The Australian marketing ROI conversation.
Marketing ROI conversations are typically misleading because the metrics that get cited aren't the metrics that drive business outcomes. How to have honest ROI conversations, what the metrics actually mean, what they mislead about, what the questions are.
Three metrics that mislead
1. ROAS (Return on Ad Spend)
Platform-reported ROAS (Meta's reported ROAS, Google's reported ROAS) overstates incremental revenue dramatically, typically 2-4x. The platform takes credit for purchases that would have happened anyway. Real incrementality is much lower than reported ROAS.
Use instead: Server-side attributed revenue + hold-out incrementality testing. Honest cost-per-acquired-customer measurement.
2. Click-through rate
CTR is a vanity metric for high-consideration purchases. A high CTR on Meta with low conversion rate often reflects creative that's clickbait without being intent-aligned. Optimising for CTR can degrade actual outcomes.
Use instead: Cost-per-acquired-customer + payback period + customer lifetime value.
3. Cost per lead (raw)
Raw cost-per-lead misleads when lead quality varies wildly. A campaign producing 100 leads at each (90 unqualified) is worse than 50 leads at each (40 qualified). Lead-quality-adjusted measurement matters.
Use instead: Cost-per-qualified-lead + sales-pipeline conversion + booked revenue per lead source.
Three metrics that matter
1. Incrementality-adjusted CAC
What's the actual incremental cost-per-acquired-customer after subtracting the customers who would have purchased anyway? Hold-out testing produces honest measurement. Most agencies don't run hold-outs because the data is uncomfortable.
2. Customer Lifetime Value (with payback period)
What's the customer LTV vs the CAC? What's the payback period (months to recover CAC)? Long-term unit economics depend on this, not on month-to-month ROAS.
3. Cohort retention by acquisition channel
Customers acquired via different channels behave differently over time. Meta-acquired customers vs Google-acquired vs Klaviyo-organic vs referral, different cohort retention curves. Attribution-quality decisions depend on channel-cohort retention measurement.
Three honest ROI questions
For agencies + clients to ask each other:
- "What's the actual incremental CAC after hold-out incrementality adjustment?" If neither side can answer with hold-out-tested data, the conversation is opinion not fact.
- "What's the channel-cohort retention curve at month 6 + month 12?" Attribution-quality decisions depend on this. Most engagements skip the analysis.
- "What's the payback period vs the CAC vs the LTV at current scale?" Operationally, this is the question that determines whether the marketing programme is commercially sustainable.
Read more:
Want an honest ROI conversation?
Book a Discovery