Marketing ROI and Attribution: Measuring Which Channels Drive Revenue
Master marketing ROI: understand attribution models, calculate CAC by channel, and optimize spend for best returns.
Key Takeaways
- Attribution = which channel gets credit for customer acquisition; problem: customer touches multiple channels before signing up (sees ad, reads blog, attends webinar, talks to sales); which channel caused conversion? Multi-touch attribution distributes credit (first-touch 40%, last-touch 40%, middle 20%), or linear (equal credit all channels); single-touch (last-click only) oversimplifies and favors retargeting/sales over awareness
- CAC by channel varies 10-100x: Organic search £100-500 (lowest, earned), content marketing £300-800, paid ads £500-3K, events £1-5K, sales £5-20K (highest); allocate budget to lowest CAC channels (organic > content > ads > events); but CAC shouldn't be only metric (organic might have lower quality, higher churn); track CAC + payback + LTV by channel
- Marketing ROI = revenue from channel ÷ marketing spend; example: Google Ads £10K spend → £50K revenue → 5x ROI (£5 revenue per £1 spend); if payback 12 months and CAC £1K, customer LTV £10K = ROI 10x over lifetime (but 5x year 1); focus on unit economics by channel, not just short-term ROAS
Understanding Marketing Attribution
Attribution answers the question: Which marketing channel gets credit for this customer acquisition? **The Attribution Challenge** Typical customer journey: Day 1: Sees Google Ad → Clicks, lands on website Day 3: Reads blog post (organic search) Day 10: Downloads whitepaper (email nurture) Day 15: Attends webinar Day 20: Talks to sales rep Day 25: Signs contract Which channel caused the signup? All of them (at different points). **Attribution Models** 1. **Last-Touch Attribution** (easiest, most biased) - Credit final touchpoint only - In above example: Sales gets 100% credit - Problem: Undervalues awareness (ads, content) - Common in: CRM systems (where last interaction tracked) 2. **First-Touch Attribution** (opposite problem) - Credit initial touchpoint only - In above example: Google Ads gets 100% credit - Problem: Undervalues nurture and sales - Common in: Attribution tools that track first source 3. **Linear Attribution** (fair, middle ground) - Equal credit to all touchpoints - In above example: Each of 6 touchpoints gets 16.7% - Problem: Assumes all channels equally valuable - Common in: Balanced analytics approaches 4. **Time-Decay Attribution** (middle touches matter) - Recent touches weighted higher - Example: First touch 10%, middle 20%, last touch 70% - In above example: Sales 70%, webinar 20%, others 10% - Problem: Complex, hard to explain - Common in: Advanced marketing analytics 5. **Custom Attribution** (org-specific) - Define rules specific to your business - Example: Awareness 30%, consideration 30%, decision 40% - Ad touchpoint → 30% of credit - Content touchpoint → 30% - Sales touchpoint → 40% - Problem: Biased toward what you think is important - Common in: Enterprise companies **Recommended: Linear or Time-Decay** - Fair to all channels - Doesn't oversimplify (as last-touch does) - Easier to explain than custom **CAC by Attribution Model** Same company, different attribution: Marketing spend (monthly): - Google Ads: £5,000 - Content: £2,000 - Events: £3,000 - Sales: £10,000 - Total: £20,000 Customers acquired: 10 (£10K ACV each) **Last-Touch Attribution (biased):** - All 10 customers credit to "sales" (last touchpoint) - Sales CAC: £10,000 ÷ 10 = £1K per customer - Other channels appear worthless (£0 CAC) **Linear Attribution (fair):** - Each touchpoint 25% credit (4 channels = 1/4 each) - Google Ads: 2.5 customers × £5K spend = £2K CAC - Content: 2.5 customers × £2K spend = £800 CAC - Events: 2.5 customers × £3K spend = £1.2K CAC - Sales: 2.5 customers × £10K spend = £4K CAC **Insights from linear attribution:** - Content is most efficient (£800 CAC) - Sales is least efficient (£4K CAC) - But sales helps close deals (necessary in cycle) Action: Increase content budget (highest ROI), optimize sales efficiency. **Multi-Touch Attribution Tools** Advanced tools model customer journey: - Google Analytics 4: First-touch, last-touch, time-decay - Mixpanel: Custom attribution rules - Marketo/Pardot: Lead scoring (implicit attribution) - Salesforce: CRM touchpoint tracking - Stripe/Churn: Revenue attribution These tools track: - Which source brought customer in - All subsequent touchpoints (emails, ads, content) - Time between touches - Final conversion Then distribute credit per attribution model. **CAC Payback by Channel** Different channels have different payback periods: | Channel | CAC | Monthly value | Payback months | |---------|-----|----------|----------| | Organic search | £400 | £200 | 2 months | | Content marketing | £600 | £200 | 3 months | | Paid ads | £1000 | £200 | 5 months | | Sales | £3000 | £500 | 6 months | | Events | £2000 | £300 | 6.7 months | **Organic search** has fastest payback (customers highly qualified). **Events** has slowest (high CAC, but might have higher LTV). Best strategy: Combination approach - Organic search (base, SEO takes time to build) - Sales (enterprise, long cycle, high value) - Paid ads (fill gaps, scalable) **Marketing Mix Optimization** Goal: Allocate budget across channels for highest total ROI. Constraint: Total budget £20K/month Scenario 1: All into organic search (best CAC) - Problem: Takes 12-24 months to build (not fast) - Miss near-term revenue Scenario 2: All into paid ads (fastest) - Problem: Expensive (£1K CAC vs. £400 organic) - Revenue high, but burn high - May not be sustainable Scenario 3: Blended (realistic) - Organic search: £3K (long-term, build SEO) - Paid ads: £7K (fill gaps, fast scaling) - Events: £5K (mid-market, high ACV) - Sales: £5K (enterprise, existing pipeline) - Result: Balanced CAC, sustainable growth **Measuring Marketing ROI** ROI = (Revenue from channel − Marketing spend) ÷ Marketing spend Example: Google Ads: - Spend: £10,000 - Revenue from customers acquired: £50,000 (5 customers × £10K ACV) - ROI: (£50K − £10K) ÷ £10K = 4.0x (£4 profit per £1 spent) Content marketing: - Spend: £5,000 - Revenue: £40,000 (4 customers × £10K ACV) - ROI: (£40K − £5K) ÷ £5K = 7.0x (£7 profit per £1 spent) **Content has better ROI** (7x vs. 4x). But ROI is year-1 only. Over customer lifetime: If customers stay 3 years: - Google Ads customer LTV: £30K (3 × £10K) - ROI over lifetime: (£30K − £10K) ÷ £10K = 2.0x - This is lower return (because customer doesn't expand) Content marketing customer LTV: £35K (expansion revenue) - ROI over lifetime: (£35K − £5K) ÷ £5K = 6.0x - Higher ROI (because customer expands more) **Best ROI channels tend to have:** - Lower CAC (efficient acquisition) - Higher retention/LTV (loyal customers) - Better product-market fit (right customer) **Avoid vanity metrics:** - Impressions (how many people saw ad) - Clicks (how many clicked) - CTR (click-through rate) Focus on: - CAC (cost per customer) - Payback (months to recover) - LTV/CAC (lifetime profit per acquisition spend) - ROI (£ returned per £ spent) Vanity metrics look good but hide poor economics.