Paid Ads Specialist

---
name: paid-ads-specialist
description: Plan and optimize paid advertising — Google, Meta, and multi-channel campaigns, budgets, creative testing, and ROAS.
category: business-marketing
---

## Overview

Paid ads buy speed: immediate traffic and measurable results, at a price. This skill covers full-funnel paid media — channel selection, campaign structure, audience targeting, creative testing, budget management, and optimization toward ROAS/CAC goals. Vendor-neutral principles; channel specifics noted where they matter.

## When to use

- Launching paid campaigns from scratch
- Choosing between Google, Meta, TikTok, LinkedIn, etc.
- Fixing campaigns with high spend and low return
- Scaling winning campaigns
- Structuring accounts for testing and learning
- Setting budgets and ROAS/CAC targets

- Auditing an agency's or predecessor's ad account
- Planning seasonal or holiday campaign pushes
- Deciding between in-house and agency management
- Scaling winning campaigns profitably
- Launching on new ad platforms (TikTok, Reddit)
- Fixing tracking after iOS privacy changes
## Core concepts

**Channel-intent match.** Search ads capture existing demand (high intent, high CPC); social ads create demand (interruption, needs strong creative); LinkedIn for B2B targeting (expensive, precise); TikTok/Shorts for young audiences and creative-led discovery. Match channel to where your buyers are and what intent looks like.

**Campaign structure.** Separate by objective and funnel stage: prospecting (broad/interest audiences) vs. retargeting (site visitors, engagers) vs. brand search (defensive). Keep structures simple enough to manage — complexity without volume just fragments learning.

**Creative is the targeting.** As platforms automate targeting, creative quality drives performance. Test hooks, formats (UGC-style video, static, carousel), and angles systematically. Refresh creative before fatigue sets in (watch frequency and CTR decay).

**Landing page congruence.** Ad promise → landing page delivery must match exactly. Message mismatch is the #1 conversion killer in paid traffic. Dedicated landing pages beat homepage sends.

**Measurement.** Platform-reported conversions overstate; use UTMs + analytics + CRM data for truth. Track CAC and ROAS by campaign, not just clicks. Account for attribution windows and view-through claims skeptically.

**Budget strategy.** Start with test budgets (enough for statistical significance — typically 10–20x target CPA per ad set), kill losers fast, scale winners gradually (20–30% budget increases to avoid resetting learning).


**Full-funnel structure.** Prospecting (cold audiences, broad messaging) → consideration (engaged audiences, social proof) → conversion (high-intent, offer-led) → retention (existing customers, upsell). Budget follows the funnel: most spend on prospecting, highest ROAS on retargeting — both are necessary.

**Creative fatigue signals.** Frequency rising + CTR falling + CPA rising = fatigue. Refresh creative before performance collapses: new hooks on winning concepts, new formats, new angles. Build a creative pipeline, not one-off ads.

**Attribution in a privacy-first world.** iOS ATT, cookie deprecation, and ad blockers broke deterministic tracking.
Adapt with: conversion APIs (server-side tracking), modeled conversions, incrementality testing, and media mix modeling for big budgets.
Stop optimizing to flawed pixel data alone — triangulate.
**Creative testing systems.** Test matrix: hooks (first 3 seconds) × angles (pain points) × formats (UGC, demo, testimonial) × CTAs.
Produce 10–20 new creatives weekly for scale accounts; kill losers fast (3x target CPA spend), scale winners aggressively.
Creative is the new targeting — broad audiences + great creative beats narrow targeting.
**Landing page synergy.** Message match (ad promise = page headline), speed (< 3s load), single CTA, social proof above fold, mobile-first.
Sending paid traffic to generic pages wastes spend — dedicated landing pages per campaign theme.
Test landing pages as rigorously as ads; they are half the equation.
## Practical workflow

1. **Set targets.** Define target CAC or ROAS from unit economics. This is the optimization north star — every decision references it.
2. **Pick channels.** Start with 1–2 where intent and audience align. Search for existing demand; one social channel for demand creation. Expand only after one works.
3. **Build the foundation.** Conversion tracking (pixel + server-side where possible), UTM taxonomy, dedicated landing pages per major campaign, creative bank (5–10 variants per concept).
4. **Launch structured tests.** Campaign per objective; ad sets testing one variable (audience OR creative, not both). Run 1–2 weeks minimum before judging.
5. **Optimize weekly.** Kill ad sets above target CAC with statistical confidence. Double down on winners. Refresh creative showing fatigue. Tighten or expand audiences based on data.
6. **Scale carefully.** Increase winning budgets incrementally, expand to lookalikes/broad once core works, add channels sequentially. Watch blended CAC — scaling always costs more per acquisition eventually.

**Creative testing matrix:** 3 hooks × 2 formats × 2 CTAs = 12 variants. Test hooks first (biggest lever), then format, then CTA. Keep winners, iterate on angles.


**Weekly optimization routine:** 1) check spend pacing vs. budget, 2) kill ad sets 2x over target CPA with 3x+ target spend, 3) scale winners (budget +20–30%), 4) review search terms (search) / placement reports (social), 5) check creative fatigue metrics, 6) verify tracking (conversions firing correctly).

**Account audit checklist:** tracking verified (test conversions) → campaign structure clean (no overlapping audiences) → budget allocation matches goals → creative freshness (no fatigue) → search terms / placements reviewed → bid strategies appropriate → attribution settings correct.
Audit quarterly or when performance shifts suddenly.
**Scaling protocol:** verify profitability at current spend (7+ days stable) → increase budgets 20–30% every 3–4 days → expand audiences (lookalikes, broad) → add creative variations → monitor CPA daily.
Scale gradually — doubling budgets overnight resets learning and spikes CPA.
## Common pitfalls

- **No conversion tracking.** Optimizing for clicks instead of revenue. Fix tracking before spending.
- **Sending traffic to the homepage.** Generic pages convert poorly. Match landing pages to ad intent.
- **Judging too early.** Killing tests before statistical significance. Let tests run.
- **Scaling too fast.** Doubling budgets overnight resets learning and spikes CAC. Scale gradually.
- **Creative fatigue.** Running the same ads for months while CTR decays. Refresh on a schedule.
- **Trusting platform attribution.** Platforms take credit generously. Triangulate with your own analytics.
- **Ignoring landing page.** Blaming ads for poor conversion when the page is the problem. Test both.
- **Too many campaigns.** Fragmented budgets that never exit learning. Consolidate until each ad set gets meaningful volume.
- **Optimizing for the wrong event.** Training algorithms on page views instead of purchases. Always optimize for the deepest reliable conversion event.
- **Ignoring post-click.** Blaming ads when the landing page converts at 1%. Audit the full journey before judging the traffic.
- **Premature scaling.** Increasing spend on 3 days of good data. Statistical noise bankrupts impatient advertisers — wait for significance.
- **Ignoring creative fatigue.** Running the same ads for months. Refresh creative before performance decays, not after.