Cross-channel creative grid

Admaxxer is a DTC analytics platform with built-in Meta + Google ad ops. The same creative — say, a 15-second product demo — can be a winner on Meta and a loser on Google for completely rational reasons. The short answer: Meta measures social response (CTR, engagement, creative fatigue); Google measures intent match (Quality Score, expected CTR at query); and a creative that performs on one but not the other tells you something specific about why.

TL;DR

• Meta signal = social proof, scroll-stop, fatigue curve (best on UGC-style and demo video)
• Google signal = intent match, keyword fit, landing page relevance (best on product-forward creative)
• A creative that wins Meta but flops Google usually means targeting mismatch, not creative quality
• A creative that wins Google but flops Meta usually means it's too literal/direct-response for social feed
• Use the Admaxxer creative grid to see cross-platform performance on the same asset

What each platform measures

Meta: social response

Meta's algorithm is tuned for scroll-stop, engagement, and fatigue. Signals that Meta weights:

• First-second view rate (does the creative stop the scroll?)
• 25%, 50%, 75% completion (video holding power)
• CTR and post-engagement (comments, shares, saves)
• Creative fatigue (same user seeing the ad >5 times)
• Relevance score (post-launch quality signal)

Creatives that win on Meta: UGC testimonials, before/after transformations, punchy first-3-seconds hooks, culturally-relevant humor. Creatives that flop: static product shots without context, long product descriptions, anything that requires >3 seconds to "get."

Google: intent match

Google Search algorithm weights:

• Keyword-to-ad relevance (does the ad match the query?)
• Landing page relevance (does the page deliver what the ad promises?)
• Expected CTR at that query (will users click vs. competitors?)
• Ad Strength score (for Responsive Search Ads)

Creatives that win on Google: product-forward images with clear USP in headline, price or offer in description, specific benefit in extension. Creatives that flop: vague lifestyle photography, emotional ads without concrete product mention, UGC-style creative (lacks the clarity Google Search users expect).

TikTok: native-format response

TikTok is closer to Meta than Google in that it measures scroll-stop and completion, but weights native format heavily — creative that looks like an ad gets penalised. See the TikTok integration docs when that connector ships.

Diagnostic steps

Step 1: Pull the cross-channel creative grid

In the Admaxxer dashboard, navigate to Creatives → Cross-channel grid. Each creative shows its Meta CPA, Google CPA, and the ratio. Look for:

• Both low (winner — double down)
• Meta low, Google high (investigate Google targeting)
• Google low, Meta high (investigate Meta placement or hook)
• Both high (kill)

Step 2: Check ratio vs. baseline

A creative with Meta CPA of $32 and Google CPA of $45 isn't necessarily failing Google — Google baseline CPAs are often 30–50% higher than Meta for similar products because Search traffic is bottom-funnel and Meta is mixed. Compare to the account-level baseline for each platform.

Step 3: Inspect the specific mismatch

For a "Meta wins, Google loses" creative:

• Is the Google ad text too close to Meta ad copy? Rewrite for query intent.
• Is the landing page fit wrong? Google rewards pages that match the ad headline verbatim.
• Is the keyword set too broad? Narrow to bottom-funnel queries.

For "Google wins, Meta loses":

• Is the Meta hook strong enough? First 3 seconds must stop scroll.
• Is the creative format wrong (static vs. video)? Meta Feed favors video; Stories favors 9:16.
• Is the placement limiting reach? Test Advantage+ placements.

Step 4: Test same creative across placements

Run the winning Meta creative as a Google Display asset (not Search — different intent). If it performs on Display, the creative itself is good but doesn't match Search intent. If it flops on Display too, the creative has a deeper targeting or positioning issue.

Step 5: Check the Admaxxer MMM attribution

The MMM contribution model decomposes channel attribution independently. If a creative's MER-level contribution is positive while platform ROAS says it's losing, the creative is doing halo work that shows up in blended MER but not platform-attributed ROAS.

Divergence patterns

Three common patterns and their diagnoses:

Pattern 1: Meta CPA $30, Google CPA $75, same creative Hypothesis: the creative is social-native (UGC, humor, story-driven) and doesn't match Search intent. Google Search users want answers, not vibes. Fix: rewrite Google ad copy to be query-matched; move the Meta creative to Google Display with demographic targeting instead.

Pattern 2: Meta CPA $85, Google CPA $35, same creative Hypothesis: the creative is product-forward and clinical — great for intent-matched Search but boring in a social feed. Fix: repurpose the product-forward creative as a Meta Collection ad or Catalog Shopping ad where "boring but clear" works. Don't force it into Feed video slots.

Pattern 3: Meta CPA $45, Google CPA $45, both near target Hypothesis: creative is versatile. Double down with more budget on the lower-fatigue channel first. Test slight variations on each channel to find the ceiling.

The Admaxxer Claude agent can call query_metrics with a creative-level breakdown and flag these patterns automatically.

Common mistakes

Three mistakes we see:

1. Using identical creative assets on Meta and Google Search. Search users are in a different mindset. A Feed-optimized hook will often fail in a Search expanded-text slot. Always create platform-specific variants.
2. Ignoring landing page fit per platform. Google checks for ad-to-landing-page keyword relevance; a Meta-optimised landing page (long, story-driven) often fails Google's Quality Score.
3. Comparing CPA without normalising for intent. A $60 Google Search CPA for "buy organic protein powder" is not the same as a $60 Meta prospecting CPA. Search is closer to conversion; Meta spend has more pre-click demand-generation work.

See Blended MER vs ROAS for why creative-level CPA alone doesn't tell the full story — some creatives drive Meta awareness that Google harvests.

FAQs

Should the same creative run on every platform? No. Platform-specific variants outperform universal creatives by 20–40% on CPA. Cut each variant for the native format and intent.

How do I know if a creative is fatiguing on Meta? Frequency >4 with a CTR drop of >20% from peak. The Admaxxer creative grid shows fatigue score per creative.

What about TikTok and YouTube? TikTok follows Meta-like dynamics (social, scroll-stop). YouTube is mixed — pre-roll is Meta-like, search is Google-like. Treat each placement separately.

Can I use AI to generate platform-specific variants? Yes — the Admaxxer Claude agent can draft platform-specific copy given a base creative brief. It's a starting point, not a final asset.

How does the creative grid integrate with my media mix? The Admaxxer grid pulls spend and conversion data from Meta and Google connectors, joins on creative fingerprint (hash of image/video + copy), and surfaces cross-platform performance in one view. Works best when creatives have consistent naming across platforms.