RGM-202 · Paid Social Mastery · Module 7 of 7

Operating Across Platforms

Cross-platform paid social is its own discipline. Budget allocation, creative adaptation, audience strategy, and unified measurement that scale beyond any single platform.

What you will learn10 sections
Meta+1 claimedGoogle+1 claimedTikTok+1 claimed1 real salein your CRMSummed dashboards over-report 150–300%

Why cross-platform operations matters

Cross-platform operations matter because every platform optimizes for itself and claims the same conversions, so running them as silos overspends and double-counts. A unified operating system — shared budget logic, creative pipeline, and one source of truth — is what turns several channels into one machine.

Most paid social programs are run platform-by-platform. A Meta specialist manages Meta. A TikTok specialist manages TikTok. They report into different dashboards, use different creative, attribute conversions differently, and rarely talk. The result is silo'd performance, duplicated effort, and a measurement system that systematically overstates platform contribution.

Cross-platform operating discipline solves this by treating paid social as a single capacity: a roster of channels with shared budget, shared creative inputs, shared audience strategy, and unified measurement. The competitive advantage is not in any single platform — everyone has access to the same Meta, TikTok, LinkedIn. The advantage is in how the pieces work together.

By the numbers The walled gardens each grade their own homework
Run platforms in silos and the numbers stop adding up — literally
150–300%
aggregate ROAS inflation from overlapping attribution windows across platforms.
30%+
gap between summed platform conversions and real CRM sales signals duplication.
56%
of US ad buyers leaning further into marketing-mix modeling in 2025 (IAB).
71%
of brands reducing reliance on user-level data — the driver behind MMM’s return.

Sources: Ruler Analytics · eMarketer / IAB.

Budget allocation across platforms

Allocate budget across platforms by marginal return and role, not by habit: fund the channel that returns the most on the next dollar, hold a test budget for emerging channels, and rebalance on a cadence — not in a daily panic.

Start with the question: what is each platform good at? Then allocate budget against the role.

PlatformStrengthsWeaknessesRole in mix
Meta (Facebook + Instagram)Broadest reach, best conversion API, robust commerce features, mature targetingSaturated audiences, rising CPMs, attribution under iOS 14.5+Always-on workhorse; 35–65% of paid social budget for most DTC and SMB
TikTokCheap impressions, creator-driven creative, fastest-growing inventory, strong for new product launchesLess mature measurement, narrower commerce features in some marketsAwareness + creative discovery; 15–35% of budget
LinkedInOnly first-party professional graph; B2B targeting precisionHigh CPMs, finite inventory for niche ICPsB2B essential; consumer brands skip
PinterestHigh-intent visual discovery; women 25–54 over-indexed; long content half-lifeLower volume than Meta/TikTok; weaker for impulse purchasesLifestyle, home, beauty, fashion, wedding; 5–20% of budget
SnapchatGen Z reach (especially 13–24); AR-first formatsLimited targeting depth; smaller mature-market audienceYouth brands, AR-native products; 2–10% of budget
RedditHigh-engagement communities; intent-rich subreddits; tech/finance/gaming audiencesBrand-safety perception, lower volume, scale ceilingNiche verticals, community-led brands; 2–8% of budget
X (Twitter)Real-time news, B2B influence nichesPerformance ad product instability post-2022Optional; test, don't commit

The 70-20-10 rule

A defensible default for most consumer brands: 70% to your proven primary (usually Meta), 20% to a strong secondary (TikTok for DTC, LinkedIn for B2B, Pinterest for lifestyle), 10% across experimental channels. Adjust quarterly based on incrementality, not last-click ROAS.

What changes the split

RGM Expert Trick
We allocate budget by marginal return, not last year’s split

Most cross-platform budgets are inherited percentages nobody re-earns. The right question is where the next dollar returns most, not where last year’s dollars went.

We watch marginal CPA/ROAS as we scale each platform and move money to the rising edge — budgets follow incremental return, not habit.

WHY IT’S RARE · Fixed platform splits are the most expensive default in media.
Interactive · move the budget Where does the next dollar actually belong?
Average ROAS tells you the past; marginal ROAS tells you where to spend next

Search has the higher average return here — but watch the marginals. Slide the split and find where total return peaks (it’s where the marginals meet).

Search
$15,000
marginal: 0.0 conv / $1k
Social
$15,000
marginal: 0.0 conv / $1k
Total conversions from a $30,000 budget0

Budget is optimal when marginal ROAS is equal across channels — often the lower-average, less-saturated channel deserves the next dollar. Source: Precis.

RGM EXPERT TRICK
Adapt the first second per platform; reuse everything after

‘Build once, adapt many’ gets mistranslated into ‘run the same cut everywhere’ on one extreme, or ‘produce everything from scratch per platform’ on the other — one underperforms, the other doesn’t scale.

My middle path: re-shoot only the first second or two — the native hook each platform demands — and reuse the shared body. The hook is what determines the scroll; the rest travels fine.

You get native performance with a fraction of the production cost, because you’re only remaking the part that has to be native.

WHY IT’S RARE · Most teams treat cross-platform creative as all-or-nothing. Re-cutting just the hook per platform is the efficient middle that captures most of the native lift for little extra work.

Creative: build once, adapt many

‘Build once, adapt many’ means producing a core concept then cutting it natively per platform — vertical sound-on for TikTok/Reels, professional and specific for LinkedIn — rather than running one master cut everywhere. Adaptation is mandatory; reuse-as-is is the mistake.

The single most expensive mistake in cross-platform operations is producing platform-specific creative from scratch every time. The right model: a creative concept gets produced once with a master asset, then adapted to each platform's native format and aspect ratio.

PlatformPrimary aspect ratiosNative style cues
Meta1:1, 4:5, 9:16 (Reels/Stories)Polished, branded, can mix UGC and brand-led
TikTok9:16 onlyUGC, creator-led, no overt branding in opening 3s, sound-on, captions burned in
LinkedIn1:1, 16:9, 4:5Insight-led, professional tone, captions essential (mute viewing)
Pinterest2:3 (1000x1500 ideal)Lifestyle, aspirational, text overlay common, vertical Pin templates
Snapchat9:16 onlyCasual, AR-friendly, Gen Z mood
Reddit1:1 most commonAuthentic, low-production-value, community voice; native ads rarely look like ads

Production workflow that scales

  1. Creative brief at concept level. Brief the idea, the proof, the CTA — not the format. One brief, many outputs.
  2. Shoot for the most demanding format. Vertical 9:16 with safe-zone framing. You can crop to 1:1, 4:5, 16:9 from a 9:16 master; you cannot go the other way.
  3. Native edits per platform. TikTok needs captions burned in; LinkedIn needs different headline; Pinterest needs text overlay. Build templates so the variants take hours, not days.
  4. Asset library with metadata. Every asset tagged by concept, persona, funnel stage, hook, format. Searchable. Reused across platforms when performance suggests it should be.
  5. Performance feedback loop. Best-performing concepts on Platform A become first hypotheses on Platform B — with platform-native adaptation.
RGM Expert Trick
We build creative once at the concept level, then cut native

Reframing a 16:9 video into 9:16 is not adapting — it’s how you get ignored on every platform at once. Each feed has its own native grammar.

We design the concept once, then cut platform-native edits: a TikTok hook, a LinkedIn caption-first version, a Meta thumb-stop open. One idea, many native expressions.

WHY IT’S RARE · ‘Resize and ship’ quietly halves the performance of good ideas.

Audience strategy across platforms

Across platforms, the same principles hold: first-party data and broad targeting plus strong creative, with exclusions to avoid paying twice for the same person. Don’t rebuild a different audience philosophy per channel.

You will use different mechanics on each platform but the underlying audience strategy should be consistent.

RGM Expert Trick
We trust no platform’s self-reported ROAS — they all claim the same sale

Every platform takes credit for the same conversion, so summing their dashboards conjures revenue you never earned. Believe them and you over-invest everywhere.

We anchor to one source of truth — a blended view or geo/holdout lift — and treat platform ROAS as a directional hint, never the scoreboard.

WHY IT’S RARE · Added up, the dashboards always ‘prove’ more revenue than the bank shows.
Benchmark Stop summing the dashboards
Added up, the platforms claim far more than your business actually earned

Each walled garden counts the conversions it touched. Sum them and you conjure revenue that never happened — then over-invest on the strength of a number that isn’t real.

Platforms, summed
~250% of real
Actual (CRM)
100% — the truth

Window overlap inflates aggregate ROAS 150–300%. Source: Ruler Analytics.

Attribution is not incrementality.
— Avinash Kaushik, Occam’s Razor — on marketing analytics

Claim: Because each ad platform attributes conversions to itself, the sum of platform-reported conversions routinely exceeds actual orders — over-claim grows with the number of channels running. Source: RGM analysis. Context: Anchor on one source of truth (GA4/orders) and MER, and use holdout tests to find which channel truly drives incremental sales.

INTERACTIVE TOOL Cross-platform over-claim calculator
How much are your platforms double-counting?
over-claim

Over-claim = (platform-reported − actual) ÷ actual. The bigger it is, the more your channels are re-slicing one pie. Also a standalone tool.

RGM EXPERT TRICK
Reconcile to MER weekly and treat the over-claim gap as a KPI

Teams stare at each platform’s ROAS and never notice that the platforms collectively claim 130% of actual revenue. The silos each look fine; the business is overspending.

Every week I sum platform-reported conversions, divide by actual orders, and track that over-claim ratio over time. When it climbs, I know credit is being re-sliced, not grown — and I lean harder on MER and a holdout.

The gap between claimed and real is itself a number worth watching; it tells you when your channels are fighting over the same pie.

WHY IT’S RARE · Almost no one tracks the over-claim ratio explicitly. Making it a weekly KPI exposes the double-counting that platform dashboards are designed to hide.

Unified measurement when every platform claims credit

When every platform claims credit, the sum of platform-reported conversions exceeds reality — often badly. Anchor on a single source of truth (GA4/orders), use blended metrics like MER, and confirm causation with holdout or geo tests rather than trusting any platform’s self-report.

The cross-platform measurement problem: if Meta last-touch attributes a conversion, and TikTok view-through attributes the same conversion, and LinkedIn click-attributes it, you have 3× the credit for 1× the revenue. Unified measurement resolves this.

Interactive · tap a layer Measure with three lenses, not one
No single method is right — the truth is where all three agree

Triangulation beats any one number. Tap each layer to see what it’s good and bad at.

Platform reporting — fast, biased

Real-time and granular, but every platform self-credits and double-counts. Use it for in-flight optimisation — never as the scoreboard.

Incrementality tests — causal truth

Geo or audience holdouts show what would have happened anyway. The tie-breaker when models disagree — expensive to run, so reserve for big bets.

Marketing-mix modeling — privacy-durable

Models the whole mix from aggregate data — no user-level tracking needed. Best for long-run budget allocation across channels and brand vs performance.

Governance: who owns what

Governance is who owns what: a clear RACI for budget, creative, tracking, and reporting so platforms don’t drift, tracking doesn’t silently break, and nobody optimizes a channel in isolation against the others.

Operating across platforms requires governance or you get chaos. A defensible structure:

RoleOwns
Paid Social Director / LeadCross-platform budget allocation, creative strategy, measurement framework, vendor relationships
Platform Specialists (Meta, TikTok, LinkedIn)Campaign structure, bid strategy, daily optimization on their platform
Creative Lead / ProducerMaster asset production, adaptation pipeline, asset library, brief intake
Analytics / MeasurementConversion tracking integrity, attribution models, incrementality tests, dashboards
Audience / Data OpsCustomer list syncs, segmentation logic, audience refresh cadence

Small teams compress these into 1–3 people, but the responsibilities still exist — just on one person's plate. Document them so nothing falls through cracks.

Interactive · tap a lane Who owns what, across platforms
Cross-platform programs fail on governance, not tactics — give every lane an owner

When three people can edit everything and no one owns measurement, results drift. Tap each lane to see where the line sits.

Strategy / Brand — owns the plan

Sets objectives, audience priorities, and the brand-vs-performance split. Everyone else executes against this — one owner, not a committee.

Creative — owns the concepts

Feeds the native-cut pipeline across platforms. Builds once at the concept level, then adapts per feed. Accountable for the asset flow, not the media settings.

Media buying — owns execution

Builds and runs campaigns, manages bids and budgets within the plan. Shifts spend toward marginal return — but doesn’t grade its own results.

Analytics — owns the truth

Single owner of measurement: dedupes across platforms, runs incrementality, maintains the MMM. The neutral scoreboard everyone trusts.

Finance — owns the constraint

Sets the budget and the efficiency targets, and signs off on reallocation. Keeps the program honest to the P&L.

Tools and tech stack

The cross-platform stack ties the channels together: a server-side tagging layer feeding every platform’s API, a creative asset system, a unified dashboard on a single source of truth, and a CDP or list pipeline for consented audiences.

Advanced playbook

The advanced playbook runs incrementality across the mix (geo holdouts, MMM), reallocates on marginal return, and treats the platforms as one portfolio — shifting budget to the channel creating demand, not just the one claiming the last click.

How to · step by step Operate paid social as one system, not five accounts
Six moves that turn scattered platforms into a coordinated program
  1. Plan budget by marginal return.Shift the next dollar to the highest marginal ROAS until channels converge — not last year’s split.
  2. Build creative once, cut native.One concept, platform-native edits — never a resized 16:9.
  3. Coordinate audiences and exclusions.Suppress overlap so you don’t pay to fatigue the same people on every feed.
  4. Triangulate measurement.Platform reporting for speed, incrementality for proof, MMM for allocation.
  5. Give every lane an owner.Strategy, creative, buying, analytics, finance — one owner for measurement.
  6. Rebalance on evidence, on a cadence.Review marginal returns regularly; validate big shifts with a lift test before committing.
What the discipline shows · triangulation
3 lensesplatform + lift + MMM1 truththat survives privacy

The brands measuring well in 2026 stopped trusting any single dashboard. They run platform reporting for speed, geo lift tests for causal proof, and MMM for whole-mix allocation — and when two disagree, the incrementality test settles it. Source: eMarketer.

Each platform shows positive ROAS but the business isn't growing — why?
Classic over-claim. Each platform attributes the same conversions to itself, so the silos look profitable while blended performance is flat. Switch to MER and run a holdout to find real incremental contribution.
Who should own tracking in a cross-platform setup?
One named owner, explicitly. Tracking that’s ‘everyone’s job’ breaks silently and nobody notices for weeks. Put it in a RACI alongside budget, creative, and reporting.
How often should I rebalance budget across platforms?
On a deliberate cadence (often weekly or bi-weekly) based on marginal return — not daily reactions to noise, which just resets learning and chases variance.

Common mistakes

Cross-platform mistakes: managing channels in silos, summing platform-reported conversions as if they were real, reusing creative unadapted, no single source of truth, and no owner for tracking — so it breaks unnoticed.

Operating checklist

A sound cross-platform operation has one source of truth, blended targets (MER), a server-side feed to every platform, a build-once-adapt-many creative pipeline, clear ownership, and periodic incrementality tests — run as one portfolio, not five silos.

Sources and further reading

  • Meta Business, TikTok for Business, LinkedIn Marketing Solutions, Pinterest Business, Snap Ads Manager, Reddit Ads — native documentation
  • Tinuiti, Wpromote, Common Thread Collective, Power Digital — agency cross-platform playbooks
  • Search Engine Land, Marketing Land, AdExchanger — industry coverage of cross-platform measurement evolution
  • Haus.io, Recast, Northbeam — incrementality and MMM methodology
  • Andrew Faris, Common Thread Collective — channel role frameworks for DTC
  • Mike Taylor, Vexpower — media mix modeling open-source frameworks
  • MASA (Mobile Attribution and Analytics Standards Association) — cross-platform measurement standards
  • The Drum, Marketing Brew, Modern Retail — cross-platform program case studies
  • WARC and Effie award case studies — cross-platform integrated campaign winners
  • IAB Cross-Platform Measurement Guidelines
  • RGM internal benchmarks — cross-platform budget allocation by vertical and funnel stage
CASE-method test

Prove it. Earn your passcode.

Ten questions, CASE method. Pass at 90% to unlock this module’s completion passcode — retake as many times as you like.