---
title: Weekly Meta media-buying cadence and attribution settings
slug: meta-weekly-cadence-and-attribution
summary: Launch new ads on one fixed day each week and judge after 7 days; use 7-day click for evergreen campaigns and 1-day click for flash sales; starter bid/attribution defaults; how teams run weekly pillar and hypothesis reviews; when third-party attribution tools are worth it.
domain: marketing
tags: [meta-ads, attribution, workflow, reporting, bidding]
related: [meta-incremental-attribution-model, meta-breakdown-effect-judge-in-aggregate, meta-creative-testing-by-budget, meta-consolidation-vs-sprint-approach, meta-ads-manager-metrics-setup, meta-ad-diagnostic-cheatsheet]
sources: [yt:kn4ByanAe-s, yt:CCsty8R0UaA, yt:aCkulCPAYzw, yt:oO_9CefhuDE, yt:cnWpC1g5qek, yt:vaBxYgZ7MAU, yt:r51Tj5qQGa0, yt:-BYAC0ALW78, yt:dJlkNgpXhVQ, yt:Xx_hBs9Re8s]
updated: 2026-10-24
---

# Weekly cadence and attribution

## TL;DR
- Pick **one launch slot per week** (e.g. Monday 10 am): launch all new ads/tests then, wait **7 days**, then optimise, scale and launch the next batch. Monitor daily, but don't "day-trade".[^1][^2]
- The more consistent the cadence, the more stable the account (Smart Marketer).[^3]
- **Attribution:** 7-day click for evergreen campaigns; **1-day click** for flash sales, BFCM and anything with a hard end date (it also forces faster learning).[^4][^5]
- Starter defaults (Dara Denney): bid strategy **highest volume**, attribution **7-day click / 1-day view**.[^6]
- **Incremental attribution** is now a third option (optimisation + reporting) — see [[meta-incremental-attribution-model]]; for scaling decisions read 1-day click or incremental results.[^17][^18]
- Master Ads Manager first; add Triple Whale/Northbeam only when you invest in several channels (TV, podcasts, influencers, Google).[^7]

## Details
**Why weekly.** Molly Pittman says Meta's AI learnings refresh every seven days, so launching every day or two "isn't using the platform the way it was built".[^2]

**Slower cadences.** Ben Heath's version for new/low-volume accounts: fixed review cycles of 5, 7 or 10 days, making all changes in one go so the ad set re-enters learning once (learning usually lasts 24–48 h); at a few conversions a week, decisions can take months.[^19] Jon Loomer launches a batch, waits a week (plus a week after any creative test), and reviews new batches after about **two weeks**, judging the ad set in aggregate.[^20]

**Weekly creative meeting (Mars Men).** Mondays: review what won last week across dashboards (Motion, landing pages, creative, spend by channel), then rank a backlog of pillar ideas. Each idea is sanity-checked (compliance, Google Trends — is anyone searching for it yet?) and either shelved or briefed; seasonal moments (Father's Day → "fatherhood" pillar with whitelisting partners) get pulled forward.[^8][^9][^10] Spend-velocity alerts ($2K in a few days) trigger action any day.[^11]

**Hypothesis log (Evan Lee, Motion).** Treat every ad as a hypothesis at individual, team or org level; mark each as validated, invalidated or "needs more time" and keep a log. Run weekly/bi-weekly reviews from the active hypotheses, and event reports comparing e.g. BFCM year over year to plan the next one.[^12][^13][^14][^16]

**Tracking gap (2023).** A Smart Marketer cross-account check found Meta under-reporting by ~25% on average versus their third-party tool — so Ads-Manager-only buyers may be under-spending.[^15]

## Disagreements / caveats
- **Default window:** CTC's canon uses a 7-day-click optimisation window by default.[^21]
- **View-through:** Dara's starter setting includes 1-day view; Smart Marketer standardises on 7-day click. View attribution inflates reported ROAS for awareness-heavy ads — pick one and stay consistent when comparing.
- The 25% under-reporting figure is from 2023 (post-iOS 14) and one agency's tool.

## See also
- [[meta-creative-testing-by-budget]]
- [[meta-ads-manager-metrics-setup]]
- People: [[people/molly-pittman]], [[people/zach-stuck]], [[people/evan-lee]]

[^1]: yt:kn4ByanAe-s @ 34:34 — Molly Pittman
[^2]: yt:kn4ByanAe-s @ 36:03 — Molly Pittman
[^3]: yt:kn4ByanAe-s @ 39:23 — Molly Pittman
[^4]: yt:kn4ByanAe-s @ 36:57 — Molly Pittman
[^5]: yt:kn4ByanAe-s @ 37:47 — Molly Pittman
[^6]: yt:CCsty8R0UaA @ 01:12 — Dara Denney
[^7]: yt:CCsty8R0UaA @ 19:48 — Dara Denney
[^8]: yt:aCkulCPAYzw @ 28:26 — Zach Stuck
[^9]: yt:aCkulCPAYzw @ 29:17 — Zach Stuck
[^10]: yt:aCkulCPAYzw @ 30:07 — Zach Stuck
[^11]: yt:aCkulCPAYzw @ 19:25 — Zach Stuck
[^12]: yt:oO_9CefhuDE @ 13:28 — Evan Lee
[^13]: yt:oO_9CefhuDE @ 15:02 — Evan Lee
[^14]: yt:oO_9CefhuDE @ 56:51 — Evan Lee
[^15]: yt:cnWpC1g5qek @ 40:59 — Molly Pittman
[^16]: yt:oO_9CefhuDE @ 1:00:07 — Evan Lee
[^17]: yt:Xx_hBs9Re8s @ 02:30 — Jon Loomer
[^18]: yt:dJlkNgpXhVQ @ 02:22 — Jon Loomer
[^19]: yt:vaBxYgZ7MAU @ 11:37 — Ben Heath
[^20]: yt:r51Tj5qQGa0 @ 04:03 — Jon Loomer
[^21]: yt:-BYAC0ALW78 @ 06:37 — Tony Chopp