026 | Jul 2026 | Aug 2026 (MTD) | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | Spend | Orders | Revenue | ROAS | |
| All platforms | $924 | 2.0 | $1,310 | 1.42x | $11,002 | 170.2 | $194,733 | 17.70x | $17,386 | 330.3 | $338,214 | 19.45x | $21,471 | 273.2 | $259,214 | 12.07x | $23,022 | 292.3 | $282,401 | 12.27x | $22,863 | 205.5 | $199,405 | 8.72x | $19,626 | 153.6 | $146,479 | 7.46x | $15,654 | 120.0 | $89,771 | 5.73x | $14,122 | 74.4 | $66,234 | 4.69x | $14,919 | 95.2 | $61,057 | 4.09x | $15,742 | 103.5 | $65,949 | 4.19x | $10,978 | 66.1 | $43,167 | 3.93x | $14,408 | 75.4 | $57,269 | 3.97x | $13,700 | 81.0 | $68,329 | 4.99x | $20,902 | 168.7 | $167,293 | 8.00x | $23,186 | 210.5 | $255,109 | 11.00x | $25,651 | 154.3 | $183,645 | 7.16x | $24,130 | 140.6 | $167,535 | 6.94x | $18,845 | 89.7 | $85,240 | 4.52x | $11,945 | 47.5 | $42,581 | 3.56x |
| Bing | $39 | 0.0 | $0 | 0.00x | $990 | 12.0 | $11,774 | 11.90x | $2,265 | 41.8 | $43,007 | 18.98x | $3,298 | 31.0 | $26,951 | 8.17x | $3,505 | 52.2 | $51,858 | 14.79x | $3,885 | 24.1 | $21,485 | 5.53x | $3,061 | 20.1 | $15,374 | 5.02x | $1,003 | 11.3 | $7,164 | 7.14x | $979 | 7.6 | $5,485 | 5.60x | $1,003 | 7.9 | $8,394 | 8.37x | $995 | 8.5 | $8,632 | 8.68x | $1,478 | 5.5 | $5,441 | 3.68x | $1,223 | 7.1 | $4,410 | 3.61x | $718 | 7.2 | $10,798 | 15.04x | $2,047 | 16.9 | $19,752 | 9.65x | $2,590 | 30.9 | $48,902 | 18.88x | $2,591 | 14.7 | $20,140 | 7.77x | $2,590 | 14.4 | $15,461 | 5.97x | $1,935 | 10.3 | $5,618 | 2.90x | $1,232 | 6.3 | $5,463 | 4.43x |
| $104 | 0.0 | $0 | 0.00x | $2,292 | 24.6 | $35,571 | 15.52x | $2,899 | 45.4 | $43,732 | 15.08x | $2,551 | 28.3 | $24,919 | 9.77x | $2,924 | 41.7 | $36,590 | 12.51x | $3,271 | 29.6 | $24,167 | 7.39x | $3,010 | 11.5 | $13,512 | 4.49x | $3,105 | 6.9 | $3,480 | 1.12x | $2,294 | 1.7 | $1,819 | 0.79x | $2,505 | 2.4 | $1,774 | 0.71x | $3,149 | 11.0 | $9,149 | 2.91x | $2,609 | 6.4 | $4,975 | 1.91x | $2,495 | 7.1 | $4,110 | 1.65x | $2,416 | 6.5 | $7,384 | 3.06x | $1,737 | 32.2 | $16,635 | 9.58x | $2,908 | 46.6 | $16,035 | 5.51x | $3,092 | 15.2 | $10,965 | 3.55x | $2,793 | 9.8 | $10,357 | 3.71x | $1,963 | 4.7 | $2,079 | 1.06x | $1,111 | 1.8 | $884 | 0.80x | |
| $780 | 2.0 | $1,310 | 1.68x | $7,720 | 133.6 | $147,389 | 19.09x | $12,221 | 243.2 | $251,475 | 20.58x | $15,623 | 214.0 | $207,344 | 13.27x | $16,593 | 198.4 | $193,952 | 11.69x | $15,707 | 151.9 | $153,753 | 9.79x | $13,555 | 122.0 | $117,593 | 8.68x | $11,546 | 101.8 | $79,127 | 6.85x | $10,849 | 65.1 | $58,930 | 5.43x | $11,411 | 84.9 | $50,889 | 4.46x | $11,599 | 84.0 | $48,167 | 4.15x | $6,891 | 54.2 | $32,752 | 4.75x | $10,689 | 61.2 | $48,749 | 4.56x | $10,566 | 67.3 | $50,147 | 4.75x | $17,118 | 119.6 | $130,906 | 7.65x | $17,689 | 133.0 | $190,172 | 10.75x | $19,968 | 124.4 | $152,540 | 7.64x | $18,747 | 116.4 | $141,718 | 7.56x | $14,947 | 74.7 | $77,544 | 5.19x | $9,603 | 39.5 | $36,234 | 3.77x | |
Inventory mission — to healthy stock by 2026-12-01 ✎ edit deadline ✎ edit cover weeks
Healthy = on-hand equals 18 weeks of seasonally-adjusted FORWARD demand (bulky imported durables run 2-4 turns/yr; ~10-14 wk container lead times set the cover). Deadline 2026-12-01: excess clears before the Sep-Jan trough so cash and warehouse are free ahead of the Feb-May peak. Both numbers are plan settings -- tap ✎ edit beside them to change them right here.
⚠ COUNT WARNING: Shopify on-hand values at ~$1,926,590 at REAL unit cost vs $900,000 in QuickBooks. Unit figures below are QB-scaled–claimed ranges; the plan is set against the conservative bound. Know the real number? Tap ✎ enter actual count on any family row -- the range collapses on the next refresh (~5 min).
Rhode Island Homestead 1,277 on hand · operator count 2026-07-24 ✎ enter actual count target 421 sell 856 need 59.9–59.9/wk · season gives ~12.9/wk CLIENT DECISION BEHIND
BEYOND ADS ALONE -- needs a decision: deeper cut, later deadline, or accept carry · at full ads+price intensity, healthy by ~2027-04-11 · 0 units moved since baseline (required so far: 359.4).
| Product / run combination | On hand | Sale price | Unit cost | Margin |
|---|---|---|---|---|
| RIH Coop + Run | 1,191 | $1,577 | $314 | 15.71/wk |
| ↳ Coop + Original Run | 207 | $1,499 | $303 | $1,196 |
| ↳ Coop + Walk-In Run | 351 | $1,699 | $342 | $1,357 |
| ↳ Coop + Brahma Run | 275 | $2,099 | $425 | $1,674 |
| ↳ Coop Only | 356 | $1,099 | $206 | $893 |
| RIH Hen House Only | 356 | $1,099 | $206 | 1.24/wk |
| ↳ Default | 355 | $1,099 | $206 | $893 |
| RIH Roof Bundle | 173 | $1,905 | $375 | 3.5/wk |
| ↳ Orig Run + Roof | 36 | $1,749 | $337 | $1,412 |
| ↳ Walk-In + Roof | 136 | $1,899 | $388 | $1,511 |
| ↳ Orig Run + Roof (BR) | 69 | $1,749 | $337 | $1,412 |
| ↳ Walk-In + Roof (BR) | 182 | $1,999 | $388 | $1,611 |
| RIH Original Run Ext | 36 | $500 | est. | 1.24/wk |
Orpington 709 on hand · operator count 2026-07-24 ✎ enter actual count target 145 sell 564 need 39.5–39.5/wk · season gives ~4.5/wk CLIENT DECISION BEHIND
BEYOND ADS ALONE -- needs a decision: deeper cut, later deadline, or accept carry · at full ads+price intensity, healthy by ~2027-11-08 · 0 units moved since baseline (required so far: 237.0).
| Product / run combination | On hand | Sale price | Unit cost | Margin |
|---|---|---|---|---|
| Orpington Lodge | 824 | $985 | $232 | 4.9/wk |
| ↳ Coop + Original Run | 124 | $999 | $209 | $790 |
| ↳ Coop + Walk-In Run | 35 | $1,399 | $281 | $1,118 |
| ↳ Coop + Brahma Run | 107 | $1,799 | $361 | $1,438 |
| ↳ Coop Only | 557 | $799 | $209 | $590 |
| Orpington Roof Bundle | 159 | $1,338 | $230 | 1.71/wk |
| ↳ Orig + Roof | 124 | $1,199 | $209 | $990 |
| ↳ Walk-In + Roof | 35 | $1,649 | $281 | $1,368 |
| ↳ Orig + Roof (BR) | 83 | $1,215 | $209 | $1,006 |
| ↳ Walk-In + Roof (BR) | 51 | $1,665 | $281 | $1,384 |
| Orpington Run Ext | 124 | $299 | $69 | 0.86/wk |
| ↳ Default | 124 | $299 | $69 | $230 |
Leghorn 633 on hand · operator count 2026-07-24 ✎ enter actual count target 112 sell 521 need 36.5–36.5/wk · season gives ~3.4/wk CLIENT DECISION BEHIND
BEYOND ADS ALONE -- needs a decision: deeper cut, later deadline, or accept carry · at full ads+price intensity, healthy by ~2028-02-05 · 0 units moved since baseline (required so far: 219.0).
| Product / run combination | On hand | Sale price | Unit cost | Margin |
|---|---|---|---|---|
| Leghorn Cottage | 758 | $598 | $145 | 4.43/wk |
| ↳ Coop + Run | 634 | $549 | $145 | $404 |
| ↳ Coop + Extra 3ft Run | 124 | $849 | $145 | $704 |
| Leghorn Roof Bundle | 125 | $999 | $246 | 0.31/wk |
| ↳ Default | 125 | $999 | $246 | $753 |
| Leghorn Run Ext | 124 | $299 | $69 | 1.01/wk |
| ↳ Default | 122 | $299 | $69 | $230 |
Brahma 672 on hand · operator count 2026-07-27 ✎ enter actual count target 210 sell 462 need 32.3–32.3/wk · season gives ~6.4/wk CLIENT DECISION BEHIND
BEYOND ADS ALONE -- needs a decision: deeper cut, later deadline, or accept carry · at full ads+price intensity, healthy by ~2027-04-30 · 0 units moved since baseline (required so far: 193.8).
| Product / run combination | On hand | Sale price | Unit cost | Margin |
|---|---|---|---|---|
| Brahma Coop | 1,061 | $2,493 | $433 | 8.17/wk |
| ↳ Coop + Walk-In Run | 373 | $2,599 | $468 | $2,131 |
| ↳ Coop + Brahma Run | 273 | $3,099 | $550 | $2,549 |
| ↳ Coop Only | 415 | $1,999 | $324 | $1,675 |
| XL Brahma Run Ext | 654 | $1,299 | $226 | 2.64/wk |
| ↳ Orpington Coop | 108 | $1,299 | $226 | $1,073 |
| ↳ Rhode Island Coop | 273 | $1,299 | $226 | $1,073 |
| ↳ Brahma Coop | 273 | $1,299 | $226 | $1,073 |
Walk-In Run 0 on hand · operator count 2026-07-27 ✎ enter actual count target 18 sell 0 need 0.0–0.0/wk · season gives ~0.6/wk ON TRACK —
ON TRACK -- season alone gets there · 0 units moved since baseline (required so far: 0.0).
| Product / run combination | On hand | Sale price | Unit cost | Margin |
|---|---|---|---|---|
| Walk-In Run Ext | 825 | $703 | $140 | 0.93/wk |
| ↳ Orpington Coop | 35 | $799 | $144 | $655 |
| ↳ Rhode Island Coop | 395 | $699 | $140 | $559 |
| ↳ Brahma Coop | 395 | $699 | $140 | $559 |
Rollup by family (a coop and its run combinations sell as a set); tap a family for its products and variants with real Shopify sale prices and unit costs. Verdicts are planned on the conservative (QB-scaled) count. Pace reads weekly.
Executive summary
$632,000 in coop inventory is sitting in the warehouse going into the three slowest months of the year. August through October runs at about 60 cents on spring volume and December is worse. The window to clear is now -- a 15% sale still hits positive margin on every model, and the space we free up is space we need for February when demand comes back. RIH and Brahma ship first: together they account for about half of coop contribution margin and they take up the most floor space per unit.
Non-brand Google search is returning roughly $8 in revenue for every dollar spent and it is only capturing 10% of available searches -- three-quarters of what it loses goes to competitors because the bids are not competitive enough, not because demand is thin. Bing is confirmed at $4 to $5 back per dollar and runs under $1,000 a month with room to grow. The SMS list is the most underrated asset we have: 14,740 subscribers, zero cost to reach them, and the top campaigns on record pulled $8,500 to $9,600 per send.
Three structural problems are costing us real money and all three are fixable this week. Our two highest-returning Google campaigns -- both branded, returning $7 to $8 per dollar -- are bidding to maximize clicks instead of revenue, and they are also hitting a daily budget cap that cuts off roughly 20% of available impressions; we are switching the bid strategy and raising the cap. The PMax shopping campaign has its revenue target set at 8x return, more than three times our break-even, so it is automatically declining profitable sales to protect a number that is too tight; we are dropping it to 5.5x. On Meta, the ad set that should catch people who abandoned their cart has an exclusion list that blocks the very people it is supposed to reach -- a one-time audience fix that goes in before any budget moves. We are not adding money to Google or Meta right now because our own confirmed revenue numbers and what the platforms report are too far apart to bet new dollars on either.
This week: switch the Google branded bid strategies to revenue-optimized and raise those budgets to match the proven lost impressions, drop the PMax target to 5.5x, fix the Meta remarketing audience exclusions, and send the RIH and Brahma clearance to the SMS list. We are deliberately not launching anything new on Meta, not increasing overall Google spend while the last seven days are near break-even, and not touching Bing budgets until the Bing conversion tracking tag -- currently completely inactive -- is repaired and confirmed firing.
How we will know if we are wrong:
- If Google branded revenue per dollar falls in the ten days after the bid strategy switch, we revert immediately.
- If PMax spend rises after the target drop but confirmed order volume does not follow within two weeks, we raise the target back and cut budget instead.
- If the RIH and Brahma clearance SMS send does not produce at least 25 coop orders in 72 hours, the offer depth or the framing needs adjusting before the next push.
Recommendations
Each card: what to do, why (source-tagged evidence), and the expected, measurable outcome. Status persists across days — resolved items move to the archive below. Every executed change is logged to the change engine and measured on its stated window.
Get Rhode Island Homestead from ~1,277 on hand down to a healthy 421 (18 weeks of forward seasonal demand).
Details — evidence, verification, testing window (tap to expand)
What
Get Rhode Island Homestead from ~1,277 on hand down to a healthy 421 (18 weeks of forward seasonal demand). That means selling ~856 units in 14 weeks: ~60/wk vs ~12.9/wk the season gives unaided -- a 4.63x lift at the conservative count. Verdict: BEYOND ADS ALONE -- needs a decision: deeper cut, later deadline, or accept carry.
Why
Selling through generates ~$1,275,732 of revenue at current sale prices. Target and deadline are benchmark-based plan settings (editable right on the Mission board), seasonally adjusted from 30 months of Shopify sales. COUNT WARNING: Shopify on-hand values at ~$1,926,590 at cost vs $900,000 in QuickBooks -- unit figures shown as QB-scaled..claimed ranges until reconciled. Source: Shopify inventory+velocity cache 2026-07-11, real unit economics where pulled.
Expected outcome
Tracking: the Inventory Mission board reads weekly -- family units/wk vs the required ~60/wk line; pace flips ON PACE / BEHIND / AHEAD (currently BEHIND). Stop or escalate at each Monday read, not mid-week.
Corpus evidence
Retail and apparel advertisers should ensure product feeds have high-quality images and accurate pricing to benefit from price-tracking and virtual try-on surfaces; confirm Google Lens eligibility by auditing image quality and structured data in Merchant Center before peak back-to-school traffic in
Prioritize Merchant Center feed hygiene (titles, images, GTIN, availability) as a first-order ecommerce ROAS lever for CCC.
Get Brahma from ~672 on hand down to a healthy 210 (18 weeks of forward seasonal demand).
Details — evidence, verification, testing window (tap to expand)
What
Get Brahma from ~672 on hand down to a healthy 210 (18 weeks of forward seasonal demand). That means selling ~462 units in 14 weeks: ~32/wk vs ~6.4/wk the season gives unaided -- a 5.02x lift at the conservative count. Verdict: BEYOND ADS ALONE -- needs a decision: deeper cut, later deadline, or accept carry.
Why
Selling through generates ~$941,399 of revenue at current sale prices. Target and deadline are benchmark-based plan settings (editable right on the Mission board), seasonally adjusted from 30 months of Shopify sales. COUNT WARNING: Shopify on-hand values at ~$1,926,590 at cost vs $900,000 in QuickBooks -- unit figures shown as QB-scaled..claimed ranges until reconciled. Source: Shopify inventory+velocity cache 2026-07-11, real unit economics where pulled.
Expected outcome
Tracking: the Inventory Mission board reads weekly -- family units/wk vs the required ~32/wk line; pace flips ON PACE / BEHIND / AHEAD (currently BEHIND). Stop or escalate at each Monday read, not mid-week.
Corpus evidence
Retail and apparel advertisers should ensure product feeds have high-quality images and accurate pricing to benefit from price-tracking and virtual try-on surfaces; confirm Google Lens eligibility by auditing image quality and structured data in Merchant Center before peak back-to-school traffic in
Prioritize Merchant Center feed hygiene (titles, images, GTIN, availability) as a first-order ecommerce ROAS lever for CCC.
Get Orpington from ~709 on hand down to a healthy 145 (18 weeks of forward seasonal demand).
Details — evidence, verification, testing window (tap to expand)
What
Get Orpington from ~709 on hand down to a healthy 145 (18 weeks of forward seasonal demand). That means selling ~564 units in 14 weeks: ~40/wk vs ~4.5/wk the season gives unaided -- a 8.86x lift at the conservative count. Verdict: BEYOND ADS ALONE -- needs a decision: deeper cut, later deadline, or accept carry.
Why
Selling through generates ~$540,687 of revenue at current sale prices. Target and deadline are benchmark-based plan settings (editable right on the Mission board), seasonally adjusted from 30 months of Shopify sales. COUNT WARNING: Shopify on-hand values at ~$1,926,590 at cost vs $900,000 in QuickBooks -- unit figures shown as QB-scaled..claimed ranges until reconciled. Source: Shopify inventory+velocity cache 2026-07-11, real unit economics where pulled.
Expected outcome
Tracking: the Inventory Mission board reads weekly -- family units/wk vs the required ~40/wk line; pace flips ON PACE / BEHIND / AHEAD (currently BEHIND). Stop or escalate at each Monday read, not mid-week.
Corpus evidence
Retail and apparel advertisers should ensure product feeds have high-quality images and accurate pricing to benefit from price-tracking and virtual try-on surfaces; confirm Google Lens eligibility by auditing image quality and structured data in Merchant Center before peak back-to-school traffic in
Prioritize Merchant Center feed hygiene (titles, images, GTIN, availability) as a first-order ecommerce ROAS lever for CCC.
More recommendations (39) — work the top 3 first; open these once those are executed, snoozed, or declined
Get Leghorn from ~633 on hand down to a healthy 112 (18 weeks of forward seasonal demand).
Details — evidence, verification, testing window (tap to expand)
What
Get Leghorn from ~633 on hand down to a healthy 112 (18 weeks of forward seasonal demand). That means selling ~521 units in 14 weeks: ~36/wk vs ~3.4/wk the season gives unaided -- a 10.63x lift at the conservative count. Verdict: BEYOND ADS ALONE -- needs a decision: deeper cut, later deadline, or accept carry.
Why
Selling through generates ~$318,341 of revenue at current sale prices. Target and deadline are benchmark-based plan settings (editable right on the Mission board), seasonally adjusted from 30 months of Shopify sales. COUNT WARNING: Shopify on-hand values at ~$1,926,590 at cost vs $900,000 in QuickBooks -- unit figures shown as QB-scaled..claimed ranges until reconciled. Source: Shopify inventory+velocity cache 2026-07-11, real unit economics where pulled.
Expected outcome
Tracking: the Inventory Mission board reads weekly -- family units/wk vs the required ~36/wk line; pace flips ON PACE / BEHIND / AHEAD (currently BEHIND). Stop or escalate at each Monday read, not mid-week.
Corpus evidence
Retail and apparel advertisers should ensure product feeds have high-quality images and accurate pricing to benefit from price-tracking and virtual try-on surfaces; confirm Google Lens eligibility by auditing image quality and structured data in Merchant Center before peak back-to-school traffic in
Prioritize Merchant Center feed hygiene (titles, images, GTIN, availability) as a first-order ecommerce ROAS lever for CCC.
Attach the existing shared negative keyword lists (7 lists, 3,344 terms) to (NB) Chicken Coops.
Details — evidence, verification, testing window (tap to expand)
What
Attach the existing shared negative keyword lists (7 lists, 3,344 terms) to (NB) Chicken Coops.
Why
(NB) Chicken Coops spent $1,134.17 in the last 30 days with zero shared negative lists attached, while 3,344 already-built negative terms sit on other campaigns. Sources: Bing Campaign Management API (settings) + Bing Reporting API (30d spend).
Expected outcome
Reduces irrelevant matched traffic without new spend. Uses lists already written and in use, so no new negatives need authoring.
Corpus evidence
Paid media managers running Microsoft Search campaigns should immediately audit AI Max settings on existing campaigns to determine if URL expansion and broad match-style query expansion are enabled by default, and apply negative keywords or opt-out controls where brand safety or budget efficiency is
Attach the existing shared negative keyword lists (9 lists, 3,374 terms) to (2) Demand Gen - Video Placement.
Details — evidence, verification, testing window (tap to expand)
What
Attach the existing shared negative keyword lists (9 lists, 3,374 terms) to (2) Demand Gen - Video Placement.
Why
(2) Demand Gen - Video Placement spent $716.45 in the last 30 days with zero shared negative lists attached, while 3,374 already-built negative terms sit on other campaigns. Sources: Bing Campaign Management API (settings) + Bing Reporting API (30d spend).
Expected outcome
Reduces irrelevant matched traffic without new spend. Uses lists already written and in use, so no new negatives need authoring.
Corpus evidence
When a Budget & Bid Strategist agent recommends blocking a wasteful search term inside a PMax campaign, it must route the negative to an account-level list, not attempt a campaign-level negative -- the latter silently fails to apply the way it would on a Search campaign.
For any PMax campaign, run a brand leak audit by pulling search term reports (via Insights or Search Terms where available), isolating branded query spend, and comparing ROAS with vs. without brand traffic; if brand spend is material, add brand exclusion lists or a dedicated brand campaign with PMax
Add brand and irrelevant-term exclusions to (1) PMax - Shopping Only - Coops & Runs.
Details — evidence, verification, testing window (tap to expand)
What
Add brand and irrelevant-term exclusions to (1) PMax - Shopping Only - Coops & Runs.
Why
(1) PMax - Shopping Only - Coops & Runs (PerformanceMax) spent $571.37 in 30 days with zero campaign negatives and zero shared lists attached. Sources: Bing Campaign Management API + Bing Reporting API.
Expected outcome
Stops spend on queries the campaign should not serve.
Corpus evidence
Paid media managers running Microsoft Search campaigns should immediately audit AI Max settings on existing campaigns to determine if URL expansion and broad match-style query expansion are enabled by default, and apply negative keywords or opt-out controls where brand safety or budget efficiency is
Extend the winner: duplicate this ad into adjacent ad sets/audiences and brief the next creative batch on its angle.
Details — evidence, verification, testing window (tap to expand)
What
Extend the winner: duplicate this ad into adjacent ad sets/audiences and brief the next creative batch on its angle. Do NOT edit the original.
Why
CTR 7.389564 is >=1.5x the campaign median (4.662) with 6 purchases on $270.68 spend, BOFU stage. Source: Meta ad-level insights vs campaign median, 2026-08-23.
Expected outcome
The proven angle earns more reach; watch for fatigue at the stage frequency bar.
Corpus evidence
For clients running Meta Ads in dense urban or highly competitive local markets, audit current targeting, bidding, and creative approaches against competitive-market benchmarks; generic national or suburban playbooks likely underperform in these environments.
Default CCC/ecommerce Meta prospecting to ASC; keep a small manual/testing campaign to graduate winning creative in.
'(1) PMax - Shopping Only - Low Priced Coops' (google) is profitable but under its 2025 good-year bar by 2.9x.
Details — evidence, verification, testing window (tap to expand)
What
'(1) PMax - Shopping Only - Low Priced Coops' (google) is profitable but under its 2025 good-year bar by 2.9x. Diagnose (search terms, creative age, bid strategy) before adding or cutting budget.
Why
Verified ROAS 2.88x is above break-even but below the channel's own funnel-adjusted 2025 median -- history proves this campaign class can do better. Sources: platform API + ThoughtMetric, 2026-08-23. COORDINATE: Intero (agency) (cole.johanson@interodigital.com) changed maximize_conversion_value.target_roas on this campaign 2026-08-14 -- align before overlapping.
Expected outcome
Closing the gap returns ~$2,040/mo additional contribution at current spend if ROAS recovers to the median.
Corpus evidence
When a Budget & Bid Strategist agent recommends blocking a wasteful search term inside a PMax campaign, it must route the negative to an account-level list, not attempt a campaign-level negative -- the latter silently fails to apply the way it would on a Search campaign.
When auditing Google Ads accounts, use Quality Score sub-scores to pinpoint which of the three components is underperforming and prioritize fixes there (e.g., rewrite ad copy for relevance, improve page load and message match on landing pages) rather than chasing the score number itself.
This campaign positively targets Georgia -- almost certainly a picker mistake (the US state was already targeted).
Details — evidence, verification, testing window (tap to expand)
What
This campaign positively targets Georgia -- almost certainly a picker mistake (the US state was already targeted). A freight-heavy US-only product should not serve ads there.
Why
campaign_criterion LOCATION audit, live 2026-08-23. Positive geo targets are OR'd, so this actively expands serving abroad.
Expected outcome
Zero foreign impressions; no US impact -- the United States country target is unchanged.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
This campaign positively targets Georgia -- almost certainly a picker mistake (the US state was already targeted).
Details — evidence, verification, testing window (tap to expand)
What
This campaign positively targets Georgia -- almost certainly a picker mistake (the US state was already targeted). A freight-heavy US-only product should not serve ads there.
Why
campaign_criterion LOCATION audit, live 2026-08-23. Positive geo targets are OR'd, so this actively expands serving abroad. COORDINATE: Intero (agency) (cole.johanson@interodigital.com) changed maximize_conversion_value.target_roas on this campaign 2026-08-14 -- align before overlapping.
Expected outcome
Zero foreign impressions; no US impact -- the United States country target is unchanged.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Campaigns carry 38-63 negative zip exclusions each, in structured patterns (zips positive on one campaign are negative on another).
Details — evidence, verification, testing window (tap to expand)
What
Campaigns carry 38-63 negative zip exclusions each, in structured patterns (zips positive on one campaign are negative on another). Nobody has claimed them; they silently block whatever demand lives in those zips.
Why
campaign_criterion LOCATION audit, live 2026-08-23. Actor unknown: zero Google location-criterion records in the trailing-30d API change history; all 73 stored geo changes are Meta ad-set edits.
Expected outcome
Either a documented reason for each list, or removal -- then watch zip-adjacent state revenue at the next weekly read.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Dead creative still enabled.
Details — evidence, verification, testing window (tap to expand)
What
Dead creative still enabled.
Why
Meta ad-level insights 2026-08-23: 0.0% CTR across the window.
Expected outcome
Cleaner ad-set reporting, no spend impact.
Corpus evidence
For clients running Meta Ads in dense urban or highly competitive local markets, audit current targeting, bidding, and creative approaches against competitive-market benchmarks; generic national or suburban playbooks likely underperform in these environments.
Default CCC/ecommerce Meta prospecting to ASC; keep a small manual/testing campaign to graduate winning creative in.
'(NB) Chicken Coops' (bing) is profitable but under its 2025 good-year bar by 1.4x.
Details — evidence, verification, testing window (tap to expand)
What
'(NB) Chicken Coops' (bing) is profitable but under its 2025 good-year bar by 1.4x. Diagnose (search terms, creative age, bid strategy) before adding or cutting budget.
Why
Verified ROAS 4.72x is above break-even but below the channel's own funnel-adjusted 2025 median -- history proves this campaign class can do better. Sources: platform API + ThoughtMetric, 2026-08-23. COORDINATE: Other/unknown (Administrator) changed Budget Paused -> Not paused; via Other on this campaign 2026-08-23 -- align before overlapping.
Expected outcome
Closing the gap returns ~$943/mo additional contribution at current spend if ROAS recovers to the median.
Corpus evidence
Any new Microsoft Ads campaigns created on or after Oct. 1 must use automated bidding; audit all planned campaign builds before that date and identify any workflows or templates that default to Max CPC so they can be updated to an approved automated strategy (e.g., Target CPA or Target ROAS) before
Before October 1, audit any Microsoft Advertising campaigns that rely on standalone automated bidding strategies with Max CPC caps; document current CPC ceilings and assess budget exposure, since removal of Max CPC means automated strategies will have unconstrained per-click spending—consider settin
'(1) PMax - Shopping Only - Coops & Runs' (bing) is profitable but under its 2025 good-year bar by 2.0x.
Details — evidence, verification, testing window (tap to expand)
What
'(1) PMax - Shopping Only - Coops & Runs' (bing) is profitable but under its 2025 good-year bar by 2.0x. Diagnose (search terms, creative age, bid strategy) before adding or cutting budget.
Why
Verified ROAS 4.11x is above break-even but below the channel's own funnel-adjusted 2025 median -- history proves this campaign class can do better. Sources: platform API + ThoughtMetric, 2026-08-23. COORDINATE: Intero (agency) (cole.johanson@interodigital.com) changed maximize_conversion_value.target_roas on this campaign 2026-08-14 -- align before overlapping.
Expected outcome
Closing the gap returns ~$587/mo additional contribution at current spend if ROAS recovers to the median.
Corpus evidence
Any new Microsoft Ads campaigns created on or after Oct. 1 must use automated bidding; audit all planned campaign builds before that date and identify any workflows or templates that default to Max CPC so they can be updated to an approved automated strategy (e.g., Target CPA or Target ROAS) before
Before October 1, audit any Microsoft Advertising campaigns that rely on standalone automated bidding strategies with Max CPC caps; document current CPC ceilings and assess budget exposure, since removal of Max CPC means automated strategies will have unconstrained per-click spending—consider settin
'(1) PMax - Shopping Only - Coops & Runs' (google) is profitable but under its 2025 good-year bar by 0.4x.
Details — evidence, verification, testing window (tap to expand)
What
'(1) PMax - Shopping Only - Coops & Runs' (google) is profitable but under its 2025 good-year bar by 0.4x. Diagnose (search terms, creative age, bid strategy) before adding or cutting budget.
Why
Verified ROAS 5.39x is above break-even but below the channel's own funnel-adjusted 2025 median -- history proves this campaign class can do better. Sources: platform API + ThoughtMetric, 2026-08-23. COORDINATE: Intero (agency) (cole.johanson@interodigital.com) changed maximize_conversion_value.target_roas on this campaign 2026-08-14 -- align before overlapping.
Expected outcome
Closing the gap returns ~$1,546/mo additional contribution at current spend if ROAS recovers to the median.
Corpus evidence
When a Budget & Bid Strategist agent recommends blocking a wasteful search term inside a PMax campaign, it must route the negative to an account-level list, not attempt a campaign-level negative -- the latter silently fails to apply the way it would on a Search campaign.
When auditing Google Ads accounts, use Quality Score sub-scores to pinpoint which of the three components is underperforming and prioritize fixes there (e.g., rewrite ad copy for relevance, improve page load and message match on landing pages) rather than chasing the score number itself.
23.2% of this PMax's conversion value comes from BRAND search terms it did not create.
Details — evidence, verification, testing window (tap to expand)
What
23.2% of this PMax's conversion value comes from BRAND search terms it did not create. Apply a brand exclusion list (Google support form / brand lists) so PMax earns non-brand demand; branded Search already captures brand at 15.9x verified.
Why
campaign_search_term_insight brand-category share over ['2026-05-25', '2026-08-23']. Source: Google Ads API, Google-attributed. COORDINATE: Intero (agency) (cole.johanson@interodigital.com) changed maximize_conversion_value.target_roas on this campaign 2026-08-14 -- align before overlapping.
Expected outcome
PMax reported ROAS drops (honest), true incremental non-brand contribution becomes measurable.
Corpus evidence
For CCC or any PMax-driven client, check conversion volume per campaign before recommending a new segmented campaign. If splitting would drop any resulting campaign below ~30 conv/month, recommend consolidation instead -- directly relevant to the CCC deep-clearance-lane idea, which should live insid
For any PMax campaign, run a brand leak audit by pulling search term reports (via Insights or Search Terms where available), isolating branded query spend, and comparing ROAS with vs. without brand traffic; if brand spend is material, add brand exclusion lists or a dedicated brand campaign with PMax
'(B) Branded' (google) is profitable but under its 2025 good-year bar by 0.2x.
Details — evidence, verification, testing window (tap to expand)
What
'(B) Branded' (google) is profitable but under its 2025 good-year bar by 0.2x. Diagnose (search terms, creative age, bid strategy) before adding or cutting budget.
Why
Verified ROAS 7.58x is above break-even but below the channel's own funnel-adjusted 2025 median -- history proves this campaign class can do better. Sources: platform API + ThoughtMetric, 2026-08-23.
Expected outcome
Closing the gap returns ~$93/mo additional contribution at current spend if ROAS recovers to the median.
Corpus evidence
When a Budget & Bid Strategist agent recommends blocking a wasteful search term inside a PMax campaign, it must route the negative to an account-level list, not attempt a campaign-level negative -- the latter silently fails to apply the way it would on a Search campaign.
When auditing Google Ads accounts, use Quality Score sub-scores to pinpoint which of the three components is underperforming and prioritize fixes there (e.g., rewrite ad copy for relevance, improve page load and message match on landing pages) rather than chasing the score number itself.
Completes the ladder: current vs LaL vs broad.
Details — evidence, verification, testing window (tap to expand)
What
Completes the ladder: current vs LaL vs broad.
Why
Platform best practice in the knowledge corpus favors broad on mature pixels, but this account has never measured it -- account memory requires a measured window, not doctrine.
Expected outcome
A ranked audience playbook for this account grounded in its own measured CPA.
Corpus evidence
Start Meta prospecting broad; use purchaser-list lookalikes as a soft signal and let creative do the qualifying.
Gives the Meta Analyst agent a concrete pre-check before recommending a client move budget into ASC: verify catalog size, weekly purchase volume, and daily budget against these floors first, rather than defaulting to 'Advantage+ is the 2026 standard.'
Decision for Chris/Danny, not an automated call: reconcile what Intero bills for against the change logs in this dashboard, and set expectations or re.
Details — evidence, verification, testing window (tap to expand)
What
Decision for Chris/Danny, not an automated call: reconcile what Intero bills for against the change logs in this dashboard, and set expectations or restructure.
Why
Meta account audit log (Meta's own authoritative record, 365 days): Intero made 4 changes in ALL of 2026 vs 1850 in Jul-Dec 2025. Google (API caps at 30 days): 1064 events, 182 of them on a single day (bulk asset upload). Last 90 days across both: 1 Meta events. Sources: Meta /activities + Google change_event, pulled 2026-08-23. CAVEAT: Google's full-year history needs a manual UI export (2-yr history is not API-accessible); unresolved actor names (Ashley King, Sierra Liegler) may be agency staff -- if so, agency Meta activity extends to May 2026, then still stops.
Expected outcome
Either restored agency output or recovered fees; both show up as change-log volume and G2/G4 movement.
Corpus evidence
Agency and in-house marketers should anticipate continued client-side budget and headcount pressure at big tech; pitch lean, AI-augmented retainers and prioritize demonstrable ROI reporting rather than expanding scope assumptions.
Prospecting rides the proven creative (CTR 5.257%, $371/28d) into a colder, larger pool.
Details — evidence, verification, testing window (tap to expand)
What
Prospecting rides the proven creative (CTR 5.257%, $371/28d) into a colder, larger pool.
Why
Meta ad-level insights 2026-08-23: 'Store collection - Best Coops - April 20' is the strongest FRESH performer; audience_ops is the last PARTIAL takeover category and the account has no recent operator-run audience experiment in the change history.
Expected outcome
A read on whether LaL-3% beats current targeting on CPA within 14 days at <=$350 total risk.
Corpus evidence
Start Meta prospecting broad; use purchaser-list lookalikes as a soft signal and let creative do the qualifying.
Gives the Meta Analyst agent a concrete pre-check before recommending a client move budget into ASC: verify catalog size, weekly purchase volume, and daily budget against these floors first, rather than defaulting to 'Advantage+ is the 2026 standard.'
Formally pause the two boosted-post ad sets flagged in the 2026-07-18 audit (roadmap item 1).
Details — evidence, verification, testing window (tap to expand)
What
Formally pause the two boosted-post ad sets flagged in the 2026-07-18 audit (roadmap item 1). They optimize for Landing Page Views and were never built to drive sales.
Why
Audit 2026-07-18, Part 4 item 1. They already triggered the false 'losing money' alert once; paused-for-real removes the risk of silent resumption.
Expected outcome
Zero further spend possible on non-Purchase boosted posts.
Google's own open recommendation on both PMax campaigns -- free ad real estate, no downside.
Details — evidence, verification, testing window (tap to expand)
What
Google's own open recommendation on both PMax campaigns -- free ad real estate, no downside.
Why
Audit 2026-07-18, Part 3 (34 open Google recommendations) and Part 4 item 3.
Expected outcome
Higher CTR on PMax Search-surface placements at zero added spend.
Close the confirmed creative gap on the top non-brand campaign -- the one with the verified YoY efficiency decline.
Details — evidence, verification, testing window (tap to expand)
What
Close the confirmed creative gap on the top non-brand campaign -- the one with the verified YoY efficiency decline.
Why
Audit 2026-07-18, Part 3 + Part 4 item 4; Google change history 2026-07-16 (Intero's RSA swap, values recovered 2026-07-19).
Expected outcome
Better Ad Strength and CTR on the account's most important non-brand campaign.
Stop paying for competitor-brand searches that never convert.
Details — evidence, verification, testing window (tap to expand)
What
Stop paying for competitor-brand searches that never convert.
Why
Audit 2026-07-18, Part 3: ~$6,675/90d on zero-conversion terms, competitor brands bleeding into Branded specifically.
Expected outcome
A measurable share of the $6.7K/90d zero-conversion spend eliminated.
Replace the assets Microsoft rates Low in (NB) Chicken Coops / Chicken Coop: 'Featured In Leading Magazines'; 'Engineered For Strength'; 'Excellent Cu.
Details — evidence, verification, testing window (tap to expand)
What
Replace the assets Microsoft rates Low in (NB) Chicken Coops / Chicken Coop: 'Featured In Leading Magazines'; 'Engineered For Strength'; 'Excellent Customer Service'; 'Simple & Easy Assembly'; 'Accessories Also Available'; 'The Chicken Coop Company's Coops Are Sturdy & Designed To Be Easy to Clean & Maintain.'...
Why
Microsoft's own AssetPerformanceLabel marks these 9 assets Low on actively serving ads. Source: Bing Campaign Management API, ad asset labels.
Expected outcome
Raises the share of impressions served by Best/Good assets. Reversible -- the originals can be restored.
Corpus evidence
Real estate lead-gen campaigns should test messaging that foregrounds potential rate stability as a buying opportunity; update ad copy and landing page headlines within 1-2 weeks while this macro narrative is active in the news cycle.
Real estate agents running buyer or seller campaigns should avoid messaging that equates rising inventory with a fully balanced or buyer-friendly market; instead, frame listings around the 4.6-month supply figure to set accurate client expectations and differentiate ad copy by local demand signals r
Do NOT exclude Oklahoma yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Details — evidence, verification, testing window (tap to expand)
What
Do NOT exclude Oklahoma yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern. Platform-reported ROAS 0.0x on $449 spent (90d), below the 2.232x breakeven floor. Re-check at the next quarterly read.
Why
Oklahoma: $449 spend, 0.0 conversions, $0 attributed value over the 90-day window (Google geographic_view, platform-reported, 2026-08-23). not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Expected outcome
No action -- monitoring only until the longer-window read confirms or clears it.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Do NOT exclude South Dakota yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Details — evidence, verification, testing window (tap to expand)
What
Do NOT exclude South Dakota yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern. Platform-reported ROAS 0.22x on $136 spent (90d), below the 2.232x breakeven floor. Re-check at the next quarterly read.
Why
South Dakota: $136 spend, 1.0 conversions, $30 attributed value over the 90-day window (Google geographic_view, platform-reported, 2026-08-23). not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Expected outcome
No action -- monitoring only until the longer-window read confirms or clears it.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Do NOT exclude Rhode Island yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Details — evidence, verification, testing window (tap to expand)
What
Do NOT exclude Rhode Island yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern. Platform-reported ROAS 0.0x on $188 spent (90d), below the 2.232x breakeven floor. Re-check at the next quarterly read.
Why
Rhode Island: $188 spend, 0.0 conversions, $0 attributed value over the 90-day window (Google geographic_view, platform-reported, 2026-08-23). not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Expected outcome
No action -- monitoring only until the longer-window read confirms or clears it.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Do NOT exclude Wyoming yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Details — evidence, verification, testing window (tap to expand)
What
Do NOT exclude Wyoming yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern. Platform-reported ROAS 0.0x on $225 spent (90d), below the 2.232x breakeven floor. Re-check at the next quarterly read.
Why
Wyoming: $225 spend, 0.0 conversions, $0 attributed value over the 90-day window (Google geographic_view, platform-reported, 2026-08-23). not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Expected outcome
No action -- monitoring only until the longer-window read confirms or clears it.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Do NOT exclude Idaho yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Details — evidence, verification, testing window (tap to expand)
What
Do NOT exclude Idaho yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern. Platform-reported ROAS 1.15x on $430 spent (90d), below the 2.232x breakeven floor. Re-check at the next quarterly read.
Why
Idaho: $430 spend, 2.0 conversions, $493 attributed value over the 90-day window (Google geographic_view, platform-reported, 2026-08-23). not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Expected outcome
No action -- monitoring only until the longer-window read confirms or clears it.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Do NOT exclude Arizona yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Details — evidence, verification, testing window (tap to expand)
What
Do NOT exclude Arizona yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern. Platform-reported ROAS 1.56x on $682 spent (90d), below the 2.232x breakeven floor. Re-check at the next quarterly read.
Why
Arizona: $682 spend, 4.0 conversions, $1,065 attributed value over the 90-day window (Google geographic_view, platform-reported, 2026-08-23). not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Expected outcome
No action -- monitoring only until the longer-window read confirms or clears it.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Do NOT exclude West Virginia yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Details — evidence, verification, testing window (tap to expand)
What
Do NOT exclude West Virginia yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern. Platform-reported ROAS 0.0x on $267 spent (90d), below the 2.232x breakeven floor. Re-check at the next quarterly read.
Why
West Virginia: $267 spend, 0.0 conversions, $0 attributed value over the 90-day window (Google geographic_view, platform-reported, 2026-08-23). not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Expected outcome
No action -- monitoring only until the longer-window read confirms or clears it.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Do NOT exclude Alabama yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Details — evidence, verification, testing window (tap to expand)
What
Do NOT exclude Alabama yet -- not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern. Platform-reported ROAS 0.36x on $502 spent (90d), below the 2.232x breakeven floor. Re-check at the next quarterly read.
Why
Alabama: $502 spend, 2.0 conversions, $182 attributed value over the 90-day window (Google geographic_view, platform-reported, 2026-08-23). not below breakeven over the trailing ~365 days -- this looks like a seasonal window, not a pattern.
Expected outcome
No action -- monitoring only until the longer-window read confirms or clears it.
Corpus evidence
When clients suspect click fraud, first audit location targeting settings (especially 'people showing interest in' settings), review placement reports for unfamiliar domains, and evaluate ad creative for accidental click risk before pursuing fraud-specific solutions or platform refund requests.
Review 2 paused ad group(s) holding 44 keywords: Competitor (41 kws), Competitor - Lowes (3 kws).
Details — evidence, verification, testing window (tap to expand)
What
Review 2 paused ad group(s) holding 44 keywords: Competitor (41 kws), Competitor - Lowes (3 kws)
Why
These keywords cannot serve while their ad group is paused. No spend is at risk; this usually indicates a restructure that was left unfinished. Source: Bing Campaign Management API, ad group status.
Expected outcome
Either re-activate the intended structure or retire it.
Corpus evidence
Audit CCC/GSME campaigns for over-segmentation; merge thin ad groups so each campaign clears the volume Smart Bidding needs to learn.
Review 3 paused ad group(s) holding 52 keywords: Chicken Coop 5+ (18 kws), Large Chicken Coop (16 kws), Chicken Coop 10+ (18 kws).
Details — evidence, verification, testing window (tap to expand)
What
Review 3 paused ad group(s) holding 52 keywords: Chicken Coop 5+ (18 kws), Large Chicken Coop (16 kws), Chicken Coop 10+ (18 kws)
Why
These keywords cannot serve while their ad group is paused. No spend is at risk; this usually indicates a restructure that was left unfinished. Source: Bing Campaign Management API, ad group status.
Expected outcome
Either re-activate the intended structure or retire it.
Corpus evidence
Audit CCC/GSME campaigns for over-segmentation; merge thin ad groups so each campaign clears the volume Smart Bidding needs to learn.
Review 25 paused ad group(s) holding 392 keywords: Chicken Pens (5 kws), Chicken Coops Small (9 kws), Chicken Coops Urban (12 kws), Chicken Coop Price.
Details — evidence, verification, testing window (tap to expand)
What
Review 25 paused ad group(s) holding 392 keywords: Chicken Pens (5 kws), Chicken Coops Small (9 kws), Chicken Coops Urban (12 kws), Chicken Coop Prices (5 kws)
Why
These keywords cannot serve while their ad group is paused. No spend is at risk; this usually indicates a restructure that was left unfinished. Source: Bing Campaign Management API, ad group status.
Expected outcome
Either re-activate the intended structure or retire it.
Corpus evidence
Practitioners managing LSA accounts for local service businesses should audit their current LSA setup and monitor official Google Ads announcements for specifics on the PMax integration, preparing to restructure campaign management workflows if LSAs shift under the PMax umbrella.
Audit CCC/GSME campaigns for over-segmentation; merge thin ad groups so each campaign clears the volume Smart Bidding needs to learn.
Review 2 paused ad group(s) holding 14 keywords: Chicken Coops Home (2 kws), Chicken Coop 10 (12 kws).
Details — evidence, verification, testing window (tap to expand)
What
Review 2 paused ad group(s) holding 14 keywords: Chicken Coops Home (2 kws), Chicken Coop 10 (12 kws)
Why
These keywords cannot serve while their ad group is paused. No spend is at risk; this usually indicates a restructure that was left unfinished. Source: Bing Campaign Management API, ad group status.
Expected outcome
Either re-activate the intended structure or retire it.
Corpus evidence
Practitioners managing LSA accounts for local service businesses should audit their current LSA setup and monitor official Google Ads announcements for specifics on the PMax integration, preparing to restructure campaign management workflows if LSAs shift under the PMax umbrella.
Audit CCC/GSME campaigns for over-segmentation; merge thin ad groups so each campaign clears the volume Smart Bidding needs to learn.
Review 2 paused ad group(s) holding 44 keywords: Competitor (41 kws), Competitor - Lowes (3 kws).
Details — evidence, verification, testing window (tap to expand)
What
Review 2 paused ad group(s) holding 44 keywords: Competitor (41 kws), Competitor - Lowes (3 kws)
Why
These keywords cannot serve while their ad group is paused. No spend is at risk; this usually indicates a restructure that was left unfinished. Source: Bing Campaign Management API, ad group status.
Expected outcome
Either re-activate the intended structure or retire it.
Corpus evidence
Practitioners managing LSA accounts for local service businesses should audit their current LSA setup and monitor official Google Ads announcements for specifics on the PMax integration, preparing to restructure campaign management workflows if LSAs shift under the PMax umbrella.
Audit CCC/GSME campaigns for over-segmentation; merge thin ad groups so each campaign clears the volume Smart Bidding needs to learn.
Review 3 paused ad group(s) holding 82 keywords: Branded (36 kws), Competitor (42 kws), Competitor - Lowes (4 kws).
Details — evidence, verification, testing window (tap to expand)
What
Review 3 paused ad group(s) holding 82 keywords: Branded (36 kws), Competitor (42 kws), Competitor - Lowes (4 kws)
Why
These keywords cannot serve while their ad group is paused. No spend is at risk; this usually indicates a restructure that was left unfinished. Source: Bing Campaign Management API, ad group status.
Expected outcome
Either re-activate the intended structure or retire it.
Corpus evidence
Practitioners managing LSA accounts for local service businesses should audit their current LSA setup and monitor official Google Ads announcements for specifics on the PMax integration, preparing to restructure campaign management workflows if LSAs shift under the PMax umbrella.
Audit CCC/GSME campaigns for over-segmentation; merge thin ad groups so each campaign clears the volume Smart Bidding needs to learn.
Review 1 paused ad group(s) holding 11 keywords: Pre-Fabricated (11 kws).
Details — evidence, verification, testing window (tap to expand)
What
Review 1 paused ad group(s) holding 11 keywords: Pre-Fabricated (11 kws)
Why
These keywords cannot serve while their ad group is paused. No spend is at risk; this usually indicates a restructure that was left unfinished. Source: Bing Campaign Management API, ad group status.
Expected outcome
Either re-activate the intended structure or retire it.
Corpus evidence
Practitioners managing LSA accounts for local service businesses should audit their current LSA setup and monitor official Google Ads announcements for specifics on the PMax integration, preparing to restructure campaign management workflows if LSAs shift under the PMax umbrella.
Audit CCC/GSME campaigns for over-segmentation; merge thin ad groups so each campaign clears the volume Smart Bidding needs to learn.
Review 2 paused ad group(s) holding 44 keywords: Competitor (41 kws), Competitor - Lowes (3 kws).
Details — evidence, verification, testing window (tap to expand)
What
Review 2 paused ad group(s) holding 44 keywords: Competitor (41 kws), Competitor - Lowes (3 kws)
Why
These keywords cannot serve while their ad group is paused. No spend is at risk; this usually indicates a restructure that was left unfinished. Source: Bing Campaign Management API, ad group status.
Expected outcome
Either re-activate the intended structure or retire it.
Corpus evidence
Practitioners managing LSA accounts for local service businesses should audit their current LSA setup and monitor official Google Ads announcements for specifics on the PMax integration, preparing to restructure campaign management workflows if LSAs shift under the PMax umbrella.
Audit CCC/GSME campaigns for over-segmentation; merge thin ad groups so each campaign clears the volume Smart Bidding needs to learn.
Housekeeping — 2 money moves under $1,000/mo (~$508/mo combined — batch on a slow day, never the headline)
Raise '(NB) Chicken Coops' (google) from ~$132/day to $151/day (+15%).
Details — evidence, verification, testing window (tap to expand)
What
Raise '(NB) Chicken Coops' (google) from ~$132/day to $151/day (+15%). Half-step (+15%, not the usual +30%) into soft demand -- the full step is earned by a held verdict, not assumed. Hold 72h, re-check verified ROAS at the 7-day read, and step again while it holds at/above the good bar. This is where the needle moves -- not the trims. SEASONALITY: demand ahead is soft (next-month index 0.6x annual mean); hold each step 7 days and size against next month's demand, not the trailing 90 -- do not let PMax pace into weak weeks.
Why
Verified 6.39x on $11,853 in the 90-day window -- at/above its funnel-adjusted good bar with proven demand. Sized with a marginal-ROAS haircut to 5.1x (new impressions never match the core). Sources: platform API + ThoughtMetric, 2026-08-23. Last year's same window -- 2025-09: 5.43x on $10,849 spend; 2025-10: 4.46x on $11,411 spend (ThoughtMetric monthly). The trough cuts VOLUME; this is the check on whether it also cut efficiency. COORDINATE: Other/unknown (Administrator) changed Budget Paused -> Not paused; via Other on this campaign 2026-08-23 -- align before overlapping.
Expected outcome
Seasonally adjusted for next month's 0.60x demand index: ~+$459/mo contribution (unadjusted ~+$765/mo at trailing-90 demand). ~+$3,030/mo revenue before adjustment. Direct measurement: verified channel ROAS + revenue at +7/+30d in this dashboard.
Corpus evidence
For CCC or any PMax-driven client, check conversion volume per campaign before recommending a new segmented campaign. If splitting would drop any resulting campaign below ~30 conv/month, recommend consolidation instead -- directly relevant to the CCC deep-clearance-lane idea, which should live insid
When auditing Google Ads accounts, consolidate fragmented campaigns to meet minimum 30 conversions/month per campaign, apply brand exclusions to PMax to prevent branded traffic cannibalization, and review PMax Listing Groups to verify budget distribution across the full product catalog rather than j
Raise '(2) Demand Gen - Video Placement' (google) from ~$25/day to $29/day (+15%).
Details — evidence, verification, testing window (tap to expand)
What
Raise '(2) Demand Gen - Video Placement' (google) from ~$25/day to $29/day (+15%). Half-step (+15%, not the usual +30%) into soft demand -- the full step is earned by a held verdict, not assumed. Hold 72h, re-check verified ROAS at the 7-day read, and step again while it holds at/above the good bar. This is where the needle moves -- not the trims. SEASONALITY: demand ahead is soft (next-month index 0.6x annual mean); hold each step 7 days and size against next month's demand, not the trailing 90 -- do not let PMax pace into weak weeks.
Why
Verified 4.81x on $2,270 in the 90-day window -- at/above its funnel-adjusted good bar with proven demand. Sized with a marginal-ROAS haircut to 3.8x (new impressions never match the core). Sources: platform API + ThoughtMetric, 2026-08-23. Last year's same window -- 2025-09: 5.43x on $10,849 spend; 2025-10: 4.46x on $11,411 spend (ThoughtMetric monthly). The trough cuts VOLUME; this is the check on whether it also cut efficiency.
Expected outcome
Seasonally adjusted for next month's 0.60x demand index: ~+$49/mo contribution (unadjusted ~+$82/mo at trailing-90 demand). ~+$437/mo revenue before adjustment. Direct measurement: verified channel ROAS + revenue at +7/+30d in this dashboard.
Corpus evidence
For CCC or any PMax-driven client, check conversion volume per campaign before recommending a new segmented campaign. If splitting would drop any resulting campaign below ~30 conv/month, recommend consolidation instead -- directly relevant to the CCC deep-clearance-lane idea, which should live insid
When auditing Google Ads accounts, consolidate fragmented campaigns to meet minimum 30 conversions/month per campaign, apply brand exclusions to PMax to prevent branded traffic cannibalization, and review PMax Listing Groups to verify budget distribution across the full product catalog rather than j
Recently resolved (6)
Raise '(B) Branded' (google) from ~$38/day to $44/day (+15%).
Details — evidence, verification, testing window (tap to expand)
What
Raise '(B) Branded' (google) from ~$38/day to $44/day (+15%). Half-step (+15%, not the usual +30%) into soft demand -- the full step is earned by a held verdict, not assumed. Hold 72h, re-check verified ROAS at the 7-day read, and step again while it holds at/above the good bar. This is where the needle moves -- not the trims. SEASONALITY: demand ahead is soft (next-month index 0.6x annual mean); hold each step 7 days and size against next month's demand, not the trailing 90 -- do not let PMax pace into weak weeks.
Why
Verified 7.86x on $3,434 in the 90-day window -- at/above its funnel-adjusted good bar with proven demand. Sized with a marginal-ROAS haircut to 6.3x (new impressions never match the core). Sources: platform API + ThoughtMetric, 2026-08-22. Last year's same window -- 2025-09: 5.43x on $10,849 spend; 2025-10: 4.46x on $11,411 spend (ThoughtMetric monthly). The trough cuts VOLUME; this is the check on whether it also cut efficiency.
Expected outcome
Seasonally adjusted for next month's 0.60x demand index: ~+$187/mo contribution (unadjusted ~+$312/mo at trailing-90 demand). ~+$1,080/mo revenue before adjustment. Direct measurement: verified channel ROAS + revenue at +7/+30d in this dashboard.
Corpus evidence
For CCC or any PMax-driven client, check conversion volume per campaign before recommending a new segmented campaign. If splitting would drop any resulting campaign below ~30 conv/month, recommend consolidation instead -- directly relevant to the CCC deep-clearance-lane idea, which should live insid
When auditing Google Ads accounts, consolidate fragmented campaigns to meet minimum 30 conversions/month per campaign, apply brand exclusions to PMax to prevent branded traffic cannibalization, and review PMax Listing Groups to verify budget distribution across the full product catalog rather than j
Prospecting rides the proven creative (CTR 5.004%, $379/28d) into a colder, larger pool.
Details — evidence, verification, testing window (tap to expand)
What
Prospecting rides the proven creative (CTR 5.004%, $379/28d) into a colder, larger pool.
Why
Meta ad-level insights 2026-08-22: 'Store collection - Best Coops - April 20' is the strongest FRESH performer; audience_ops is the last PARTIAL takeover category and the account has no recent operator-run audience experiment in the change history.
Expected outcome
A read on whether LaL-3% beats current targeting on CPA within 14 days at <=$350 total risk.
Corpus evidence
Start Meta prospecting broad; use purchaser-list lookalikes as a soft signal and let creative do the qualifying.
Gives the Meta Analyst agent a concrete pre-check before recommending a client move budget into ASC: verify catalog size, weekly purchase volume, and daily budget against these floors first, rather than defaulting to 'Advantage+ is the 2026 standard.'
Completes the ladder: current vs LaL vs broad.
Details — evidence, verification, testing window (tap to expand)
What
Completes the ladder: current vs LaL vs broad.
Why
Platform best practice in the knowledge corpus favors broad on mature pixels, but this account has never measured it -- account memory requires a measured window, not doctrine.
Expected outcome
A ranked audience playbook for this account grounded in its own measured CPA.
Corpus evidence
Start Meta prospecting broad; use purchaser-list lookalikes as a soft signal and let creative do the qualifying.
Gives the Meta Analyst agent a concrete pre-check before recommending a client move budget into ASC: verify catalog size, weekly purchase volume, and daily budget against these floors first, rather than defaulting to 'Advantage+ is the 2026 standard.'
Add brand and irrelevant-term exclusions to (1) PMax - Shopping Only - Coops & Runs.
Details — evidence, verification, testing window (tap to expand)
What
Add brand and irrelevant-term exclusions to (1) PMax - Shopping Only - Coops & Runs.
Why
(1) PMax - Shopping Only - Coops & Runs (PerformanceMax) spent $568.00 in 30 days with zero campaign negatives and zero shared lists attached. Sources: Bing Campaign Management API + Bing Reporting API.
Expected outcome
Stops spend on queries the campaign should not serve.
Corpus evidence
Paid media managers running Microsoft Search campaigns should immediately audit AI Max settings on existing campaigns to determine if URL expansion and broad match-style query expansion are enabled by default, and apply negative keywords or opt-out controls where brand safety or budget efficiency is
Attach the existing shared negative keyword lists (7 lists, 3,344 terms) to (NB) Chicken Coops.
Details — evidence, verification, testing window (tap to expand)
What
Attach the existing shared negative keyword lists (7 lists, 3,344 terms) to (NB) Chicken Coops.
Why
(NB) Chicken Coops spent $1,114.07 in the last 30 days with zero shared negative lists attached, while 3,344 already-built negative terms sit on other campaigns. Sources: Bing Campaign Management API (settings) + Bing Reporting API (30d spend).
Expected outcome
Reduces irrelevant matched traffic without new spend. Uses lists already written and in use, so no new negatives need authoring.
Corpus evidence
Paid media managers running Microsoft Search campaigns should immediately audit AI Max settings on existing campaigns to determine if URL expansion and broad match-style query expansion are enabled by default, and apply negative keywords or opt-out controls where brand safety or budget efficiency is
Replace the assets Microsoft rates Low in (NB) Chicken Coops / Chicken Coop: 'Featured In Leading Magazines'; 'Engineered For Strength'; 'Excellent Cu.
Details — evidence, verification, testing window (tap to expand)
What
Replace the assets Microsoft rates Low in (NB) Chicken Coops / Chicken Coop: 'Featured In Leading Magazines'; 'Engineered For Strength'; 'Excellent Customer Service'; 'Simple & Easy Assembly'; 'Accessories Also Available'; 'The Chicken Coop Company's Coops Are Sturdy & Designed To Be Easy to Clean & Maintain.'...
Why
Microsoft's own AssetPerformanceLabel marks these 9 assets Low on actively serving ads. Source: Bing Campaign Management API, ad asset labels.
Expected outcome
Raises the share of impressions served by Best/Good assets. Reversible -- the originals can be restored.
Corpus checked — no item in the learning corpus met the relevance bar for this recommendation.
In flight — are the changes working?
Every executed change with its testing window and verdict date. Ads take time to read: nothing here is a verdict until its window completes. Daily: glance for VERDICT DUE. Weekly: the Monday digest rolls these up.
No changes in measurement. Mark a recommendation done and it appears here with its verdict date.
Opportunities — think bigger
Strategist-agent proposals: structurally different plays — Meta rethought from first principles, new platforms weighed honestly. Every play carries a test budget, a verification path, a success threshold, and a kill criterion; ideas without kill criteria are not shown.
Meta Advantage+ Shopping with Overstock Catalog Priority
The play
Rebuild Meta from awareness-only into a direct-response Advantage+ Shopping (ASC) campaign fed a filtered product feed of ONLY the four overstock SKUs (Brahma, RIH, XL Brahma Run Extension, Orpington Lodge). Layer DPA retargeting on 30d site visitors against the same feed. Creative: 6 static + 3 UGC video ads pulled from existing CCC social corpus, with the 15% sale as the hook. Kill the current prospecting structure in parallel — this replaces it, not adds to it.
Why now
Meta is currently 89.05% divergent (worst of all channels) and treated as awareness only, meaning ~$8.9k/90d is likely mispriced. Inventory mission ships 2026-07-11 with 849 Brahma + 783 RIH units to move; a catalog-first structure maps 1:1 to the overstock list instead of driving generic traffic. 15% sale holds margin on every SKU, giving ASC room to clear at contribution-positive levels.
Test terms
Budget: $2,500/mo · Measured by: ThoughtMetric last-click revenue on the ASC campaign UTMs cross-checked against Shopify order tags with fb_asc source parameter; unit-level velocity delta on the 4 target SKUs from the Shopify inventory cache (independent of ad platform). · Success: Verified ROAS >= 2.23x AND >= 120 incremental units moved across the 4 overstock SKUs within the window (vs trailing 30d baseline velocity). · KILL IF: After 21 days: verified ROAS < 1.8x OR overstock unit velocity delta < +40 units vs baseline OR ThoughtMetric-to-Shopify divergence stays >40% (means we still can't trust the channel).
YouTube Shorts Demand-Gen on Coop Comparison Intent
The play
Launch a Demand Gen campaign in the existing Google Ads account using Shorts-format creative repurposed from the CCC YouTube channel. Two ad groups: (1) 'coop buyer education' targeting custom segments built from branded + non-brand coop search queries (leveraging the verified 11x non-brand Search signal), (2) product-feed-linked Shorts featuring Brahma and RIH walkthroughs with a 15% offer end-card. Route all clicks through a dedicated /clearance landing page with UTM taxonomy.
Why now
Non-brand Search runs 11x at only ~10% impression share — there is verified upper-funnel intent Google Ads is not capturing on Search alone. YouTube infrastructure already exists (channel, Shorts production, GAds account, ThoughtMetric attribution), so incremental infra cost is zero. August 17 Smart Bidding change gives us a 5-week window to baseline a new campaign before the shift, and Demand Gen inventory is priced softer than Search.
Test terms
Budget: $1,800/mo · Measured by: ThoughtMetric attributed conversions on /clearance landing page UTMs vs GA4 session-to-purchase path (GA4 unlocker required — flag: if GA4 credential remains missing, drops to single-source on ThoughtMetric only). · Success: Verified ROAS >= 2.23x on last-click AND assisted-conversion lift on branded Search of >= 8% during the flight window (measured via Google Ads brand-search query volume vs prior 30d). · KILL IF: After 28 days: verified ROAS < 1.5x AND no measurable branded search lift; OR CPM > $18 with CTR < 0.8% (signals creative-market fit failure, not bid problem).
Pinterest Catalog Shopping for Backyard-Lifestyle Overstock
The play
Stand up a Pinterest Shopping campaign using the already-integrated Windsor product feed, filtered to the 4 overstock SKUs plus 6 adjacent lifestyle products. Two ad groups: Consideration (broad backyard/homesteading interest) and Conversion (product-tagged retargeting from feed engagers). Creative uses existing lifestyle imagery, not net-new production. Pair with a Refersion affiliate push to the 94 active affiliates using the same clearance landing page — two owned surfaces, one clearance narrative.
Why now
Windsor Pinterest feed is already live infrastructure (zero setup cost), and the audience — backyard/homestead lifestyle — matches CCC's codified buyer profile from the 19d-stale audience doc. Pinterest planning-cycle intent runs 4-8 weeks ahead of purchase, which aligns with clearing 5,974 units before Q4 season winds down. Doubling up with the Refersion affiliate surface (owned, 94 active) means the same clearance creative gets two independent distribution shots without doubling spend.
Test terms
Budget: $1,500/mo · Measured by: ThoughtMetric attributed revenue by utm_source=pinterest cross-checked against Pinterest Tag conversion count (proxy only — Pinterest is a new channel with no verification history, so treat Pinterest-reported numbers as directional and Shopify order UTMs as truth). · Success: Verified (ThoughtMetric) ROAS >= 2.0x with explicit inventory-clearance rationale (below 2.23x floor is acceptable ONLY if >= 80 overstock units clear in the window, priced in contribution). · KILL IF: After 30 days: ThoughtMetric revenue < $3,000 OR overstock unit clearance < 40 units OR ThoughtMetric-vs-Pinterest divergence > 50% (means we cannot get to two-source verification even directionally).
Proposals generated 43 day(s) ago; refreshed weekly so each stays stable long enough to evaluate. Nothing here moves money without explicit approval.
Channels
Is it working? — measured trend
ROAS and revenue are ThoughtMetric-measured. 2026-08 is a partial month; the arrow compares its run-rate ROAS to 2026-07. Same month last year shows what seasonality alone predicts — beat it and the changes are adding something.
| 2026-08 (partial) | 2026-07 | 2025-08 | ||||
|---|---|---|---|---|---|---|
| ROAS | Revenue | ROAS | Revenue | ROAS | Revenue | |
| ACCOUNT (blended) ▼ | 3.6x | $42,581 | 4.5x | $85,240 | 5.7x | $89,771 |
| Bing ▲ | 4.4x | $5,463 | 2.9x | $5,618 | 7.1x | $7,164 |
| Facebook ▼ | 0.8x | $884 | 1.1x | $2,079 | 1.1x | $3,480 |
| Google ▼ | 3.8x | $36,234 | 5.2x | $77,544 | 6.8x | $79,127 |
ROAS = revenue measured by ThoughtMetric multi-touch / platform spend. Google self-reported numbers run 26% above measured this quarter; money decisions use measured only.
BING
ROAS = revenue measured by ThoughtMetric multi-touch / platform spend. Bing self-reported numbers run 9% above measured this quarter; money decisions use measured only.
ROAS = revenue measured by ThoughtMetric multi-touch / platform spend. Facebook self-reported numbers run 89% above measured this quarter; money decisions use measured only.
Geo targeting — Google, analyzed 2026-05-25 → 2026-08-23
Performance verdict: geography is healthy -- spend concentrates where it converts (top 6 states carry 36% of spend). Configuration verdict: needs cleanup -- see the audit below.
✓ No exclusion candidates: every state with material 90-day spend (≥$500) has attributed conversions.
✓ All 8 enabled campaigns target by physical PRESENCE — the expert setting for freight-heavy ecommerce; none run Google’s leaky PRESENCE_OR_INTEREST default.
✖ 2 campaign(s) positively target a FOREIGN COUNTRY (Georgia) — see the geo defect card in Recommendations.
⚠ 222 negative ZIP-CODE exclusions across campaigns, origin unknown (predate the 30-day API history) — review card in Recommendations. Until claimed or removed, these silently block demand in those zips.
⚠ 393 redundant positive targets (states/zips/cities layered under a US country target, zero bid modifiers) — no serving effect, pure clutter; batch-remove for legibility whenever convenient.
| Campaign | Positive targets | Negative targets | Bid mods |
|---|---|---|---|
| (1) PMax - Shopping Only - Coops & Runs | City:2, Country:1, Postal Code:25, State:46 | City:1, Country:1, County:1, Postal Code:39, State:2 | 0 |
| (1) PMax - Shopping Only - Low Priced Coops | City:2, Country:2, State:47 | Country:1, Postal Code:63, State:2 | 0 |
| (B) Branded | City:2, Country:1, Postal Code:25, State:46 | City:1, Country:1, Postal Code:41, State:2 | 1 |
| (B) Branded No Image Extensions Test | City:2, Country:1, Postal Code:25, State:46 | City:1, Country:1, Postal Code:41, State:2 | 1 |
| (NB) Chicken Coops | City:2, Country:1, Postal Code:26, State:47 | Country:1, Postal Code:38, State:2 | 0 |
| (NB) Non-Brand (Max Conversion Test) | City:2, Country:2, State:47 | Country:1, State:2 | 0 |
| Branded Trial 7 | Country:1, State:1 | Country:1, State:2 | 1 |
| State | Spend (90d) | Share | ROAS* |
|---|---|---|---|
| California | $4,895 | 10% | 8.3x |
| Texas | $3,468 | 7% | 8.3x |
| New York | $2,693 | 6% | 8.8x |
| Florida | $2,436 | 5% | 10.1x |
| Virginia | $1,818 | 4% | 4.5x |
| North Carolina | $1,786 | 4% | 8.9x |
Underweighted states worth watching (strong ROAS, small share — expansion candidates if the trend holds): Minnesota (13.2x on $1,056), Illinois (12.4x on $1,216), Oregon (12.3x on $880), Missouri (10.8x on $863).
*Geo-grain ROAS is Google-attributed only — ThoughtMetric cannot verify at state grain, so geo reads size conservatively. Bing and Meta geo grain: not yet pulled (Bing unlocks with the API creds; Meta geo breakdown is a planned pull).
Search terms — Google, analyzed 2026-05-25 → 2026-08-23
5,787 distinct search terms, $12,297 term-attributable spend. Category vocabulary (tokens of converting and high-impression terms) is structurally protected from ever becoming a negative.
✓ No material negative-keyword waste: 54.6% of the $12,297 term-level spend converts (Google-attributed), and off-category residue is $466/90d — below the $500 action threshold.
Flagged for review (zero attributed conversions, off-category tokens) — judgment call, not auto-negative:
| Term | Campaign | Spend | Clicks | Off-category tokens |
|---|---|---|---|---|
| chicken coop company reviews | (B) Branded No Image Extensions Test | $157 | 4 | reviews |
| the chicken coop company reviews | (B) Branded No Image Extensions Test | $110 | 4 | reviews |
| jumbo chicken coop | (NB) Chicken Coops | $97 | 1 | jumbo |
| chicken coop company reviews | (NB) Chicken Coops | $51 | 7 | reviews |
| chicken coop company reviews | (B) Branded | $51 | 11 | reviews |
Attribution watch (category terms with zero ATTRIBUTED conversions — a tracking-grain artifact, do NOT negative): chicken coops for sale near me ($250); chicken coop ($213); large chicken coop ($63); chicken coop kits ($62); chicken coop and run ($57).
Campaigns
Top 5 — scale candidates
| Campaign | Channel | Funnel | Spend 90d | Platform ROAS 30d | Verified ROAS 90d | Tier |
|---|---|---|---|---|---|---|
| Intero | Catalog | Sales | bofu | $4,243 | – | 32.42x | inactive | |
| Intero | ATC | Sales | bofu | $145 | – | 15.60x | inactive | |
| (B) Branded No Image Extensions Test | bofu | $1,495 | – | 11.81x | ON TARGET | |
| (B) Branded | bofu | $3,458 | – | 7.58x | BELOW TARGET | |
| (NB) Chicken Coops | mofu | $11,853 | – | 6.39x | ON TARGET |
Bottom 5 — money at risk
| Campaign | Channel | Funnel | Spend 90d | Platform ROAS 30d | Verified ROAS 90d | Tier |
|---|---|---|---|---|---|---|
| Post: "Summer flock season is here ☀️🐓" | tofu | $83 | – | 0.00x | inactive | |
| Post: "☀️ We're kicking off summer with our biggest coop..." | tofu | $349 | – | 0.00x | inactive | |
| Post: "🇺🇸 Celebrate the Fourth with FREE gifts for your..." | tofu | $89 | – | 0.00x | inactive | |
| Intero | Prospecting | Traffic | tofu | $1,846 | – | 0.63x | inactive | |
| (1) PMax - Shopping Only - Low Priced Coops | mofu | $4,646 | – | 2.88x | BELOW TARGET |
All campaigns (14)
| Campaign | Channel | Funnel | Spend 90d | Platform ROAS 30d | Verified ROAS 90d | Tier |
|---|---|---|---|---|---|---|
| (NB) Chicken Coops | bing | mofu | $4,477 | – | 4.72x | BELOW TARGET |
| (1) PMax - Shopping Only - Coops & Runs | bing | mofu | $1,946 | – | 4.11x | BELOW TARGET |
| Intero | Catalog | Sales | bofu | $4,243 | – | 32.42x | inactive | |
| Intero | Prospecting | Traffic | tofu | $1,846 | – | 0.63x | inactive | |
| Post: "☀️ We're kicking off summer with our biggest coop..." | tofu | $349 | – | 0.00x | inactive | |
| Intero | ATC | Sales | bofu | $145 | – | 15.60x | inactive | |
| Post: "🇺🇸 Celebrate the Fourth with FREE gifts for your..." | tofu | $89 | – | 0.00x | inactive | |
| Post: "Summer flock season is here ☀️🐓" | tofu | $83 | – | 0.00x | inactive | |
| (1) PMax - Shopping Only - Coops & Runs | mofu | $24,083 | – | 5.39x | BELOW TARGET | |
| (NB) Chicken Coops | mofu | $11,853 | – | 6.39x | ON TARGET | |
| (1) PMax - Shopping Only - Low Priced Coops | mofu | $4,646 | – | 2.88x | BELOW TARGET | |
| (B) Branded | bofu | $3,458 | – | 7.58x | BELOW TARGET | |
| (2) Demand Gen - Video Placement | tofu | $2,270 | – | 4.81x | PEAK | |
| (B) Branded No Image Extensions Test | bofu | $1,495 | – | 11.81x | ON TARGET |
How to run this — the daily loop
Open the guide (one-time read; the page is self-explanatory day to day)
1. Read the top cards. Recommendations are ordered by business impact × strategy priority — inventory moves lead while overstock is objective #1. Money moves under $1,000/mo are batched under Housekeeping; real, but never the day’s focus.
2. Check the timer chip before acting. Every actionable card states its testing window up front (⏱ 7-day or 14-day verdict). An amber ⏸ HOLD chip means that campaign already has a change mid-measurement — do not touch it until the hold date, or the test is contaminated.
3. Make the change on the platform, then tap “I made this change.” Your Google sign-in identifies you automatically — no PIN or name entry. The check-off is logged, becomes a measured experiment automatically, and shows on the board within about 15 minutes.
4. In Flight answers “is it working?” Amber MEASURING = too early to judge, and the card says exactly when the verdict lands. Green VERDICT DUE = the window is complete; the card shows the measured before → after ROAS from the change engine.
5. Cadence. You will not act every day — that is by design. Most days: read, confirm nothing is on fire, check In Flight for green cards. Mondays: the weekly digest rolls up every measured change.
Never move money on a single source’s numbers — unverified channels are flagged red under Channels. Deep-clearance (20–30%) pricing only on SKUs Danny/Chris have explicitly approved; the standard sale is 15%.
Change management
Agency activity audit — Intero Digital, trailing 12 months
Finding: Meta (authoritative full-year audit log): Intero made 1850 changes Jul–Dec 2025, then 4 in ALL of 2026 — and 1 in the last 90 days. Google (API caps history at 30 days): 1064 events in the visible window, 182 of them on a single day (a bulk asset upload on 2026-06-15); 3 budget changes, 2 campaign changes, and nothing at all since 2026-06-30.
Caveats per verification standard: (1) Google's 2-year change history is not API-accessible — the full-year Google verdict needs a one-time manual export from the Change History page in the Google Ads UI (import path ready). (2) Meta actors 'Ashley King' and 'Sierra Liegler' are unresolved — if they are Intero staff, agency Meta activity extends to May 2026 and then stops; if not, it stopped in Dec 2025. Either way the last 60+ days show zero human agency activity on Meta. (3) Meta's /activities retention beyond 12 months is undocumented; the 12 months shown are complete. Sources: Google Ads API change_event + Meta Graph /activities, pulled 2026-07-09.
Searchable change log
Google: API serves the trailing 30 days only (2-year history is UI-export only); the log accumulates daily from here forward. Meta: full trailing year from Meta's own audit log. Bing: Microsoft exposes no change-history API — permanent gap, changes inferred from daily snapshot diffs going forward. Routine noise (ad delivered / billing) excluded.
| When | Channel | Actor | Action | Object | Detail |
|---|