Ask three people in the same company how marketing is performing and you will often get three different answers. The paid search specialist points to a healthy cost per conversion. The social media manager shares a strong return on ad spend. The finance team notes that revenue is flat while marketing costs have grown. Each person is looking at real numbers — just not the same ones.
Good measurement is not about collecting more data. It is about agreeing which numbers reflect real business outcomes, how they relate to each other, and which ones should drive decisions. This article sets out a practical framework for doing that.
Why platform metrics are not enough
Every advertising platform reports its own results using its own rules. Each one decides which interactions count, how long after an ad view or click a conversion can be credited, and how to estimate conversions it cannot observe directly. Those rules are reasonable from each platform's point of view, but they are not designed to agree with one another.
The practical consequence is that if you add up conversions reported by every platform, the total frequently exceeds the number of customers you actually acquired. The same purchase may be claimed by a search ad, a social ad the customer saw a week earlier, and a display retargeting impression. None of the platforms is necessarily wrong by its own definition, but none of them is answering the question your business cares about: what did our marketing investment actually produce?
Platform metrics remain essential for day-to-day optimization. They are simply the wrong place to start when judging overall performance.
Start with the business outcome
Begin with the outcome that matters to the business, expressed in the terms your leadership and finance team already use. Depending on your model, that might be:
- Revenue or gross profit for e-commerce and transactional businesses.
- Qualified pipeline and closed revenue for B2B and high-consideration services.
- Booked appointments or jobs for local service businesses.
- Enrolments, subscriptions, or activated users for education and software.
Where possible, move beyond revenue toward profit. A campaign that drives sales of low-margin products, or attracts customers with high return rates, can look excellent on revenue and still lose money. Contribution margin — revenue minus the variable costs of delivering it — is often the most honest measure of whether marketing is paying for itself.
Build a hierarchy of metrics
Once the outcome is clear, organize metrics into layers so that everyone knows which numbers inform which decisions:
- Business outcome metrics. One or two measures, such as marketing-driven revenue, profit, or pipeline, reviewed by leadership.
- Efficiency metrics. Measures that relate outcomes to spend, such as blended customer acquisition cost or marketing efficiency ratio (total revenue divided by total marketing spend).
- Channel performance metrics. Cost per qualified lead, return on ad spend, or cost per acquisition for each channel, interpreted in light of that channel's role.
- Diagnostic metrics. Click-through rate, conversion rate, impression share, frequency, and similar measures that explain why channel performance is changing.
The hierarchy prevents a common failure: optimizing a diagnostic metric — click-through rate, for example — in a way that harms the outcome above it.
Define conversions consistently
Measurement breaks down quickly when "a conversion" means different things in different places. Agree on a small set of primary conversions that represent genuine business value, and treat everything else as secondary.
A primary conversion for a consulting firm might be a qualified enquiry submitted through the contact form. Secondary signals might include brochure downloads, newsletter sign-ups, or visits to the pricing page. Secondary signals are useful for understanding behaviour, but if they are set as optimization targets alongside real leads, automated bidding will happily pursue the cheaper, easier actions.
Where different conversions have different value, assign values that reflect reality. If one service line is worth several times more than another, telling the platforms so allows their bidding to prioritize accordingly.
Close the loop with downstream data
For businesses where the sale happens offline or later — after a phone call, a proposal, or a sales meeting — the most valuable data lives in your CRM, not in your analytics tool. Feeding qualified-lead and closed-sale information back to the advertising platforms, through their offline conversion import features, lets optimization focus on leads that become customers rather than on form fills alone.
Even without automated imports, a monthly reconciliation of leads by source against sales outcomes will reveal which channels produce enquiries that actually close.
Use blended and channel views together
Blended metrics, which compare total marketing spend with total business outcomes, are immune to attribution disagreements because they do not attempt to split credit. Their weakness is that they cannot tell you which channel to change. Channel-level metrics have the opposite profile.
Used together, they keep each other honest. If channel reports show improving efficiency while the blended figure deteriorates, it is a strong signal that platforms are claiming credit for conversions that would have happened anyway.
Set baselines and a review rhythm
A number only means something relative to a baseline. Before making significant changes, record current performance over a stable period, note seasonality and known external factors, and agree how you will judge success. Then review on a rhythm that suits each layer of the hierarchy:
- Diagnostic metrics: frequently, by the people managing campaigns.
- Channel performance: weekly or fortnightly.
- Business outcomes and blended efficiency: monthly and quarterly, with leadership.
A practical checklist
- Is there a single, agreed definition of the business outcome marketing is responsible for?
- Are primary and secondary conversions clearly separated in every platform?
- Do conversion values reflect genuine differences in business value?
- Is downstream data from your CRM or sales process connected back to marketing?
- Do you review a blended efficiency metric alongside channel reports?
- Is every significant change measured against a documented baseline?
None of this requires expensive software. It requires clear definitions, consistent tracking, and the discipline to judge marketing by the outcomes the business actually needs.
A Dune marketing audit applies these principles to your accounts, tracking, and funnel. See how the audit works or get in touch.
This article is general information, not advice for your specific situation. Platform features and policies change; confirm current details in each platform's official documentation.
