Video summary

Subir video de cbo

Main summary

Key takeaways

Business

Summary (Business/Ops + Execution Focus)

What this is

  • A Meta CBO (Campaign Budget Optimization) playbook for running one campaign that tests whether an offer + ads have market demand.
  • Designed for teams with tighter budgets that need more sales stability, but it requires patience.
  • Core idea: let Meta optimize, but force spending fairness across ad sets so the system doesn’t starve promising audiences or overfund the wrong ones.

Who it’s for / When to use

  • For teams with slightly tighter budgets who want more stable sales.
  • Works for teams in Latin America and the U.S. (with references to Brazil).
  • Requires:
    • Patience: don’t deoptimize daily.
    • Humility: if metrics indicate “loss,” remove the offer—even if you like it.

Campaign Structure (Single CBO)

Recommended structure

  • Use 3–5 ad sets (audience splits).
  • Use exactly 3 ads per ad set (avoid going above; “a little less is okay”).
  • Do not mix creative formats within the same campaign:
    • If you use videos, include all videos.
    • If you use images, use a separate CBO (don’t mix video + image in the same setup).

Audience targeting

  • Public targeting is allowed except “sex” must be segmented.
  • Exclude certain countries described as “typical excluded” (examples mentioned: Nicaragua, Venezuela, Brazil).

Budget Math (Key Formula)

  • Total campaign budget = 1.5% × your product commission (commission-based sizing).
    • Example: if commission is 8, budget = 8 × 1.5 = 12.
  • Minimum spend is used to force spending distribution across ad sets:
    • With 3 ad sets: set 10% minimum per set (total forced = 30%).
    • With 5 ad sets: set 6% minimum per set (total forced = 30%).

The “Minimum Spend” Mechanism (Safeguard)

Problem observed

  • Without minimums, Meta may allocate spend unevenly—not necessarily to the ad sets you’d pick as winners.

Solution

  • Force a baseline spend in each ad set:
    • Force 30% of the campaign budget across ad sets (e.g., 10% + 10% + 10%).
    • Leave 70% for Meta to distribute based on performance.

Why this works (practical reason)

  • Meta must “use” the forced portion in each set.
  • Then you allow Meta to place the remaining budget using enough early signal.

Timeline + Optimization Rule (Major Constraint)

  • Run the test for 3 days.
  • Don’t change anything daily—optimization is over the 3-day learning window, not “per day.”

Operational exception

  • You can turn ads off at night if spend/cost is very high.
  • Reactivate in the morning if still within limits.

What to Analyze After 3 Days (Decision Rules)

Primary objective

  • Determine whether offer + ads produce demand (sales conversion).

Decision logic

  1. If the offer is losing after 3 days
    • Check supporting metrics (e.g., initiated payments, CPC, etc.).
    • If metrics are “wrong” overall → kill the offer (“goodbye offer”).
  2. If it’s a draw / break-even-like
    • Optimize creative angles and scale with improvements.
  3. If you already have profit / good performance
    • Start scaling budget for the next day.

Scaling constraint

  • Do not scale blindly during the 3-day test.
  • Scaling happens after the 3-day window.

Scaling Playbook (After the 3-Day Test)

Budget increases (next day after confirming performance)

  • Increase by 20%–30% if performance is acceptable.
  • If it performs well:
    • Increase by up to 50%
    • Then continue iterating upward.
  • If performance starts trending down, lower slightly.
  • Always respect an ideal CPA ceiling:
    • Scale only as long as it doesn’t exceed your ideal CPA.

Concrete Examples / Outcomes Mentioned

Platform demo example

  • Structure: 1 campaign, 3 ad sets, 3 ads per set (9 ads total).
  • Budget shown: 102 soles (~$30).
  • Commission inferred: ~$20 (because $30 / 1.5 ≈ 20).
  • 3-day results:
    • Day 1: 1 sale, ROAS ~0.33 (“terrible”).
    • Later: ROAS improved toward break-even by the end of 3 days.
    • Overall outcome described as not-loss / optimized after waiting.

Why minimums mattered in the demo

  • In the demo, without minimum spend, an ad set didn’t spend.
  • This is exactly what minimums are intended to prevent.

Additional performance claim (conversion example)

  • “46 payment sent, eight sales” used as an example of a strong conversion level.

How to Implement Minimums (Practical Build Steps)

Minimum spend creation approach (build method)

  • Build your CBO with:
    • Campaign level CBO budget = commission × 1.5
    • 3 ad sets
    • Duplicate ads so each set has 3 ads
    • Set minimum spend per ad set so the forced portion totals 30%:
      • For 3 sets: each set minimum = 10%
      • For 5 sets: each set minimum = 6%
  • When scaling by adding more ads/sets:
    • Duplicate and name correctly
    • Verify minimum is applied (it can “get lost” during duplication)
    • Turn off ads that exceeded the CPA threshold.

Frameworks / Playbooks Explicitly Implied

  • CBO learning window playbook
    • Test for 3 days → then decide → then scale
  • Spending fairness constraint
    • Force 30% baseline across ad sets via minimum spend; allow 70% freedom
  • Creative variance control
    • 3 ads per set; distinct creative angles; avoid mixing formats in the same CBO

Key Metrics / KPIs Referenced

  • Sales (primary success signal for the 3-day test)
  • ROAS
    • Example: ROAS ~0.33 on day 1, later improving toward break-even over 3 days
  • CPA / ideal CPA
    • Used for kill vs keep decisions and for scaling ceilings
  • CPC
    • Used as a supporting metric (especially when interpreting “loss”)
  • Payments initiated
    • Treated as a positive early sign even if early sales/ROAS look bad
  • Spend vs CPA ceiling
    • If an ad set/ad spends too much relative to target CPA, it should be turned off

Presenters / Sources

  • Miguel (speaker/instructional lead throughout)
  • Aaron (referenced as sharing another approach; “usual” method is cited though not shown in the transcript)

Original video