Impressions Are Not Evidence: The Case for Reporting on Perception, Not Reach

Somewhere in Canada this week, a chief marketing officer will present a quarterly report built entirely on numbers that prove nothing. Impressions are up forty percent. Engagement rate beat benchmark. Share of voice climbed two points. The room will nod, the report will be filed, and not one person in it will be able to say whether a single customer now believes anything different about the brand than they did in January.

This is metric myopia: the industry-wide habit of reporting on the delivery of messages while remaining silent on their effect. And it persists because the delivery numbers are easy, abundant, and flattering. The effect numbers are harder, scarcer, and occasionally damning.

But flattering is not the same as evidence. A brand can accumulate enormous reach while its market position quietly erodes — impressions flowing in one end, perception unchanged or worsening at the other. The dial, as it were, never moves.

What Vanity Metrics Actually Measure

To be clear about the indictment: impressions measure how many times a message was placed in front of someone. Engagement measures whether that placement prompted a reflex. Share of voice measures how loudly a brand spoke relative to competitors. None of these measure whether the message worked — whether it changed what the audience believes, feels, or does.

The confusion is understandable, because delivery metrics behave like performance metrics. They trend. They benchmark. They produce charts. But a chart is not a finding. The entire edifice rests on an unstated assumption: that exposure converts to persuasion at some meaningful rate. In a media environment of infinite supply and collapsing attention, that assumption has never been weaker.

Canadian brands, competing against better-resourced American entrants for the same audiences, can least afford this confusion. We do not out-spend anyone. The only ground we can win is the ground between the ears — and we are not measuring it.

What Perception Reporting Would Look Like

The alternative is not exotic. It requires asking, on a regular cadence, a small set of questions the industry already knows how to ask and mostly chooses not to:

Belief tracking. A short, repeated survey — three to five statements about the brand and category, rated by agreement — fielded consistently among the target audience. The metric is not the score. The metric is the delta: did agreement with the intended belief move, quarter over quarter, among the people the campaign actually targeted?

Association shift. What words come to mind when the audience thinks of the brand? Unprompted. If the words are the same this year as last year, the brand has spent its budget maintaining the status quo — which may be a strategy, but should be a chosen one.

Consideration among the unconvinced. Not overall consideration, which flatters incumbents, but consideration among the segment that currently prefers a competitor. Movement there is the truest signal that messaging is doing work.

The objection ledger. What reasons do prospective customers give for not choosing the brand? If the same objections recur across four quarters of active campaigning, the messaging has failed at its primary task, however impressive the reach numbers look.

A Scorecard for the Next Agency Review

Executives heading into their next agency or internal marketing review can bring a one-page scorecard with four questions:

1. Which specific belief were we trying to change, in which audience, this quarter?
2. What is the measured movement in that belief, against baseline?
3. What did we learn about the objections we failed to move?
4. What will we attempt to change next quarter, and how will we know?

If an agency or team cannot answer the first question, the strategy is undefined. If they cannot answer the second, the reporting is incomplete. If they cannot answer the third, the learning loop is broken. And if they cannot answer the fourth, the relationship is producing activity rather than progress.

None of this abolishes reach metrics. Delivery data remains useful for diagnosing why a perception didn't move. But delivery is the means, and it has been masquerading as the end for a decade.

The Courage to Be Measured

There is a reason the industry resists this shift, and it is not methodological. Perception metrics are honest, and honesty is uncomfortable. A brand that reports on belief change can fail visibly. A brand that reports on impressions can fail invisibly, forever, at scale.

Canadian brands deserve better than comfortable reporting. The market is too competitive, the budgets too modest, and the stakes — for brands trying to establish authority against larger foreign players — too high to spend another quarter charting deliveries while the dial sits still.

Measure what matters. Report on minds, not media. That is the whole argument, and it fits on a single slide — which is more than can be said for most quarterly reports.