Summary of "How I Made £2.7M with Facebook Ads in 90 Days (Full Breakdown)"

Case study outcome (Facebook Ads, e-commerce)

Client result in 90 days / Q4 campaign overhaul: ~£2.7M additional sales attributed to fixing Facebook/Meta ads.
Business context (puzzle notebook niche):
- AOV (Average Order Value): £27 (low AOV = need very low cost per purchase)
- High margin / low production cost → break-even ROAS achievable at ~1.4–1.5x
- Seasonality: 60–70% of sales in Q4 (Christmas gifting; impulse/present purchase behavior)

Revenue: £3.69M
Increase vs Q4 2023: +£2.7M
ROAS: ~4.68x (more than 2x improvement)
Cost per purchase: £5.91 (from ~£14.12; >50% lower)
AOV: ~£27 (roughly unchanged)
Units sold: 133,000 (from 35,000; ~+98,000 units)
Meta ad spend: £788k (from £499k; ~+£288k, ~+55% spend)
Implied scaling efficiency: ~300%+ revenue growth with ~55% spend increase
Constraint / limiter: Ran out of best-seller stock, limiting further scaling
- Estimate: could have surpassed £5M+ revenue if inventory had been sufficient
Macro factor: CPM up ~20% in Q4 2024 vs Q4 2023, making the gains even more notable

Performed an ad account audit to find weaknesses.
Emphasized that Meta attribution is less reliable (privacy + attribution limits), so accurate tracking is critical.
Tracking tool recommended: Hyros (conversion/attribution beyond Meta’s view)
- Example cited: campaign spend ~£2k showed ~£15k revenue in Hyros vs ~£3k in Meta—highlighting hidden/late conversions.

Actionable process

Audit tracking first → verify true ROAS / true cost per purchase before scaling.

Problem: too many campaigns → budget spread thin → Meta can’t optimize effectively and signal gets noisy.
New structure: consolidated into only two campaign types:
- Testing campaign
- Scaling campaign
Removed/avoided “extra” layers that didn’t drive ROAS (e.g., awareness/traffic style efforts for this business).

Why this worked for their niche

With low AOV (£27), they needed a single-step funnel.
For impulse/present behavior, multi-step warmups/retargeting were too expensive at low AOV.

Framework-like rule

Optimization speed depends on conversion volume—“It’s not time, it’s conversion volume.”

Prior issue: most ads were image-based (video wasn’t failing—creative quality/style likely wasn’t strong enough).
Core winning format: UGC (user-generated content) video ads
- More recommendation-like/trustworthy feel than overly produced ads
- Fits impulse purchase behavior
Added carousel ads:
- People rarely buy only one puzzle notebook; they often buy a range
- Carousels helped users compare options and increased effective AOV via product discovery in-feed

Creative mix for scaling

Instead of betting on one format, they used a mix:
- Majority budget on UGC video
- Some image spend retained
- Some carousels
Rationale: different users respond to different creatives; Meta allocates budget based on engagement likelihood.

Examples of video ad styles they listed to drive experimentation:

Because most sales occur in Q4 within a condensed window, they needed high creative variation.
Goal sequence:
1. Get cost per purchase below £12
2. Scale volume
Creative volume prevents performance drop when audience size is limited and exposure increases quickly.

Consolidate campaigns to concentrate budget and improve optimization clarity (e.g., Testing vs Scaling).
Use a single-step funnel when AOV is low and purchases are impulsive/seasonal.
Treat creative testing as a volume game during short peak periods to avoid ad fatigue.
Test strong video properly—don’t assume “video failed” due to low-quality early tests.
Use UGC-style video for impulse/giftable products; add carousels when buyers purchase bundles/ranges.
Fix attribution/tracking so you scale based on true conversion economics, not Meta’s incomplete view.

Presenter: The video speaker (also referencing “Heath Media” and their own agency/service).
Named company/tool: Hyros (tracking attribution software).
Named agency: Heath Media (Meta/Google Ads agency).