ROI. Why marketing defies simple measurement

Marketing teams want proof. Finance wants certainty. Leaders want numbers they can defend. ROI sits in the middle and causes tension because marketing does not operate in straight lines.

Not all value shows up in a report.

Some activity links cleanly to revenue. Much of it does not. Treating all marketing as if it should convert directly leads to poor decisions.

Direct response makes this tempting. Paid search. Retargeting. Shopping ads. These channels sit close to the transaction. Spend goes in. Sales come out. ROAS works here because the distance between action and outcome is short. Attribution has value.

The problem starts when this logic gets applied to everything else.

Brand awareness, reputation, and retention work differently. They shape behaviour before intent appears. They influence choice, trust and willingness to pay. These effects show up later and across multiple touchpoints. They resist clean attribution.

You can see this clearly in UK brands.

John Lewis invests heavily in long term brand campaigns. Many of these do not show strong short term ROAS. What they do build is trust, familiarity, and emotional connection. That trust shows up later in higher basket sizes, repeat visits, and resilience during price competition. The ROI exists, but not at campaign level.

Marks & Spencer provides another example. Its brand recovery over recent years came from consistency in messaging, quality signals, and experience. No single campaign drove the turnaround. The lift came from cumulative brand work that improved perception, footfall, and loyalty. Attribution models struggle here, but the commercial impact is clear.

ROAS is effective when the channel is designed to drive immediate action. Search, retargeting, and offer led media need tight feedback loops. These should be measured hard and managed closely.

ROAS becomes misleading when it is treated as the only signal of value. Teams cut brand spend because it looks inefficient. Performance holds briefly. Then efficiency drops. Costs rise. Conversion softens. The connection feels unclear, but the cause is structural.

Strong teams separate measurement by intent.

They use ROAS and attribution where conversion is the goal.

They use trend signals where influence is the goal. Branded search growth. Direct traffic. Conversion rate over time. Retention curves. Price sensitivity.

They judge success based on what the activity is meant to change.

They also accept uncertainty. Some of the most valuable work improves the system rather than a single metric. Forcing precision where it does not belong creates false confidence.

ROI still matters. Discipline still matters. But false precision does more damage than informed judgment.

Marketing works when teams understand the system, not just the spreadsheet.