How to launch paused campaigns across Meta, TikTok and Google at once

A multi-platform launch is the same intent expressed three or four times: campaign, ad set / ad group / ad squad, budgets, geos, naming — rebuilt by hand in each ad manager, where one slipped digit becomes real spend. The fix isn't speed; it's structure: codify the convention, build everything paused, review one preview, flip live deliberately. This guide is that workflow.

Codify before you automate

A launch convention worth automating fits on one page:

• Naming — {GEO}_{Objective}_{Platform}_{MonthYear} or your variant, identical across platforms (this is also what makes prefix-scoped reporting work later)
• Budget levels — where budget lives (campaign vs ad set), defaults per market tier
• Structural defaults — placements, optimization events mapped per platform, audience baselines
• States — everything created paused; go-live is a separate, human act

If two senior buyers would build the same brief differently today, write the page first.

What each platform actually needs built

• Meta — under the right act_: campaign (objective), ad sets (audience, placement, budget if ad-set level), ads attached to creatives
• Google — under the right 10-digit customer ID: campaign (network, bidding), ad groups, ads; UAC has its own asset-driven shape
• TikTok — advertiser_id → campaign → ad groups (optimization event chosen deliberately) → ads
• Snapchat — ad account → campaign → ad squads → ads, with swipe vs view objectives explicit

The cross-platform trap is objective mapping: "conversions" means differently-shaped things per platform. Map once, in the convention page.

The paused-launch workflow

1. One brief: markets, budgets, audiences, the creative pack, the prefix
2. Build per platform from the convention — every object named, budgeted, targeted
3. Everything paused — structurally, not as a reminder
4. One preview: every campaign/ad set/ad about to exist, listed with names and budgets
5. Review against the launch checklist; fix in the preview, not in production
6. Approve → objects created, still paused → audit entry written
7. Go-live: a human flips it on, market by market, at the chosen hour

The afternoon of click-work becomes a brief plus a review — and the review is where senior judgment belongs anyway.

What breaks in manual launches

Naming drift between platforms (reporting pays for it for months), daily-vs-lifetime budget mix-ups, a geo wrong on one ad set of nine, an objective set by muscle memory, and the classic: something left active overnight because the platform's default state won an argument with a tired human.

The launch QA checklist

• ✓ Every object name matches the convention (grep the preview for the prefix)
• ✓ Budgets sum to the brief, per market, with daily/lifetime explicit
• ✓ Geos and audiences verified per ad set — not per campaign
• ✓ All states paused before approval; go-live scheduled as its own step
• ✓ Audit entry confirms what exists matches what was approved

When not to automate launches

Genuinely novel structures — a first test of a new objective or format — deserve hand-building once, then codification. Automate the pattern, not the experiment.

How Opera runs it

Campaign launch automation is this workflow with the guardrails built in: native structures per platform, your convention applied, paused always, previewed, approved, logged — see campaign operations for the edit and approval model around it.

"Create paused prospecting campaigns across Meta, TikTok and Google using this creative pack — $150/day each, JO naming."