Measurement
How to measure your marketing agency's performance.
Most agency reports are full of numbers and short on answers. Here's what the research says about why measuring marketing is hard, and a practical scorecard, review rhythm and set of red flags to judge whether your agency is moving the business.
"Is our agency doing a good job?" is one of the most common questions we hear from founders and marketing leads. It sounds simple. In practice, most companies can't answer it with confidence, and it's usually not because the agency is hiding anything. It's because nobody agreed what good looks like.
This guide covers how to fix that.
Why is agency performance so hard to judge?
Because marketing itself is hard to measure, and agencies sit in the middle of that problem. A 2024 Gartner survey, reported by Destination CRM, found only 52% of CMOs and senior marketing leaders said they succeeded in proving marketing's value and receiving credit for it. CMOs ranked CFOs (40%) and CEOs (39%) as the executives most skeptical of marketing's value.
Confidence doesn't match practice either. Nielsen's Marketing ROI Blueprint, released in October 2025, reported that 85% of marketers say they are confident in their ability to measure ROI, yet only 32% measure ROI across both traditional and digital channels. Nielsen's 2025 Annual Marketing Report surveyed 1,400 marketers with budgets of at least US$1 million.
And doubt has a price. In TransUnion's October 2025 study, fielded by EMARKETER with 196 US marketers, 62% had some confidence in their performance metrics, but 60% said internal stakeholders question those metrics at least sometimes. Nearly 29% said up to 20% of their marketing budget had been reallocated or put at risk because of doubts about measurement.
The agency relationship adds its own gap. The 4As and ANA's 2024 Cost of the Pitch II study found 90% of clients said value and long-term ROI outweighed cost, but only 5% of clients and 10% of agencies had a formal, corporate-backed definition of value. If value isn't defined, it can't be measured.
Why do platform metrics mislead?
Most agency reports lean on numbers from ad platforms: return on ad spend, cost per lead, conversions. These are useful for managing campaigns. They are a poor way to judge an agency on their own, for a few reasons:
- Every platform grades its own homework. Google, Meta and others each count the conversions they touched, so the same sale can be claimed several times.
- Data is fragmented. In the TransUnion study, 49% of marketers cited siloed or incomplete data, 48% cited cross-channel deduplication and 41% cited walled-garden reporting limits as barriers to accurate measurement.
- Correlation looks like causation. Retargeting and branded search often show strong returns because they reach people who were already going to buy.
- Signals are thinner. Privacy changes mean more conversions are modelled rather than observed. We cover this in why your marketing numbers don't add up.
The direction of travel is clear. In the same study, 67% of marketers said they now prioritize incremental ROI and 66% prioritize aligning marketing metrics to business outcomes. Your agency scorecard should do the same.
What should a marketing agency scorecard include?
A good scorecard is short, balanced and agreed in advance. We suggest four areas. Weight them to fit your goals; a performance-led brand might put more weight on outcomes and efficiency, while a brand rebuilding its positioning might weight quality of thinking higher.
| Area | Example KPIs | Source of truth | Review cadence |
|---|---|---|---|
| Business outcomes | Revenue or qualified pipeline from marketing; new customers; progress toward targets | Your CRM, e-commerce or finance data | Monthly trend, quarterly judgement |
| Efficiency | Blended customer acquisition cost; marketing efficiency ratio (revenue divided by total marketing spend); payback period | Finance data plus total spend, including fees | Monthly |
| Incrementality | Lift from holdout or geographic tests on major channels | Test results agreed in advance | Quarterly or per test |
| Quality of thinking | Meaningful tests run; learnings documented and applied; quality of recommendations | Test log and QBR materials | Quarterly |
| Collaboration | Responsiveness; reporting accuracy and timeliness; proactive ideas; team stability | Short internal survey of your team | Quarterly, with a fuller annual review |
Keep platform metrics, but treat them as diagnostics underneath the scorecard, not the scorecard itself. If you want to check the paid media foundations first, our performance marketing audit checklist covers tracking, structure and spend.
How do you run a quarterly business review?
A quarterly business review (QBR) is where the scorecard turns into decisions. Keep it to about 90 minutes, with materials sent two days ahead. A simple agenda:
- Scorecard review. Results against each area, using your source of truth, not screenshots from ad platforms.
- What we tested and learned. The two or three experiments that mattered, including the ones that failed.
- What changed in the market. Competitors, costs, platform changes and anything affecting your customers.
- Next quarter's plan. Priorities, budget shifts and the tests that will run.
- Relationship check. What each side needs from the other to do better work.
Include whoever owns your analytics. Gartner's 2024 survey found that 62% of marketing leaders who meet regularly with their top analytics leaders and teams said they could prove value, compared with 30% of those who meet infrequently.
Use QBRs to improve the relationship, not to put it on trial every quarter. The ANA and 4As 2025 tenure report found that clients without mandatory agency reviews averaged relationships of 8.1 years, versus as little as 3.8 years where reviews were frequent. Constant re-pitching isn't measurement. Agreed metrics, reviewed honestly, are. The same 2024 Cost of the Pitch II study found 56% of respondents cited trust as the most important factor in a successful long-term relationship.
What are the red flags in agency reporting?
- Reports are mostly impressions, clicks and platform ROAS, with no link to revenue or pipeline.
- Platform-reported conversions add up to more than your actual sales.
- Total cost excludes agency fees, tools or production, which flatters efficiency.
- Every month's story is positive, and nothing is ever described as a failed test.
- Metrics or definitions change from report to report without explanation.
- The agency can't give you raw data, or the dashboards live only in their tools.
- Recommendations always involve more spend in the channels the agency manages.
One or two of these can be fixed with a conversation. Several together usually point to a deeper problem. If that's where you are, our guide to switching marketing agencies without losing momentum covers how to move safely, and why brands leave their marketing agency can help you diagnose the cause first.
How Three Dots Brands approaches this
We help brands in Toronto and across Canada set up measurement leadership can trust and hold every partner, including us, to the same standard. Using our Think, Plan, Execute approach, we define what value means for your business, build the scorecard and review rhythm, and help you read the results honestly. If you're not sure whether your marketing is working, start with our marketing diagnostic.
Frequently asked questions
What KPIs should I use to measure my marketing agency?
Use a small set across four areas. Business outcomes, such as revenue or qualified pipeline from marketing. Efficiency, such as blended customer acquisition cost or marketing efficiency ratio. Quality of thinking, such as the number of meaningful tests run and learnings applied. Collaboration, such as responsiveness and accuracy of reporting. Agree the definitions and the source of truth before the quarter starts.
How do I know if my marketing agency is doing a good job?
Look for three things: business results that show up in your own systems, not just ad platforms; a clear explanation of what they tested, what they learned and what they will change; and honest reporting that includes what didn't work. If you can't tell whether results are improving, the measurement setup needs fixing before you judge the agency.
How often should I review my agency's performance?
Review operating metrics monthly, hold a quarterly business review on outcomes and strategy, and do a fuller relationship review once a year. Weekly check-ins are useful for active campaigns, but judging an agency week to week encourages short-term decisions.
Why don't my agency's reported results match my revenue?
Each ad platform tends to count conversions it touched, so several platforms can claim the same sale. Tracking gaps from privacy changes, different attribution windows and modelled conversions add to the gap. Reconcile platform reports against your CRM or sales data, and use a blended metric and incrementality tests to judge what marketing actually caused.
Sources
- Gartner, "Gartner survey finds only 52% of senior marketing leaders can prove marketing's value and receive credit for its contribution to business outcomes", 2024
- Destination CRM, "Marketing leaders can't prove their value", 2024
- Nielsen, "Nielsen unveils blueprint to achieve confident ROI", 2025
- Nielsen, "Nielsen releases its 2025 Annual Marketing Report", 2025
- TransUnion (via GlobeNewswire), "New TransUnion research reveals marketers' confidence in measurement has stalled", 2025
- 4As, "4As and ANA study finds value outweighs cost in the client-agency review process", 2024
- ANA, "New ANA and 4As report reveals client-agency relationship tenure has doubled since 2016", 2025
Data limitations: most studies cited are US or global and skew toward larger advertisers. The TransUnion study is a small survey of 196 US marketers published by a measurement vendor. The Gartner figures are taken from published coverage of the survey. Survey results are self-reported and use different definitions of confidence and value. Figures are the most recent we could verify as of October 2026.