4.3 · Social ads (Meta Ads)

Level 4 · Growth & advertising

4.3Social ads (Meta Ads)

Goal: create Facebook/Instagram Ads campaigns (objective, targeting, budget, creatives) and track performance.
Estimated time: 14 min · self-paced reading · If you only have 5 min: read “The essentials in 30s” + the diagram.

The essentials in 30 seconds

  • Strict structure: campaign (objective) → ad sets (audience + budget + placements) → ads (visual + copy + CTA) — 1 objective = 1 campaign.
  • Broad + retargeting: broad discovers, retargeting (visitors, carts, 50%+ video views) converts — Pixel and Conversions API installed and tested before the first euro.
  • Test small, scale cold: 5 €/day x 3 creatives x 7 days, keep the winning CPA, raise +20%/48h after 7 stable days — never x10.

Running case — Nour tests 20 €/day

With social proof in hand, Nour goes paid: a 'sales' campaign with one broad ad set + one cart retargeting, 3 UGC creatives at 5 €/day each for 7 days. Verdict by CPA: 12 €, 9 € and 22 €. She kills #1 and #3, puts 15 € on #2 and raises +20% every 48 h — CPA stable at 9 €. The Pixel, tested before the first euro, lets her steer instead of guess.

Continued from 4.1-4.2 — Nour moves from organic to measured paid.

1. The structure: campaign → ad sets → ads

Campaign (objective: traffic, leads, sales), ad sets (audience + budget + placements), ads (visual + copy + CTA). 1 objective = 1 campaign: mixing traffic and sales in one campaign means flying blind.

Nour creates a 'sales' campaign and nothing else inside: the single objective lets the platform optimise delivery for that exact goal.

2. Targeting: broad + retargeting

Broad (the algorithm finds, strong creatives speak) + retargeting (visitors, abandoned carts, 50%+ video views): retargeting converts, broad discovers.

Pixel and Conversions API installed and tested before the first euro: with no measurement, no optimisation. The blue 'boost' button with no structure means flying blind.

3. Test small, scale cold

5 €/day x 3 creatives x 7 days: keep the winning CPA (cost per result), kill the rest, raise +20%/48h (not x10: the algorithm panics). Scaling is earned: stable 7 days before raising.

First budgets = learning: you test which creatives and targeting work with an amount you can afford to learn with, then concentrate the budget on the winner.

Common misinterpretation: 'Hitting boost on a post is enough, and when it works you multiply by 10.' No. Blind boosting mixes everything with no structure (objective, audience, measurement) and burns the budget; and a brutal x10 panics the algorithm. Real method: 1 objective per campaign, Pixel tested first, small multi-creative test, then +20% every 48 h.
Profitable ads 1 objective per campaign, broad + retargeting, +20% Structure (1 objective) Campaign→ ad sets → adsTarget (broad +retarget) Discover, then convertScale (+20%) Test 5 € x 7d, winning CPA
1 objective per campaign, broad + retargeting, +20%
Text version of the diagram (SEO / screen readers / translation): 1) Structure: Nour creates 1 'sales' campaign (1 objective = 1 campaign), broad + retargeting ad sets, 3 UGC ads. 2) Target: broad discovers, cart retargeting converts — Pixel tested before the first euro. 3) Scale: 5 €/day x 3 creatives x 7 days → CPA 12 €, 9 €, 22 €; kill #1 and #3, 15 € on #2, +20%/48h → stable 9 € CPA. Lesson: testing designates, budget follows.

🔎 Concrete example — 20 €/day that sells (Nour's case)

Testing designates, budget follows.

'Sales' campaign: broad ad set + cart retargeting
3 UGC creatives: 5 €/day each, 7 days → CPA: 12 € / 9 € / 22 €
→ kill #1 and #3, 15 € on #2, +20%/48h → stable 9 € CPA

Reading: testing designates, budget follows.

Lesson: Pixel tested first: no measurement, no optimisation.

Section vocabulary

Social advertising (social ads)
Paid advertising on social networks, multiplying reach in a targeted, measurable way.

Simple example: Nour's 20 €/day reaching a targeted audience beyond her followers.

What it is not: organic alone when reach caps and goals demand speed.

Reflex: combine solid organic + targeted paid per objective.

Meta Ads
Meta's advertising platform (Facebook + Instagram), run via Ads Manager; very powerful targeting.

Simple example: Nour's 'sales' campaign: broad + retargeting, 3 UGC creatives.

What it is not: the blue 'boost' button with no objective or measurement.

Reflex: always go through Ads Manager, never blind boost.

Targeting
Precise definition of an ad's audience (demographics, interests, behaviours, custom and lookalike audiences).

Simple example: Nour: broad that discovers + cart and 50%+ video-view retargeting that converts.

What it is not: everyone with no distinction and the same message.

Reflex: broad to discover, retargeting to convert.

Creative
The ad's content (visual and video + copy + call to action); the decisive performance element.

Simple example: Nour's 3 UGC creatives: their CPA gap (12, 9, 22 €) decides.

What it is not: a generic visual with no message or call to action.

Reflex: test several creatives, keep the CPA winner.

Campaign objective
The targeted goal (awareness, traffic, engagement, leads, conversions) for which the platform optimises delivery.

Simple example: Nour picks 'sales': the platform optimises for buyers, not likers.

What it is not: mixing traffic and sales in the same campaign.

Reflex: 1 objective = 1 campaign, or flying blind.

In practice — Create a Meta Ads campaign (Nour's case)

  1. Set the campaign objective (awareness, traffic, engagement, leads, conversions): the platform optimises for that goal.
  2. Set targeting: broad to discover + retargeting (visitors, carts, video views) to convert.
  3. Install and test the Pixel (and Conversions API) BEFORE the first euro, then launch 5 €/day x 3 creatives x 7 days.
  4. Track CPA per creative, kill the losers, concentrate on the winner and raise +20%/48h after 7 stable days.
You create and steer Meta Ads campaigns (objective, targeting, budget, creatives, tracking): a powerful lever to multiply reach and hit precise goals, in a targeted, measurable way.

Ready-to-say script — validating your test before scaling (30 s)

'Our 3 creatives ran 7 days at 5 €: CPA 12, 9 and 22 €. We cut #1 and #3, put 15 € on #2, and only raise 20% every 2 days if CPA stays stable. No x10: we scale cold.'

Say it in the weekly check-in: figures first, decision next — testing designates, budget follows.

Copy-paste template — Meta Ads test tracking

CAMPAIGN OBJECTIVE: _______________
AD SETS (broad + retargeting): _______________
PIXEL TESTED: yes / no (date: ___)
CREATIVE 1 (budget/day + CPA at D+7): _______________
CREATIVE 2 (budget/day + CPA at D+7): _______________
CREATIVE 3 (budget/day + CPA at D+7): _______________
DECISION (kill / keep / scale +20%): _______________

Small budget

5 €/day is enough to learn: 1 campaign, 2-3 creatives, 7 days — the goal is the CPA verdict, not volume.

Local shop

Nour narrows broad to her city + shop-visitor retargeting: closeness drops the CPA.

E-commerce

Abandoned-cart retargeting first: that is where CPA runs lowest and returns fastest.

Total beginner

Boost your best organic first (proof), only then create a real campaign: proof first, budget next.

✓ Best practices

  • 1 objective per campaign, never mix.
  • Pixel tested before the first euro spent.
  • +20%/48h after 7 stable days, never x10.

✗ Common pitfalls

  • Blind blue-button boosting with no structure.
  • Launching with no Pixel (flying blind, no optimisation).
  • Multiplying the budget by 10 overnight.

Key takeaways

  • Organic reach is limited; social advertising multiplies reach in a targeted, measurable way.
  • Meta Ads (Facebook + Instagram) is the most-used platform, thanks to very powerful targeting.
  • A campaign rests on: objective, targeting, budget and duration, and creatives (the ad content, the decisive element).
  • Strengths: fine targeting, controllable budget (start small), total measurability; but it must be learned (done badly, you burn your budget).

48h challenge — apply tomorrow

Check that your Pixel is installed and tested, then write your test plan (1 objective, 3 creatives, 5 €/day x 7 days) with the template above. 20 minutes is enough.

Section exit checklist
  • I can name the 3 levels (campaign, ad sets, ads).
  • I know what CPA is and how to call a test.
  • My Pixel is tested and my test plan written.
Self-rating 1-5: do I structure 1 objective per campaign?do I scale cold (+20%) instead of x10?

Frequently asked questions

Should you pay for advertising, or can you succeed with organic (free) alone?

You can do a lot in organic (community, awareness, engagement), but organic reach declines and free growth is slow. Advertising multiplies: reach, precise targeting, speed, measurable goals (traffic, leads, sales). The right mix: build organic (content, community) + amplify with paid what works.

How much should you spend to start advertising on social networks?

You can start small (a few euros a day, controlled daily budget). Treat the start as learning: test several creatives and targeting, concentrate on what works by CPA, then raise gradually while watching profitability. Example here: 5 €/day x 3 creatives x 7 days.

Check your understanding

Why advertise (Meta Ads)?

Additional resources (optional, to go further)

Level 4 examFresh, harder questions to test what you've learned