CPC ranges are illustrative; they vary wildly by industry, geo, and auction dynamics. Use them for relative planning only.
| Platform |
Audience & formats |
Targeting highlights |
Typical CPC (indicative) |
Best for |
| Google Ads — Search, Display, Shopping, YouTube |
High-intent search; visual/video reach |
Keywords, audiences, in-market, custom intent, PMax |
Search: mid–high; Display: lower |
B2B/B2C; dev tools (search + YouTube tutorials) |
| Meta — Facebook, Instagram |
Broad consumer; visual storytelling |
Interests, lookalikes, retargeting, Advantage+ |
Mid (varies by vertical) |
B2C; prosumer; creative-heavy categories |
| LinkedIn Ads |
Professional, job-title granular |
Company, title, seniority, Matched Audiences, ABM lists |
High |
B2B; enterprise; hiring/HR adjacencies |
| Twitter / X Ads |
Newsy, tech, creator-adjacent |
Keyword, follower lookalikes, engagement retargeting |
Mid–high (volatile) |
Launch moments; dev/tech audiences (validate current policy/tools) |
| TikTok Ads |
Younger skew; short video |
Interest, behavior, Spark Ads (organic-style) |
Mid |
B2C; viral creative testing |
| Reddit Ads |
Community-niche |
Subreddit, interest, keyword |
Low–mid |
Niche B2C/B2B; authentic tone required |
Paid acquisition flow
Campaign structure
| Level |
Role |
Best practices |
| Account |
Billing, access, global settings |
Separate brands or regions; shared libraries where appropriate |
| Campaign |
Objective, budget, geo, schedule |
One primary goal per campaign; align naming to funnel stage |
| Ad group / ad set |
Audience + creative theme + placements |
Tight themes; avoid mixing unrelated intents |
| Ad |
Copy, creative, extensions |
3–5 variants per set; refresh on fatigue signals |
Naming and tracking: Use a consistent convention (region_objective_audience_vN) and UTM parameters (or platform click IDs) so web analytics, CRM, and warehouse models agree on source/medium/campaign. Document parameter ownership so product and marketing do not fork conventions.
Creative testing checklist
| Check |
Question |
| Hook |
Does the first 1–2 seconds state the problem or outcome? |
| Proof |
Is there a concrete stat, logo, or demo frame? |
| CTA |
Is the next step obvious and consistent with the landing page? |
| Format |
Static, carousel, short video — matched to platform norms? |
| Fatigue |
When did this creative last beat a challenger on CPA or CTR? |
Targeting strategies
| Strategy |
When it shines |
| Keyword targeting |
Search: capture existing demand; use negatives aggressively |
| Audience targeting |
Demographics, interests, custom segments from first-party data |
| Lookalike / similar |
Scale after a seed list of converters |
| Retargeting / remarketing |
Recover consideration-stage users; cap frequency |
| Contextual |
Brand-safe placements; topic alignment |
| Account-based (LinkedIn) |
Named accounts; sales + marketing alignment |
Bidding strategies
| Model |
Meaning |
When to use |
| CPC |
Pay per click |
Learning phases; tight keyword control |
| CPM |
Pay per thousand impressions |
Awareness with strong creative testing |
| CPA / cost per conversion |
Optimize toward conversions |
Stable pixel/SDK and volume |
| ROAS target |
Revenue per ad spend |
E-com and clear LTV signals |
| Maximize conversions |
Platform-driven volume |
After baseline tracking is trusted |
Landing page optimization
- Message match: Headline and hero mirror the ad promise and keyword intent.
- Single primary CTA: Reduce competing actions above the fold.
- Social proof: Logos, quotes, usage stats — truthful and specific.
- Form optimization: Fewer fields, smart defaults, progressive profiling post-signup.
- A/B testing: Tie experiments to hypothesis and pre-registered success metrics.
| Dimension |
Organic |
Paid |
| Goal |
Community, trust, retention signals |
Scale reach, retargeting, conversions |
| Cadence |
Sustainable publishing + engagement |
Flight-based with creative rotation |
| Risk |
Algorithm shifts |
Cost inflation, policy changes |
Platform selection (simple frame): For each candidate platform, score audience fit × content you can ship consistently × business model (PLG vs sales-led). Drop platforms where two of three are weak.
- Content calendar: Mix education, proof, behind-the-scenes, and community highlights; align peaks with product launches.
- Community management: Response SLAs, escalation paths, and moderation standards belong in ops docs — not only marketing.
| Metric |
Definition / use |
| Reach |
Unique users who saw content |
| Engagement rate |
Interactions ÷ reach or followers (define denominator consistently) |
| Share of voice |
Brand mentions vs competitors (sampled) |
| Follower growth |
Net new; quality > raw count |
| Social traffic |
Sessions and assisted conversions from social referrers |
| Social conversions |
Signups/purchases attributed to social touchpoints |
Paid + organic social flywheel
Attribution models
| Model |
Logic |
Strength / weakness |
| Last click |
100% credit to final touch |
Simple; undervalues upper funnel |
| First click |
Credit to discovery |
Highlights acquisition; ignores nurture |
| Linear |
Equal credit across touches |
Fairer blend; can dilute signal |
| Time decay |
More credit to recent touches |
Balances recency and path |
| Position-based |
Heavy first + last, light middle |
Hybrid; still somewhat arbitrary |
| Data-driven |
ML allocation (where data volume allows) |
Powerful when models are trusted and audited |
Budget allocation
- Testing budget: Fixed slice for new channels, audiences, and creative formats before scale decisions.
- Scaling rules: Many teams increase spend 20–30% when ROAS or CPA holds above target for a defined window — avoid sudden 2× jumps that reset learning.
- Seasonality: Retail peaks, fiscal-year enterprise cycles, and conference seasons should pre-allocate creative and landing updates.
- Incrementality: Where possible, run geo or audience holdouts and platform lift studies — last-click dashboards often over-credit paid social and under-credit organic + brand.
Compliance and brand safety
Ad platforms and regions impose privacy, targeting, and copy rules (cookies, sensitive categories, political ads, financial services). Maintain a short policy sheet per channel: prohibited claims, required disclaimers, and approval workflows so launches are not blocked at the last mile.
Anti-patterns
| Anti-pattern |
Why it hurts |
| Broad targeting without learning data |
Burned spend, no insight |
| No negative keywords (search) |
Irrelevant clicks and polluted learning |
| Creative fatigue ignored |
Rising CPA, declining CTR |
| Vanity metrics as north stars |
Reach without pipeline or retention |
| No attribution / conversion tracking |
False confidence in channel mix |
External references
Index: Marketing channels — index · Marketing map: Marketing body of knowledge
Keep project-specific marketing plans in docs/product/marketing/ and GTM documents in docs/product/, not in this file.
