Free Growth Diagnostic·$1,500 Value
All Case Studies
7-Figure Luxury Home Brand
Google · YouTube · Meta · Cross-Channel Rebuild

Luxury Home Brand. Cross-channel build: Google, YouTube, Meta. Peak ROAS 7.0×.

3.1× 7.0×
Google-only Entry vs Peak Cross-Channel ROAS  ·  18-Month Period Average Was Lower
3.1× Entry ROAS · Search Only
+126% Peak Improvement
18 Months Build Period
Channel Architecture  ·  Nov 2024 – May 2026  ·  3 Coordinated Channels
Dimension Google Ads YouTube Meta Ads
Funnel Role Conversion Demand Creation Demand Development
Audience Strategy Customer Match + Smart Bidding on verified purchaser data Cold audiences; viewer lists fed back to Google PMax TOF: LTV-weighted Klaviyo lookalikes; BOF: engaged site visitors
Creative Format Search, Shopping, Performance Max In-stream + non-skippable; lifestyle framing only TOF + BOF; no discount messaging across all formats
Data Flow Out Conversion signals inform cross-channel attribution Viewer lists → Google PMax audience signals Email + visitor data → Klaviyo → Google Customer Match
Budget Priority Primary: high-intent capture; defends existing demand Secondary: brand consideration, not direct response Secondary: prospecting + retargeting at managed scale
Metric Entry Period · Google Search Only Peak Period · Cross-Channel System Documented Change
Blended ROAS 3.1× 7.0× +126%
Channels Active Google only Google + YouTube + Meta +2 channels
First-Party Data Use None uploaded Customer Match + PMax signals Activated
Build Period Nov 2024 entry Peak May 2026 18 months

Performance data verified across Google Ads, Meta Business Suite, and YouTube Studio. 7.0× reflects peak blended ROAS; period average was lower. Brand identity withheld per client NDA.

Google Search was well-managed and converting most of the available search demand when we took over, at 3.1× blended ROAS. The limitation wasn't inside Google. No channel was building the next pool of buyers who would eventually show up in Search.

YouTube and Meta were added as pipeline, not as direct conversion channels. Viewer lists became Performance Max audience signals. First-party email data from all three channels flowed back into Google as Customer Match. Over 18 months, peak blended ROAS reached 7.0×. Period average was lower.

A well-managed Google account at its ceiling

Better bidding and tighter match types had already been done. The 3.1× wasn't a management problem. It reflected a single-channel account that had captured most of the available search demand. Nobody was building the pool of buyers who would become next quarter's branded search traffic.

The client was skeptical that YouTube and Meta could move Google's blended ROAS for a $2K–$10K product. That skepticism was fair. It took 18 months of data to work through.

Three gaps the account didn't know it had

01

PMax was targeting cold traffic: no customer data had ever been uploaded

Years of purchase history sat in Klaviyo. None of it had been uploaded to Google. Performance Max was starting from zero on every campaign with no reference for what a real buyer looked like. We fixed this before spending on YouTube or Meta. Any new channel we launched would compound into an account already improving.

02

Nothing above the funnel: the pool of future searchers wasn't being replenished

No YouTube, no Meta. The client's skepticism was reasonable. At this AOV, we weren't expecting direct conversions from video or social either. The point of adding upper-funnel channels was different: building the audience that would eventually show up in Google Search already familiar with the brand. That connection between upper-funnel exposure and later branded search traffic isn't always tracked, but it accumulates.

03

The channels weren't sharing data: each platform was running independently

YouTube viewer lists were connected to Google Ads as Performance Max audience signals. First-party email data from all three channels was uploaded to Google as Customer Match on a rolling basis. Without those connections in place, the channels would have run independently. That's what we built.

If Google is your only channel, you're working with a fixed pool of demand, with no visibility into whether it's shrinking. The Growth Diagnostic identifies untapped upper-funnel demand and what it would take to connect it back into Google's bidding.

See what it covers →
Free Growth Diagnostic  ·  No Pitch  ·  Yours to Keep

If you recognized your account in those findings, a diagnostic session surfaces the same structural gaps and gives you a prioritized plan to address them.

Apply for Free →

How the three channels divided demand creation, development, and capture

YouTube built awareness, Meta developed it by prospecting cold audiences and retargeting warm ones, and Google captured the high-intent search traffic that resulted. The data layer connecting all three is what made the system compound rather than coexist.

1
Create Demand
YouTube
In-stream + non-skippable formats
Lifestyle creative: no discount messaging
Brand consideration, not direct response
Data out: Viewer lists → Google PMax audience signals
2
Develop Demand
Meta Ads
TOF: LTV-weighted Klaviyo lookalikes
BOF: Engaged site visitor retargeting
Same no-discount creative across all formats
Data out: Email + visitor data → Klaviyo → Google Customer Match
3
Capture Demand
Google Ads
Search: high-intent brand + category terms
PMax: YouTube audience signals + Customer Match
Shopping: feed-driven product coverage
Converts audiences pre-warmed across stages 1 and 2
First-Party Data Foundation
Purchase history from Klaviyo (segmented by 6-month LTV)  ·  YouTube viewer lists  ·  Site visitor data uploaded to Google Ads periodically as Customer Match signals and audience inputs for Performance Max and Smart Bidding

What we built and in what order

We didn't launch YouTube and Meta first. Google got fixed first, so new channels would compound into an account already improving.

Seeded PMax with verified customer history before launching any new channel

The brand's full purchase history was in Klaviyo. We segmented by LTV, uploaded the top cohort to Google as Customer Match, and connected it to Performance Max as an audience signal. Google had a real reference point for what a buyer looked like before we spent a dollar elsewhere. The cold-start phase shortened and the baseline improved first.

Launched YouTube with brand lifestyle creative and a deliberate no-discount brief

No promotional messaging, no price anchoring. In-stream and non-skippable formats, lifestyle framing only. YouTube viewer lists fed back into Google Ads as Performance Max audience signals: not a direct conversion path, but a reference for people who had already shown interest in the brand.

Built Meta TOF and BOF anchored to LTV-weighted Klaviyo lookalikes

The TOF lookalike seed was the Klaviyo segment weighted by 6-month buyer LTV, not a flat email list. BOF retargeted engaged site visitors. Both layers ran the same no-discount creative as YouTube. Email and visitor data from all channels flowed into Klaviyo and was re-uploaded to Google as Customer Match on a rolling basis.

Aligned attribution windows across all three platforms to the actual purchase decision timeline

Attribution windows were set to 7 days by default across all three platforms. At this price point and purchase timeline, assists were dropping out before the sale registered: YouTube assists and Meta view-throughs were not appearing in reports. We extended the windows to reflect the actual decision timeline so numbers across channels were comparable.

The lift in Google's ROAS came from who was clicking, not from changes inside Google's campaign structure. As YouTube and Meta built more pre-warmed audiences, those audiences started showing up in Search. Customer signals from all three channels were uploaded to Google via Customer Match. That's what moved the blended number over 18 months.

Channel Architecture

Three channels, each with a defined role in the same system

Stage 01 YouTube Create Demand
Format
In-stream and non-skippable. Lifestyle creative only. No promotional messaging, no discount anchoring at this stage.
Feeds into Viewer lists exported to Google Ads as Performance Max audience signals
Stage 02 Meta Develop Demand
Structure
TOF: LTV-weighted Klaviyo lookalikes (6-month buyer cohort). BOF: engaged site visitors. Same no-discount creative as YouTube.
Feeds into Email and visitor data flows through Klaviyo, re-uploaded to Google as Customer Match
Stage 03 Google Capture Demand
Campaigns
Search and Performance Max. PMax anchored with YouTube viewer lists and Customer Match data from all three channels.
Result 3.1× Google-only → 7.0× peak cross-channel ROAS across 18 months
Data Layer

Klaviyo held purchase history segmented by 6-month LTV. That data seeded YouTube viewer lists, Meta lookalikes, and Google Customer Match simultaneously. Each channel drew from the same first-party foundation; the signals did not run in isolation.

Eighteen months later: 3.1× Google-only to 7.0× peak cross-channel ROAS

7.0× Peak Cross-Channel ROAS
3.1× Entry ROAS (Search Only)
+126% Documented ROAS Improvement
18 mo. Build Period

No single change produced the 7.0× peak. The +126% compounded across 18 months as YouTube built the audience, Meta developed it, and Google converted it. The data layer connecting all three is what made the channels reinforce each other, and what made Google's efficiency improve without any change to its own campaign settings.

7.0× reflects peak blended ROAS; period average was lower and returns varied across months. Results depend on category, AOV, and the state of first-party data infrastructure before expanding channels. Brand identity withheld per client NDA.

Phil Chan Head of Growth MESAscale
View on LinkedIn
About the strategist

Phil brings more than a decade of paid-media experience across agency, enterprise, and ecommerce environments. Most of that time was spent in senior account leadership, directing multi-channel portfolios at seven-figure monthly scale.

He now works directly with established ecommerce brands through MESAscale, focusing on Google, Meta, and YouTube. His work centres on account architecture: where spend is being lost, how cross-channel coordination is failing, and which structural changes compound over time.

10+ Years in Paid Media Enterprise + Ecommerce Google · Meta · YouTube
Free Ecommerce Growth Diagnostic

Find out whether your channels are reinforcing each other or running in parallel.