Marketing Analytics and Attribution

Data Tells You Where to Invest

Every marketing dollar should have a measurable output. Without analytics, you're making budget decisions on intuition. With proper attribution, you can trace revenue back to specific channels, campaigns, and content pieces — then reinvest in what works and cut what doesn't.

Your Analytics Stack

Google Analytics 4 (GA4) — website traffic, user behavior, conversion events

Google Search Console — organic search performance, keyword rankings, crawl health

Google Ads Dashboard — paid search metrics, Quality Scores, search term reports

Meta Ads Manager — paid social performance, audience insights, creative analytics

Plausible or Fathom — privacy-first alternatives to GA4, simpler interface

PostHog — product analytics for SaaS: feature adoption, retention, user journeys

Email platform analytics — open rates, click rates, revenue attribution

Setting Up GA4 for Marketing

GA4 is event-based — every user action is an event. Key events to configure:

// Key conversion events to track
gtag('event', 'sign_up', {
  method: 'email',
  source: document.referrer,
});

gtag('event', 'trial_start', {
  plan: 'pro',
  value: 49,
  currency: 'USD',
});

gtag('event', 'purchase', {
  transaction_id: 'T123',
  value: 299,
  currency: 'USD',
  items: [{ item_name: 'Annual Pro Plan' }],
});

UTM Parameters

UTM parameters tag your URLs so GA4 knows where traffic came from:

https://yourdomain.com/signup
  ?utm_source=google
  &utm_medium=cpc
  &utm_campaign=brand-q1-2025
  &utm_content=headline-a
  &utm_term=your+brand+name

Standardize your naming convention and apply it to every link in every campaign. Without this, GA4 can't tell you which campaign drove which conversions.

Your Weekly Marketing Dashboard

Build a dashboard that answers these questions every week:

Attribution: Why It's Complicated

Most customers interact with 5–7 marketing touchpoints before converting. Attribution models try to assign credit across those touchpoints:

Last-click (Google default): gives 100% credit to the final touchpoint. Undervalues awareness campaigns.

First-click: gives 100% credit to the first touchpoint. Undervalues retargeting.

Data-driven: Google and Meta's AI models that distribute credit based on actual contribution patterns. Most accurate, but requires 50+ conversions/month to function properly.

For SaaS with longer sales cycles, data-driven attribution is the most trustworthy. Until you have enough conversion volume, first-click and last-click together give you a reasonable range.

Reporting Cadence

Daily — Ad spend vs budget, major anomalies (traffic drop, conversion spike)

Weekly — Channel performance dashboard, top and bottom performing content, ad creative performance

Monthly — Strategic review: what drove growth? What underperformed? Where should next month's budget go?

Quarterly — Channel strategy changes, budget reallocation, new channel tests based on accumulated data

Acting on Data

Analytics without action is just entertainment. Build decision rules:

• If organic CTR from Search Console drops below 3% for a page → update title tag and meta description
• If a Meta ad set exceeds target CPA by 30% for 3 consecutive days → pause and investigate
• If email open rate drops below 25% → A/B test subject lines
• If trial-to-paid CVR drops below 15% → investigate onboarding flow

Key Takeaways

• A complete analytics stack requires GA4 + Search Console + Ads dashboards + email platform + product analytics
• UTM parameters are non-negotiable — without them, GA4 cannot attribute conversions to specific campaigns
• Data-driven attribution is most accurate but requires 50+ monthly conversions; use first-click and last-click together until then
• A weekly dashboard covering 8–10 key metrics is more actionable than complex reporting
• Build decision rules: specific metric thresholds that trigger specific actions — this turns analytics into a system

---

Try It Yourself: Design a marketing analytics dashboard for a SaaS product. Specify 12 metrics, the tool that provides each, the healthy benchmark, and the action you'd take when a metric falls below its threshold. Include metrics across organic, paid, email, and product analytics.