GTW <TASK> MIL.COM / LG_MILITARY_ADS / 1034037341617540 LIVE
UPD 2026-06-05 SRC MASTER_AUDIT PORTAL
SPEND 30D
$841K
matches prior audit
OPP SCORE
99/100
structural OK
ROAS
N/A
value pipe broken
MAX VAL PREMIUM
+67%
$137K/mo bleed
FREQ
6.38
burn risk
CONCEPT CONC
76%
2 families · $640K
RETARGETING
~20
people · SAC blocked
EMQ PURCHASE
9.1
CompReg 8.3
PRI
OWNER
WEEK
STATUS
FIND
PRI STATUS TASK OWNER WK DUE $ IMPACT DEPS
▎ TODAY · FRI JUN 5 · ZERO-COST CLEANUP + VERIFICATION (DO BEFORE WEEKEND)
P0 PEND A01EVENTS MGR DIAGNOSTICS · DATASET 332555186604235 · PURCHASE VALUE PARAM CHECK
→ events_manager_walkthrough.md · screenshot the malformation flag
 MO FRI 2026-06-05 GATES $1.6M/YR —
WHY THIS MATTERS
Root-cause finding from the audit. Meta returns purchase_roas: "Not available" at every level — account, campaign, ad set, ad. This means Auto Max Val (which optimizes for RETURN_ON_AD_SPEND) is paying a 67% CPL premium ($36.44 vs $21.84 BAU) against a signal that doesn't compute. Confirming this in Events Manager is the gate for everything else. Without screenshot evidence, every downstream decision (especially the Max Val pause call) is on inference, not evidence.
HOW TO FIX
  1. Go to business.facebook.com → Events Manager
  2. Select the Iron Corp / Military.com BM (containing LG Military Ads 1034037341617540)
  3. Datasets sidebar → click "Military Insurance" (ID 332555186604235) — NOT "Military.com's Pixel" or "MA Military Insurance"
  4. Click the Diagnostics tab (top nav)
  5. Look for a flag on the Purchase event value parameter: "formatting issues", "invalid format", or "recommend updating value field"
  6. If found: expand it, screenshot the affected event count + recommended format
  7. If NOT in Diagnostics: Overview → Event Details → Purchase event → check value distribution. 98%+ same number = broken pipe even if Meta doesn't flag
SUCCESS CRITERIA
A screenshot proving (or disproving) the value malformation claim. This becomes the artifact you walk into the dev conversation with. If confirmed: Auto Max Val decision becomes data-supported, not inference-supported.
DO NOT CLICK
The Diagnostics tab sometimes shows "Apply Recommendation" / "Fix" buttons. Don't click these without dev review — Meta's auto-fixes sometimes restructure the event schema and break downstream QW/Delty postbacks.
EFFORT
5 minutes. Do this before the weekend so Mon Jun 8 starts with the evidence in hand.
P0 PEND A02EVENTS MGR DIAGNOSTICS · FBC/FBCLID PARAM TRUNCATION CHECK
→ note affected ad set count
 MO FRI 2026-06-05 VALIDATES $67K BLEED —
WHY THIS MATTERS
Prior audit flagged 15 ad sets, $67,174 spend, 27% of CompleteRegistration events affected by fbc truncation. API confirms fbc is PRESENT (87% coverage on Purchase, 100% on CompleteRegistration) but cannot validate value integrity. Truncation degrades attribution and trains Meta's algorithm on partial signal.
HOW TO FIX
  1. Same dataset (332555186604235), same Diagnostics tab as A01
  2. Look for warnings on fbc: "truncated", "invalid format", "missing characters"
  3. Expand if found — note ad set count, affected event count, example fbc value
  4. Healthy fbc starts with fb.1. with 3+ dots. Truncated = fewer dots.
SUCCESS CRITERIA
Screenshot of fbc truncation flag OR confirmation it's resolved. If confirmed: hand to dev as a URL-parameter-handling fix (typically a 1-2 day engineering task).
DEPS
Doing A01 first means you're already in the right dataset — saves clicks.
EFFORT
3 minutes while you're already in Diagnostics.
P1 PEND A03ADS LIBRARY · CONFIRM "SAVE UP TO 60%" CREATIVES ZERO LIVE
→ search page 6179724269 for "60%"
 MO FRI 2026-06-05 COMPLIANCE —
WHY THIS MATTERS
May 13 audit flagged 3 active ads running "Save up to 60%" claim (4× the published carrier band, no qualifier, no disclosure). Jun 4 audit didn't see them in top 50 spenders. Sweep the full library to confirm zero live before the meeting — Hassan will ask.
HOW TO FIX
  1. Open facebook.com/ads/library
  2. Filters: Country = US, Active status = Active, Ad category = All ads
  3. Page search: Military.com (page ID 6179724269)
  4. In the keyword filter: "60%"
  5. Note any active ads — record Ad IDs
  6. If any are live: pause via Ads Manager before EOD
SUCCESS CRITERIA
Zero active ads with "60%" claim language. Or list of any active with pause action queued for Monday.
EFFORT
5 minutes.
P2 PEND A04BUSINESS SETTINGS · ARCHIVE STATUS "MIL FB CUSTOMER MATCH 4.12.21"
→ confirm replacement by Custom_Recruiting-Leads_08-05-2025
 MO FRI 2026-06-05 — —
WHY THIS MATTERS
Prior audit referenced this audience by name; Jun 4 audit didn't find it in active inventory. Confirming archive status closes the loop on the customer-match list question and prevents Lauren saying "that's the one we use" later.
HOW TO FIX
  1. Business Settings → Data Sources → Audiences
  2. Search "Mil FB Customer Match 4.12.21"
  3. Check archived audiences tab if not in active list
  4. Note last refresh date if found
SUCCESS CRITERIA
Confirmed status: archived / deleted / replaced. If found archived with recent date, confirm with Lauren which is the canonical live customer-match list.
EFFORT
2 minutes.
P1 PEND A05DELETE 28 DEAD PLACEHOLDER LOOKALIKES FREE CLEANUP
→ all sized 1,000 · INACTIVE · zero risk
 MO FRI 2026-06-05 HYGIENE —
WHY THIS MATTERS
28 lookalikes all sized at exactly 1,000 people — Meta's signature for orphaned/broken lookalikes (seed audience deleted). Zero value as targeting, zero risk to delete. Reduces account audience list from 38 → 10 active. Mo's 30-day plan task B4.
HOW TO FIX
  1. Ads Manager → Audiences
  2. Filter: Status = Inactive, Type = Lookalike
  3. Select all entries sized at 1,000 with patterns like:
    • "Lookalike (Worldwide, 1%) - ..."
    • "Lookalike (US, 1%) - ..." dated 2025-07-09 or 2025-07-22
    • "Lookalike (US, 3% / 10%) - ..." same dates
  4. Bulk delete via the action menu
SUCCESS CRITERIA
28 placeholders gone. Active audience inventory reduces to ~10 entries.
DO NOT DELETE
The active LAL ladder on Custom_Recruiting-Leads (1%/3%/5%) — those are properly sized (1.9M–10.1M) and powering the ALC + Max Val campaigns. Only delete the 1,000-size INACTIVE ones.
EFFORT
10 minutes.
▎ MON JUN 8 · WEEK 1 KICKOFF · UNBLOCK MAX VAL DECISION + REQUEST INPUTS
P0 BLKD B01DEV TIMELINE FOR PURCHASE VALUE-PIPE FIX · BEST + WORST CASE WEEKS
→ unblocks B02 (Max Val decision)
 DEV W1 2026-06-10 GATES $137K/MO A01
WHY THIS MATTERS
This number gates the Auto Max Val decision (B02). Without an honest timeline (best + worst case weeks), the Max Val call is a guess. 2 weeks = cap. 6+ weeks = pause + reallocate.
HOW TO FIX
  1. Send Anthony (or his dev lead) a Slack / email
  2. Attach the screenshot from A01
  3. Scope the question precisely: "Who passes the QW per-lead payout to the Purchase event? Where does that integration sit? What server-side changes are needed?"
  4. Ask explicitly: "Best case weeks / worst case weeks / what's the work?"
  5. Push back if you get vague answers — get a number, even with a wide range
SUCCESS CRITERIA
Written timeline estimate (e.g., "2-4 weeks depending on QW API access"). With this you can make the Max Val call Tuesday.
DEPS
A01 — need the diagnostic screenshot to scope the fix concretely.
EFFORT
30 min to send and follow up. Their response may take 1-2 business days.
P0 BLKD B02DECIDE AUTO MAX VAL · PAUSE / CAP / KEEP BIGGEST DECISION
→ auto_max_val_decision.md
 HASSAN W1 2026-06-10 +$110K/MO B01
WHY THIS MATTERS
$341K/30d optimizing against a broken signal. 67% CPL premium = $137K/mo bleed. Largest single $$$ decision in the engagement. The deeper-dive doc walks through the three options with math.
HOW TO FIX
Three options:
  • A — Pause + reallocate to BAU: saves ~$110K/mo at same volume (BAU CPL $21.84 vs Max Val $36.44)
  • B — Cap at current spend: holds line, decides later. Lowest risk, lowest reward.
  • C — Keep running: bleeds $137K/mo during fix window. Worst-of-both.
Decision rule based on B01 dev timeline:
  • 2 weeks → Option B (cap, hold)
  • 3-4 weeks → Option A (pause + reallocate)
  • 6+ weeks → Option A (pause + reallocate)
Bring the framing from auto_max_val_decision.md verbatim — Hassan-tested language.
SUCCESS CRITERIA
Hassan-signed decision in writing (text / email / Slack). Action queued for Wed.
DEPS
B01 — without a dev timeline you're guessing.
PUSHBACK PREP
If Hassan says "Max Val has been running for months, must be working" → response is in the deep-dive doc. Short version: Meta returning purchase_roas: "Not available" at every level IS the algorithm telling us optimization isn't happening.
EFFORT
30 min Hassan conversation Tuesday or Wednesday.
P1 PEND B03REQUEST FRESH 180-DAY SOLD-LEADS EXPORT FROM QW/DELTY
→ rebuild customer-list seed (current is 10mo stale)
 LAUREN W1 2026-06-10 +5-10% LAL LIFT —
WHY THIS MATTERS
Current customer-list seed (Custom_Recruiting-Leads_08-05-2025) is 10 months old. All three active LALs (1%/3%/5%) — which power ALC + Max Val campaigns — are downstream of this stale seed. A 10-month-old seed underweights recent buying behavior; refreshing yields ~5–10% LAL performance lift across the LAL-driven campaigns.
HOW TO FIX
  1. Slack / email Lauren
  2. Ask: "Can you pull a CSV export of sold leads from the last 180 days for a fresh customer-match upload? I want to rebuild the LAL ladder on current signal."
  3. Required fields: hashed or plain email (Meta will hash on upload), phone if available, name, ZIP, country
  4. Confirm format: CSV with column headers — Meta accepts most formats
SUCCESS CRITERIA
CSV file from Lauren in your inbox by Wed. Triggers C02 (rebuild seed).
EFFORT
5 min ask. Lauren's response may take 1-2 days.
P1 PEND B04PULL MAR + APR 2026 DAILY REPORTS · SAME FORMAT AS MAY
→ baseline vetting input · required before mid-June Hassan 1:1
 LAUREN W1 2026-06-09 $45-67K NEGO —
WHY THIS MATTERS
Hassan proposed $275K (May) as the baseline. May data shows the month ended at a $87K-exit-run-rate after Memorial Day collapse — meaning $275K is misleading. To argue baseline DOWN (worth $45-67K to Mo over the pilot), you need March + April for YoY-comparable context.
HOW TO FIX
  1. Slack Lauren
  2. Ask: "Can you send daily reports in the same format as the May file from July 2025 through May 2026? Full post-acquisition history."
  3. Frame as execution prep, not baseline negotiation: "Each category we're running has its own seasonal rhythm. Want to use last year's numbers as the hypothesis for budget + creative allocation in the back half."
  4. Don't telegraph the baseline fight
SUCCESS CRITERIA
CSV files for at minimum Mar + Apr 2026 in your inbox by Wed. Triggers D06 (YoY analysis).
EFFORT
5 min ask. Should be fast if pulled from internal tool.
P1 PEND B05QW CONTRACT TERMS · EXCLUSIVITY · PING-POST PERMISSIONS
→ gates form migration Tier 2 viability
 ANTHONY W1 2026-06-12 GATES $1-3M —
WHY THIS MATTERS
The form-migration play's big money is in multi-buyer ping-post (Tier 2) — posting each lead to QW + EverQuote + MediaAlpha simultaneously and taking the best bid. Potential 20-30% RPL lift. But if QW contract has exclusivity, ping-post is blocked and the migration value drops by half. Cannot scope the migration without reading the contract.
HOW TO FIX
  1. Ask Anthony for a copy of the current QW/Delty/LendingTree contract
  2. Look for: exclusivity clause, volume commitment, termination notice, ping-post permissions, revenue share / accept-rate structure
  3. If exclusivity present: scope the wind-down before any migration work
  4. If ping-post explicitly permitted: this is the big unlock
SUCCESS CRITERIA
Contract reviewed, key terms documented in a one-page memo for the form-migration scoping conversation.
EFFORT
30 min reading + ~1 hour summary memo.
P2 PEND B06DECIDE MA MILITARY INSURANCE DATASET · 1536649764252698
→ never fired server-side · activate or archive
 ANTHONY W1 2026-06-12 HYGIENE —
WHY THIS MATTERS
"MA" likely = MediaAlpha. Dataset created Dec 1, 2025. Browser fires today; server has never fired. Either an unactivated MediaAlpha integration or a stale test dataset. Anthony's earlier framing was "ignore MediaAlpha, focus on Delty" — consistent with this dataset being intentionally dormant. Just confirm and decide.
HOW TO FIX
  1. Slack Anthony
  2. Ask: "The MA Military Insurance dataset (1536649764252698) has never fired server-side. What is it for? Activate or archive?"
  3. Apply his decision: activate (engineering work) or archive (Business Settings → Datasets → Archive)
SUCCESS CRITERIA
Anthony-confirmed status. Either activated (with dev plan) or archived.
EFFORT
5 min ask + ~10 min to archive if that's the decision.
P2 PEND B07SCOPE SIGNALS GATEWAY / CAPI FOR DATASET 587241028099131
→ content pixel · server silent since Mar 2023 · +23% lever
 DEV W1 2026-06-12 +23% COST/RESULT —
WHY THIS MATTERS
"Military.com's Pixel" — the content pixel from 2016 — runs browser-only. Server hasn't fired since March 2023. Adding a Signals Gateway / CAPI here lets Meta see retargeting events from military.com's broader content footprint. Estimated impact: +23% cost/result. Engineering-only (no creative dependency).
HOW TO FIX
  1. Ask dev to scope adding Conversions API Gateway for dataset 587241028099131
  2. This is typically a 24–48 hour dev sprint (deploy Stape or Meta's CAPI Gateway, configure event mapping)
  3. Confirm event taxonomy to mirror: at minimum PageView + SubscribedButtonClick + Lead (currently browser-only)
SUCCESS CRITERIA
Dev commits to a build window. Target landing: Week 2 (Jun 15-19).
EFFORT
15 min scoping conversation. Build is dev-team estimated 24-48 hours of engineering.
▎ WEEK 1 EXECUTION · MON-FRI JUN 8-12 · SHIPPABLE
P1 PEND C01BUILD NEW AD SET · FB+IG ENGAGEMENT WARM LAYER FREE LEVER
→ page_engagement_audience_build.md · $1.5K/day · 2 creatives
 MO W1 2026-06-10 +5-10% QUAL LEADS —
WHY THIS MATTERS
Under SAC, website-Pixel retargeting is essentially nonexistent (All Site Visitors / Visited Auto LP are at ~20 people each). BUT the page-engagement audiences (FB Page Likers 1M, FB Engagers 341K, IG Followers 30K, IG Engagers 93K) are healthy and SAC-compliant — and completely unused as ad-set targeting. Building one ad set against them = the only warm layer the account has access to. Free lever.
HOW TO FIX
Full spec in page_engagement_audience_build.html. Quick version:
  1. Add ad set to Auto SAC BAU campaign (120240808864310498) — pile into existing post-learning campaign
  2. Name: Auto BAU - Page+IG Engagement (Warm)
  3. Optimization: OFFSITE_CONVERSIONS (matches sibling ad sets)
  4. Daily budget: $1,500
  5. Include: union of FB Page Likers + FB Engagers + IG Followers + IG Engagers
  6. Exclude: Custom_Recruiting-Leads_08-05-2025 (suppress existing customers)
  7. Creatives: Lower-Your-Auto_04/28 + Drive-On_05/30 (two proven winners). NOT TYFYS, NOT Veteran-VO.
SUCCESS CRITERIA
Ad set live by Wed EOD. Monitor first 72 hours for CPL trajectory. Expected CPL: $15–17 (20-30% below cold).
DECISION RULE AFTER 72H
CPL < $20 → scale to $3K/day. $20-28 → hold. $28-36 → investigate. > $36 → pause.
EFFORT
30 min to build in Ads Manager. Set-and-monitor after.
P1 PEND C02REBUILD CUSTOM-LIST SEED FROM FRESH 180D SOLD-LEADS EXPORT
→ then rebuild 1%/3%/5% LALs
 MO W1 2026-06-11 +5-10% LAL LIFT B03
WHY THIS MATTERS
Once you have Lauren's fresh 180-day sold-leads CSV (B03), the rebuild is mechanical. Fresh seed → fresh LALs → 5-10% LAL performance lift on the campaigns currently using them.
HOW TO FIX
  1. Ads Manager → Audiences → Create → Custom Audience → Customer List
  2. Upload Lauren's CSV. Meta will hash and match. Match rate of 50%+ is healthy.
  3. Name: Custom_Sold-Leads_2026-06 (or similar — keep naming convention consistent)
  4. Once the audience is ready (1-24h), create Lookalike audiences: 1%, 3%, 5% all US-targeted
  5. Swap the LAL audiences in active ad sets from the old (Aug 2025) to the new June 2026 LALs
  6. Keep old LALs for 1 week as fallback; archive after performance validates
SUCCESS CRITERIA
New LAL ladder live; old LALs swapped out of ad sets; performance comparison ready for Friday readout.
DEPS
B03 — Lauren's CSV.
EFFORT
1 hour total (15 min upload, wait for processing, 15 min create LALs, 30 min swap into ad sets).
P1 PEND C03SHIP 8-10 NET-NEW CREATIVE CONCEPTS
→ Home Inside_Icon + Drive-On + spouse-angle
 MO W1 2026-06-12 FATIGUE OFFSET —
WHY THIS MATTERS
Concept concentration is 76% of spend on 2 families. Odometer 05/15 single ad at frequency 3.27 — ~14 days from fatigue cliff. New creative isn't optional; it's the only way to slow the fatigue clock.
HOW TO FIX
Priority concept directions (from audit):
  • Home Inside_Icon variants (3-4 new) — Home format is best-CPL in the account at $12.68. Under-funded at $61K vs $371K Auto.
  • Drive-On variants (2-3 new) — newest creative trending at $18.43 BAU. Scale on its proven format.
  • Spouse-angle test (1-2 new) — competitor Veteran Life Insurance proves angle works. Military.com has zero spouse-specific creative.
  • Video — anything (1-2 new) — current 5:1 static:video split is a gap. Even one new video tests the format.
Use War Room AI variant generation pipeline. Mo+Lauren approval batch (not per-ad). Ship Mon + Thu.
SUCCESS CRITERIA
8-10 new concept ads live by Friday. Frequency on Odometer 05/15 starts declining as new variants absorb impressions.
EFFORT
~3-4 hours across the week (concept briefs, generation, QA, upload).
P1 PEND C04REALLOCATE AUTO BAU PACING GAP · $8.5K/DAY UNDEPLOYED
→ shift to Drive-On + Lower-Your-Auto BAU winners
 MO W1 2026-06-10 +$255K/MO SPEND —
WHY THIS MATTERS
Auto BAU campaign cap is $19,800/day. Actual spend is $11,296/day (57% utilization). $8,500/day of approved budget is sitting unused. At BAU's $21.84 CPL that's ~390 additional leads/day if filled smartly.
HOW TO FIX
  1. Ads Manager → Auto BAU campaign (120240808864310498)
  2. Identify the ad sets currently throttling: usually the algorithmic CBO/CPR bid strategy is limiting spend
  3. Options to fill the gap:
    • Raise bids on Drive-On + Lower-Your-Auto BAU ad sets gradually (10-20% increments)
    • Loosen the CPR target slightly (e.g., $22 → $25) on the best-performing ad sets
    • Add the new warm-layer ad set (C01) — that absorbs $1.5K/day immediately
  4. Don't over-fill in one move — Meta's algorithm punishes sudden 50%+ daily spend increases
SUCCESS CRITERIA
Spend pacing closer to $16-18K/day by Friday without CPL inflation > 10%.
WATCH FOR
If CPL climbs above $25 while filling, throttle back. The gap exists for a reason; don't force.
EFFORT
30 min + daily monitor.
P1 PEND C05SCALE HOME SPEND SHARE 8% → 15% (WK1) → 20% (WK2)
→ Home CPL 39% lower than Auto
 MO W1 2026-06-12 +8-12% NET REV —
WHY THIS MATTERS
Home blended CPL is $16.34 vs Auto $26.77 — Home is 39% cheaper. Currently Home is 9.5% of spend, Auto is 90.5%. Both have similar ROAS (1.36 vs 1.33). Pure arbitrage: shifting spend from Auto to Home at constant total budget improves blended CPL.
HOW TO FIX
  1. Raise daily budget on Home SAC ALC ($4K → $6K) and Home SAC BAU ($2K → $3K) gradually
  2. Add 1-2 net-new Home creatives (covered in C03 — Home Inside_Icon variants)
  3. Trim Auto Max Val if Hassan signs Option A on B02 (otherwise leave Auto alone for now)
  4. Week 1 target: Home share at 15% of total spend
  5. Week 2 target: Home share at 20%
SUCCESS CRITERIA
Home daily spend ~$5K (was $2.7K) by Friday. Watch CPL — if it stays under $18 the scale is healthy.
EFFORT
20 min budget adjustments + daily monitor.
P2 PEND C06NARROW CUSTOM AD SETS · CONFIRM SOURCE SIZE OR REMOVE
→ IDs 120240808864440498 / 120227142278180498
 MO W1 2026-06-12 HYGIENE —
WHY THIS MATTERS
Prior audit flagged 2 Custom ad sets as targeting ~1,000-person audiences. Live data shows reach of 716K and 52K respectively (delivery is fine) — but the underlying source audience may still be the narrow ~1K. Confirm and either fix the source or remove.
HOW TO FIX
  1. Open each ad set in Ads Manager:
    • 120240808864440498 — "Custom" in Auto BAU
    • 120227142278180498 — "Custom" in Home BAU
  2. Click into Audience definition
  3. Note the source CUSTOM audience name and size
  4. If size is ~1,000 (broken state): replace with the refreshed Custom_Sold-Leads_2026-06 from C02
  5. If size is healthy (>10K): leave as-is, no action
SUCCESS CRITERIA
Both ad sets either confirmed healthy or repaired with fresh audience.
EFFORT
15 min.
P3 PEND C07VERIFY CUSTOMER-LIST LAL ELIGIBILITY UNDER SAC IN-PRODUCT
→ 30-sec sanity check (likely confirmed by live state)
 MO W1 2026-06-12 VALIDATE —
WHY THIS MATTERS
Mo's W4-T3 task. Customer-list LALs under SAC for Financial Products are technically allowed but vary by account history. Live state shows the LAL ladder IS active and properly sized — likely confirmed already, but verify in-product.
HOW TO FIX
  1. Try creating a test Lookalike from a customer list in this account
  2. If Meta allows it without error: confirmed eligible
  3. If Meta blocks with SAC warning: document the message, surface to Anthony
EFFORT
2 min.
P1 PEND C08FRIDAY READOUT TO HASSAN · $-IMPACT FORMAT
→ first weekly readout · sets cadence
 MO W1 2026-06-12 REPORTING —
WHY THIS MATTERS
First Friday readout sets the cadence Hassan trusts. Format matters: lead with $-impact, end with what's coming. Avoid agency-speak.
HOW TO FIX — STRUCTURE
Three bullets:
  1. What we did this week: X cleanup tasks completed (28 LALs deleted, Y warm ad set live, etc.) + the Max Val decision status
  2. What we moved: early week 1 numbers vs trailing 4-week baseline — even directional
  3. What's coming: Week 2 priorities (CAPI build start, scale warm layer, ship more creative)
Length: 5 sentences max. Send by 4pm Friday.
EFFORT
20 min Friday afternoon.
▎ WEEK 2 · MON-FRI JUN 15-19 · CAPI BUILD + AUDIENCE OPTIMIZE
P0 HOLD D01PURCHASE VALUE-PIPE FIX · SERVER-SIDE BUILD START
→ enables Max Val re-light
 DEV W2 2026-06-19 +7% ROAS B01
WHY THIS MATTERS
The root-cause fix. Pass real QW per-lead payout into the Purchase event's value field. Meta estimates +7% ROAS account-wide from fixing this alone. Multi-week dev work — must start W2 to land by W4 measurement window.
HOW TO FIX (DEV)
  1. Identify where the QW postback lands (likely an Iron Corp server or webhook)
  2. Map postback payload to Meta CAPI Purchase event with proper value field (real per-lead payout in USD, decimal)
  3. Ensure dedup with browser-side Purchase via event_id
  4. Backfill last 7-14 days via offline event upload to seed Meta's optimizer
  5. Verify in Events Manager → value distribution shows real numeric range
SUCCESS CRITERIA
Purchase event value parameter shows real payout amounts. purchase_roas starts populating in Ads Manager. Diagnostics tab clears the malformation flag.
EFFORT
Dev-estimated 2-4 weeks (depends on QW API access). Started W2, lands by W4.
P1 HOLD D02CAPI GATEWAY · CONTENT PIXEL 587241028099131 BUILD
→ +23% cost/result · 24-48hr sprint
 DEV W2 2026-06-17 +23% CPR B07
WHY THIS MATTERS
The content pixel for military.com root has no CAPI since March 2023. Adding it = +23% cost/result lever. Faster build than the value-pipe fix; can land in W2.
HOW TO FIX (DEV)
  1. Deploy Meta Conversions API Gateway (or Stape.io self-hosted) for dataset 587241028099131
  2. Configure server-side firing of PageView + SubscribedButtonClick + Lead
  3. Wire match-key passing (hashed email, phone where available)
  4. Validate in Events Manager: server events firing within hours
SUCCESS CRITERIA
Dataset shows server events firing within 24h of go-live. EMQ improves.
EFFORT
24-48 hours of dev time. Scope set W1 (B07), build W2.
P1 HOLD D03SCALE WARM AD SET · $1.5K → $3K/DAY IF CPL < $20
→ decision after 72hr observation
 MO W2 2026-06-15 +$50-100K/MO C01
WHY THIS MATTERS
If the warm-layer ad set hits target CPL (<$20), scaling 2× extracts more from the unique unused audience.
HOW TO FIX
Decision rule by Monday Jun 15:
  • If CPL < $20 → raise budget to $3K/day, add a 3rd creative variant
  • If $20-28 → hold at $1.5K, tweak creative
  • If >$28 → investigate / pause
EFFORT
10 min.
P1 HOLD D04SHIP 10-12 NET-NEW CREATIVE · WK2
→ variants of W1 winners + new tests
 MO W2 2026-06-19 FATIGUE OFFSET —
HOW TO FIX
Iterate on Week 1 winners. Generate 3-4 variants of best-performing concept. Add 2-3 new tests in directions Week 1 data suggests.
P1 HOLD D05FORM MIGRATION PHASE 0 · QW CONTRACT REVIEW + DEV SCOPE
→ no code touching live · discovery only
 MO W2 2026-06-19 SCOPE $200K/MO B05
WHY THIS MATTERS
Phase 0 = paper review + dev capacity scoping. No live code touched. Sets up Phase 1 (staging build) for W3-W4 if Phase 2 (live A/B) is going to land cleanly in Q3.
HOW TO FIX
Use the QW contract review from B05. Map current form architecture (iframe / redirect / Iron Corp own page posting to QW API). Document the migration plan: Phase 1 build in staging W3-W4-Jul, Phase 2 A/B test 10-20% traffic late Jul, Phase 3 full cutover post-Aug measurement window.
P2 HOLD D06RUN YOY ANALYSIS · PREP BASELINE CASE FOR HASSAN
→ uses March + April data from B04
 MO W2 2026-06-19 $45-67K NEGO B04
WHY THIS MATTERS
The data-backed argument for arguing the baseline DOWN from Hassan's $275K. Use the four candidate baseline methods (exit run rate / trailing avg / April / YoY-adjusted) and pressure-test his +64% Jun→Dec growth assumption.
HOW TO FIX
Framework already in audit doc and June plan baseline section. Drop the new Mar+Apr data in, compute the four candidate baselines, write the one-page case file for the Hassan 1:1 (E01).
P2 HOLD D07VETERAN UGC SOURCING · MICRO-CREATOR OUTREACH
→ first 3-5 pieces target end of June
 MO W2 2026-06-22 MOAT —
WHY THIS MATTERS
Zero veteran UGC in the account. Zero across the entire competitor scan. Defensible creative moat unbuilt — and Military.com is uniquely positioned to ship it credibly.
HOW TO FIX
Outreach plan in creative engine plan. Target micro-creators ($300-600/piece). Casting brief: real veterans in current civilian roles, not actors. Aim for 3-5 pieces in market by Jun 22.
▎ WEEK 3 · MON-FRI JUN 22-26 · BASELINE LOCK + STRATEGIC CONVERSATIONS
P0 HOLD E01MID-JUNE 1:1 WITH HASSAN · BASELINE LOCK
→ 30-min slot · use case file from D06
 HASSAN W3 2026-06-22 $45-67K D06
WHY THIS MATTERS
Single highest-leverage admin task in June for Mo's comp. Arguing baseline from $275K to ~$155-185K is worth $45-67K to Mo over the pilot. Wait until Week 3 — after Week 1 wins are banked, Mo negotiates from strength.
HOW TO FIX
  1. Walk in with the one-page case file from D06
  2. Open: "Looking at March-April-May together, the trajectory tells a different story than the May headline. Want to walk you through what a fair baseline looks like."
  3. Present the four candidate baselines with the math
  4. Land on a specific number; get it in writing (text/email confirmation Tuesday)
DEPS
D06 — the case file.
EFFORT
30 min conversation + 30 min prep.
P1 HOLD E02RPL−CPL METRIC REFRAME PITCH · POST-BASELINE-WIN
→ "you only pay me to grow real net rev"
 HASSAN W3 2026-06-25 METRIC ALIGN E01
WHY THIS MATTERS
Mo's "most important strategic ask" per Master Plan W5-T1. Under SAC the standard CPL levers are blocked; CPL alone is a misleading metric. RPL−CPL spread captures both sides of the equation. Operator framing Hassan respects.
HOW TO FIX
Pitch only AFTER baseline conversation (E01) goes well. Open: "Now that we have a fair baseline, want to propose one more refinement to how we measure this. Right now we're optimizing for CPL only — but under SAC that misses half the equation. Switching to RPL−CPL spread or net rev per lead means you only pay me more when you actually make more."
P1 HOLD E03CAPI · ACCEPTEDLEAD EVENT BACKFILLED FROM QW HISTORICALS
→ enables Meta optimization switch from form-fill to accepted-lead
 DEV W3 2026-06-26 +15% CPQL D01
WHY THIS MATTERS
Once D01 (value-pipe fix) lands, the AcceptedLead event can be backfilled with real per-lead acceptance + payout data. Meta then trains on revenue-quality leads, not form-fills. Meta's published benchmark: 15% lower CPQL.
HOW TO FIX
Dev work continues from D01. Add AcceptedLead as named event in CAPI taxonomy. Upload 7-14 days of historical QW accepted-lead data as offline events. Verify in Events Manager that AcceptedLead is firing at expected volume.
P1 HOLD E04SHIP 12-15 NET-NEW CREATIVE · CONCENTRATE WINNERS
→ deeper variants on validated winners · kill losers
 MO W3 2026-06-26 SCALE PREP —
HOW TO FIX
By W3 you have W1 + W2 performance data. Triple down on validated winners with 4-5 variants each. Kill bottom-decile losers. Net add 12-15 new ads while removing under-performers — keeps ad count healthy.
P2 HOLD E05FORM MIGRATION PHASE 1 · BUILD IN STAGING (NO LIVE)
→ wire consent + certs · server-side CAPI · demo-ready
 DEV W3 2026-06-26 SCOPE Q3 D05
HOW TO FIX (DEV)
Build the own-form in staging. Wire Jornaya + TrustedForm certs properly. Set up seller-specific consent. Stand up server-side CAPI on the new form path. Do not touch live traffic. Get to "demo-ready" by end of W3 so A/B test can launch in July.
P2 HOLD E06VETERAN UGC · 3-5 PIECES IN MARKET
→ defensible moat
 MO W3 2026-06-26 MOAT D07
HOW TO FIX
Sourcing started W2 (D07). By W3 first batch lands. Upload to Ads Manager, attach to BAU campaigns as test ad sets, monitor performance vs static creative baseline.
▎ WEEK 4 · MON-FRI JUN 29 - JUL 3 · CONSOLIDATE + END-OF-MONTH REVIEW
P0 HOLD F01VALUE-PIPE FIX · GO LIVE
→ unlocks Max Val re-light · +7% ROAS account-wide
 DEV W4+ 2026-07-03 +$60K/MO D01·E03
WHY THIS MATTERS
The audit's #1 fix lands. Meta starts computing real Purchase ROAS. Auto Max Val (if it was paused) can be re-lit. Account-wide +7% ROAS lift per Meta's benchmark.
HOW TO FIX
Verify in Events Manager → Purchase event → value distribution shows real numeric range. purchase_roas starts populating in Ads Manager within 24-48h. Diagnostics flag clears.
SUCCESS CRITERIA
purchase_roas available at ad set level in Ads Manager. EMQ on Purchase holds or improves.
P0 HOLD F02SWITCH META OPTIMIZATION TO ACCEPTEDLEAD · POST VALUE-FIX
→ algorithm trains on revenue · not form-fills
 MO W4+ 2026-07-03 SCALE PREP F01
HOW TO FIX
After F01 ships and AcceptedLead has 50+ events/week, test switching one ad set's optimization from custom-conversion (current "click to quotes") to AcceptedLead. Run a holdout for 7 days. If CPQL improves: roll to all ad sets.
P1 HOLD F03END-OF-JUNE REVIEW WITH HASSAN · MONTH 1 SUMMARY + JUL/AUG FCST
→ formal monthly cadence
 HASSAN W4+ 2026-06-30 RELATIONSHIP —
HOW TO FIX
30-min slot end of June. Present: what shipped (creative count, audience builds, CAPI lands), what moved ($-impact vs locked baseline), goal-post forecast for Jul/Aug. Surface HELOC (F05) as a "later" item. Ask for end-of-July checkpoint slot.
EFFORT
1 hour prep + 30 min meeting.
P1 HOLD F04SHIP 10-12 NET-NEW CREATIVE · CONCENTRATION PHASE
→ prep Jul-Aug measurement window
 MO W4+ 2026-07-03 MEAS WINDOW —
HOW TO FIX
By end of June you have ~30-40 net-new ads in market. W4 is consolidation — concentrate spend on validated winners; ship 10-12 more variants on those winners for the measurement window.
P2 HOLD F05HELOC EXPANSION THESIS · SURFACE TO HASSAN (Q3+)
→ mention in EOM review
 HASSAN W4+ 2026-06-30 FUTURE F03
HOW TO FIX
In the EOM review, mention HELOC as Q3+ idea: "Adjacent revenue lane outside QW. Same audience. Worth scoping but not June work." Don't divert capacity. Just plant the flag.
P2 HOLD F06PHASE 2 RETAINER SCOPING · THREE-THREAD POST-PILOT
→ media + comp + recruitment partnerships
 MO W4+ 2026-07-03 $50-100K/MO —
HOW TO FIX
Internal Mo prep doc. Outline Phase 2 (Stage 3 from original pitch): full media ownership + cross-portfolio expansion + recruitment partnership program. Targeted at end-of-pilot Hassan conversation.
▎ JUL · AUG · MEASUREMENT WINDOW · 60-DAY RUN RATE TEST
P1 HOLD G01FORM MIGRATION PHASE 2 · 10-20% TRAFFIC A/B TEST
→ ONLY after measurement window closes
 MO W4+ 2026-09-01 +$200K/MO E05
WHY THIS MATTERS
The first live traffic test of the own-form. Phased to land AFTER the Aug 31 measurement window so cutover wobble doesn't cost Mo the goal-post bonus.
HOW TO FIX
Route 10-20% of traffic to the own-form via load balancer rule. Monitor accept rate hourly for 72h. If accept rate holds or improves: expand to 50% next week. If accept rate drops: investigate, fix, retry.
P1 HOLD G02GOAL POST CHECK · JUL/AUG 60-DAY RUN RATE AVG
→ comp incentive paid Aug
 MO W4+ 2026-08-31 10-45K CAD F02
WHY THIS MATTERS
Mo's comp incentive (10K / 25K / 45K CAD goal-post tiers) measured on July + August 60-day net revenue avg vs locked baseline.
HOW TO FIX
Track weekly net revenue against baseline. End-of-Aug compute average vs each goal-post threshold. Document for invoice.
P2 HOLD G03THREAD C · RECRUITMENT PARTNERSHIPS · POST-PILOT OUTREACH
→ Stage 2/3 conversation · NOT in June scope
 MO W4+ 2026-09-01 $1-3M/YR —
HOW TO FIX
Post-pilot conversation. Buyer table already built in recruitment_buyer_table.md. Outreach plan in pitch docs. 10 partners × $100K target. Hold until post-Aug.
▎ RECURRING · ONGOING CADENCE
P2 WIP R01DAILY EXEC BRIEF · $-MOVED-THIS-WEEK FORMAT
→ 15-min morning
 MO DAILY — REPORTING —
HOW TO FIX
Auto-generated daily brief format: "This week net rev: $X. WoW: ±Y%. Today's action: Z. Open question for you: W." 4 lines max. Sent to Hassan + Lauren via Slack.
P2 WIP R02WEEKLY FRIDAY READOUT · 3-BULLET STRUCTURE
→ what shipped · what's coming · flags
 MO FRI — REPORTING —
HOW TO FIX
See C08 for first Friday template. Three bullets: what we did + what we moved + what's coming. Send by 4pm Friday weekly.
P2 WIP R03WEEKLY COMPETITIVE AD LIBRARY SCAN
→ EverQuote · Zebra · Insurify · SmartFinancial · Mil Wallet
 MO WEEKLY — INTEL —
HOW TO FIX
Open facebook.com/ads/library. Search by page name for the 6 competitors. Note: format mix, top-running concept, hook style, anything new. 15 min weekly. Surface anything notable in Friday readout.
P2 WIP R04FREQUENCY MONITORING · ODOMETER 05/15 · CAP TRIGGER 4.0
→ currently 3.27 · ~14 days from fatigue
 MO DAILY — RISK MONITOR —
HOW TO FIX
Daily check in Ads Manager → Auto Max Val → Frequency. When Odometer 05/15 frequency hits 4.0: pause and let new variants absorb impressions. Track trigger threshold and act.
P3 WIP R05EVENTS MANAGER DIAGNOSTICS WEEKLY CHECK
→ catch new parameter issues early
 MO WEEKLY — HYGIENE —
HOW TO FIX
5-min weekly check of Events Manager → Diagnostics tab. Note any new flags. Surface in Friday readout if material.
▎ COMPLETED · DONE THIS SESSION
P0 DONE X01FULL META API AUDIT · ACCOUNT 1034037341617540
→ master_audit.md + 5 appendices + delta
 MO DONE 2026-06-04 FOUNDATIONAL —
OUTPUT
audit/_audit_combined.html — 12 markdown files synthesized. Live API data validated against Mo's prior 20 audit iterations.
P1 DONE X02CROSS-VALIDATION · 15/18 PRIOR CLAIMS MATCH WITHIN 5%
→ 3 corrections + 8 new findings
 MO DONE 2026-06-04 VALIDATION —
OUTPUT
audit/delta_findings.html
P1 DONE X03TUESDAY MEETING HANDOUT · PRINTABLE 2-PAGE BRIEF
→ tuesday_meeting_handout.md
 MO DONE 2026-06-04 PREP —
OUTPUT
audit/tuesday_meeting_handout.html
P1 DONE X04AUTO MAX VAL DECISION DEEP-DIVE
→ 3 options + math + framing
 MO DONE 2026-06-04 PREP —
OUTPUT
audit/auto_max_val_decision.html
P1 DONE X05EVENTS MANAGER WALKTHROUGH · STEP-BY-STEP
→ 15-min Tuesday AM
 MO DONE 2026-06-04 PREP —
OUTPUT
audit/events_manager_walkthrough.html
P1 DONE X06PAGE ENGAGEMENT AUDIENCE BUILD SPEC
→ copy-paste into Ads Mgr
 MO DONE 2026-06-04 PREP —
OUTPUT
audit/page_engagement_audience_build.html