There’s a particular kind of paid media report that makes everyone in the room feel good and changes absolutely nothing about the business. Beautiful dashboards! Green arrows everywhere! So why is your CAC creeping up? Most paid media reporting celebrates outputs that look productive without ever connecting them to the outcomes that move the business. Here’s how to tell which kind of agency you’re paying for.
Key Takeaways
- Outputs and outcomes are not the same.
- Outputs (CTR, CPC, CPL, conversion rate) are what campaigns generate. Outcomes (revenue, CAC, LTV, ROAS, payback) are what the business actually cares about.
- Outputs-only reporting is the vanity metric trap. Green-arrow dashboards can mask rising CAC, falling lead quality, and flat revenue.
- Outputs-only reporting leaves you flying blind. Revenue and ROAS are lagging indicators. Outputs are the leading indicators a sharp account manager uses to course-correct.
- Before a campaign goes live, your agency should align stakeholders on two questions: what are we trying to achieve, and how will we know if we did it? Read on for more.
Let’s dig into the details on why your SEM agency needs to speak to both outcomes and outputs.
“Can we just ask AI how the campaign performed?”
Sure. And the answer you’ll get back will look exactly like what a mediocre analyst would hand you: a pile of data, some arbitrary comparisons to other data it scraped together, and zero connection to whether your business actually moved. It’s data, but it’s not measurement.
Real performance measurement is humans aligning on expectations for business outcomes. That’s a sentence worth re-reading, because it’s the dividing line between an agency that’s a strategic partner and an agency that’s a glorified dashboard.
If you’re a CMO, CFO, or CEO evaluating your current agency (or shopping for a new one), here’s the question worth asking: Is your agency reporting on business outputs or business outcomes? And do they know the difference?
The difference between outputs and outcomes
It’s a simple distinction that gets muddied constantly, so let’s draw a clean line:
- Business outputs are the metrics campaigns generate as they run. CTR, CPC, CPM, impressions, quality score, conversion rate, cost per lead. These are what your paid media team monitors, diagnoses, and influences through creative, targeting, bid strategy, and landing page work.
- Business outcomes are what your business actually cares about. Revenue, pipeline, customer acquisition cost (CAC), lifetime value (LTV), return on ad spend (ROAS), payback period. These are why the campaigns exist in the first place.
Both matter. Reporting only on outputs is like a pilot reporting only on fuel burn rate without mentioning whether the plane is heading to the right city. Reporting only on outcomes leaves you with no way to diagnose what’s working and no way to course-correct mid-flight.
The vanity metric trap
Here’s the trap a lot of executives walk into without realizing it.
Your agency sends a beautifully designed monthly report. Click-through rate is up 32%. Cost per click is down 18%. Impressions are at an all-time high. There are green arrows everywhere.
And yet your CAC is creeping up. Your sales team is complaining about lead quality. Revenue isn’t moving the way the report would suggest it should.
What happened? Your agency optimized for outputs that look good in a deck without ever connecting them to outcomes that matter to the business. CTR going up doesn’t mean revenue is going up. A lower CPC doesn’t mean better customers. Cheaper leads aren’t the same as more valuable customers.
This is the vanity metric trap, and it’s especially dangerous because the metrics being reported aren’t wrong—they’re just incomplete.
- An agency that wants to look productive will always have something to celebrate.
- An agency that wants to make you successful will tell you when the outputs are improving but the outcomes aren’t, and then propose what to do about it.
If your reporting cycle never includes the words “we changed our mind about this strategy because the data told us to,” that’s a red flag.
Outputs aren’t useless. They’re leading indicators.
The other failure mode is the opposite problem. Imagine a CFO who’s tired of the green-arrow reports and tells the agency, “I only want to see revenue and ROAS. Stop sending me CTR and CPC.”
Understandable instinct. But also flawed.
Here’s why: by the time revenue and ROAS show up in the financial reporting, the spend has already happened. Outcomes are lagging indicators by definition. They tell you what occurred, not what’s about to occur. You can’t make Tuesday’s optimization decision based on a number that won’t be final until next quarter’s close.
Outputs are the leading indicators. They’re how a good account manager spots that something’s working (or breaking) two weeks before it shows up in revenue. CPC creeping up on your highest-intent keyword set tells you auction pressure is changing. A dropping conversion rate on a specific landing page tells you something’s wrong with the experience, not the spend. Quality score shifts tell you the algorithm’s perception of your relevance is changing.
Outputs are diagnostic. Outcomes are the verdict. A good agency uses outputs to predict, explain, and adjust toward outcomes, not as a substitute for them.
The two magic questions
Here’s what separates the agencies that drive outcomes from the ones that just report on outputs. Before any campaign goes live, the team should be able to answer two questions:
- What are we trying to achieve with this effort?
- How will we know if we’ve done that?
That second question is a two-parter. It requires identifying one or more direct or proxy measures (your KPIs) and a target for each one. Not “we’ll track conversion rate,” but an actual target or expectation we’re committing to.
No targets means no time machine. Without a target set in advance, you can’t go back and ask whether the campaign performed how you expected, because you never said what you expected. Every result becomes defensible after the fact, and “performance measurement” turns into “performance narration.”
If your agency hasn’t asked you for targets, you don’t have an agency that’s measuring performance. You have an agency that’s reporting activity.
How to actually set targets (without drama)
Here’s our target-setting trick that gets teams unstuck, and it works almost every time.
Get everyone with a stake in the outcome—agency lead, internal marketing owner, finance, sales, leadership—to come up with their target independently before anyone shares.
Then everyone shares.
What happens next is always the same: a thoughtful discussion and a quick alignment on what the target (or target range) should be. The reason it works is that no one feels individually exposed.
The conversation is about reconciling differences, not defending a position. You’ll often find the proposed numbers are closer than people expected, and where they’re not, the gap itself could be the most useful conversation you’ll have all quarter.
Where AI fits into all this
A reasonable question at this point: can’t AI just do this? Set the targets, watch the campaigns, and tell us what happened?
Not really, and not for the reason most people assume. The bottleneck isn’t AI’s analytical ability. The bottleneck is the human alignment work that has to happen before the data exists. AI can’t tell you what your business is trying to achieve. AI can’t reconcile your CMO’s growth target with your CFO’s payback expectations. AI can’t decide whether you’d rather have 1,000 cheap leads or 200 expensive ones from the right ideal customer profile.
AI is also worth measuring with the same two questions. If your agency is rolling out AI-powered initiatives like automated bidding, generative creative, or predictive audience modeling, the right questions are still: what business outcome is this trying to achieve, and how will we measure it?
What this looks like from the C-suite
The next time your agency walks you through a report, listen for whether they’re connecting the outputs they’re reporting on to outcomes that matter to the business. Are they telling you that CPL went down and what that meant for CAC? Are they showing you that ROAS improved and explaining which output changes drove it? Are they bringing forward recommendations that include reallocating the budget when the data calls for it, even if it means spending less in a channel they manage?
That’s a strategic partner and an agency operating like an owner instead of a vendor.
The agencies worth keeping speak both languages. They use outputs to drive the work and outcomes to drive the business, and they tell you the truth about both: the good, the bad, and the ugly.
Want to see what reporting looks like when an agency actually connects the work to the business? Contact us to talk with our Performance Marketing team about how we partner with C-suite leaders to drive outcomes that show up in revenue, not just dashboards.