Why your best-performing campaigns are probably your worst

I've spent much of this year looking at marketing dashboards that don't align with the businesses behind them.

The cost per lead is down. Form fills are up. Return on ad spend (ROAS) reads green across Google and Meta. The CMO walks into the quarterly review with a deck full of arrows pointing in the right direction. Then we look at closed-won, and the picture inverts. 

The campaigns driving the most leads are driving the worst pipeline. The Sales team is neglecting the audiences with the lowest cost per lead. The keywords with the highest measured ROAS are converting deals that never close.

Here's the take I'll defend: a healthy-looking paid program is no longer evidence that marketing is working. Instead, it's evidence that you're optimizing your way into a worse business. The ad platforms are doing exactly what you've told them to do- and you haven’t told them enough. This fix isn’t a new platform or a new agency. It’s completing the loop and telling the algorithm what a closed deal actually looks like.

The signal you send is the signal you get back

Google's algorithm and Meta's algorithm aren't trying to grow your revenue. They're trying to produce more of whatever has been defined as a conversion. If your conversion event is "form fill," the algorithm optimizes to achieve more form fills. If it's a button click or a PDF download, you'll get more button clicks and PDF downloads. 

This is where the loop breaks - unless you tell the platform how much profit these campaign conversions generate and how many generated leads becames customers, the algorithm will continue to optimize for the wrong signal. 

Missing this step has downstream consequences. According to ROI Revolution's research on offline conversion tracking, businesses that connect real sales outcomes back to their ad platforms often discover their actual return on ad spend is up to 300 percent higher than the online metrics suggested. This delta is what the bidding algorithm never sees — so it treats your best-performing campaigns as mediocre, starves them of budget, and doubles down on the ones generating cheap leads that never close. Most teams are bidding cautiously on their best campaigns and aggressively on their worst, in exactly the wrong direction.

Three places where the money is leaking

There’s another majormiss when it comes to modernizing paid campaigns - auditing results against a customer relationship management (CRM) system. For too long, CRMs have been treated like a Sales or Customer service-only tool. That mentality must change. The first-party data, customer lifecycle progression, and deal data are important data sets that provide marketers with important feedback to their campaign success (or failure). CRM can also show areas where your campaigns are leaking, such as:

Bidding signal

Treating every form fill as equivalent is the wrong way to weigh the value of your conversion. For example, a procurement officer associated with a $50,000 deal signing up for a product webinar doesn’t carry the same value as a student requesting a free-tier signup. Bucketing these into the same category sets the stage for a misleading interpretation of campaign success.

The solution is to swtich from conversion optimized to value-based bidding. Google's research with Making Science found a 14 percent median lift in conversion value when advertisers move to value-based bidding, where deal size flows back into the platform and the algorithm starts hunting for revenue, not events. Meta's data from the same partnership shows up to 35 percent improvement in return on investment when the shift was made from conversion-focused to value-based.

Audience Signal

The second major efficiency indicator is the audience signal. Lookalike audiences built on form-fill data are lookalikes of your tire-kickers - useless. Lookalikes built on closed-won customers, with deal value attached, are lookalikes of your actual buyers. These are not the same audience. One of them costs more upfront and pays back. The other looks cheaper and quietly burns budget for a year.

Suppression

The third is suppression. We see this constantly with B2C service businesses: a paid program retargeting people who already signed up — over the phone, in a branch, through a sales rep — because the digital channel never knew the conversion happened. That spend is pure waste. CRM-synced exclusion lists stop it the day they're turned on.

Why the gap persists

If the math is this clear, why hasn't every B2B marketing team done this already? It comes down to three roadblocks: 

1. Platform silos: Marketing owns the ad platforms. Sales owns the CRM. Nobody owns the data flow between them, meaning nobody is accountable for ensuring the connection between the platforms is set up correctly.
    
2. Knowledge: The teams running paid media often weren't hired to architect data flows between HubSpot and Google. The teams running the CRM weren't hired to think about bidding strategies. The integration sits in the gap between two job descriptions.
    
3. Data hygiene: Even when teams strive to keep their data clean, they discover the deal stages are inconsistent, the close dates are unreliable, and the deal amounts aren't trustworthy. You can't optimize an algorithm against bad data.

This is why we keep saying the answer is RevOps, not marketing ops. [Include a quick definition of  RevOps] Connecting sales data to ad platforms isn't a marketing project. It's a revenue operations project that happens to have marketing benefits. The deeper version of that conversation — how to actually structure the work — is a piece for another day.

What to do Monday

The full RevOps build is months of work. The first move is hours.

Start by picking your top three campaigns by campaign spend over the last quarter.

In your CRM, confirm you can see which deals originated from those campaigns. If you’re unable to track attribution back to the ads, start the process of getting your reporting squared away. 

Once you can associates deals with campaigns, begin digging into the campaigns with closed-won deal status and deal amount - this is the the data that needs to be funneled back into Google and Meta as offline conversions. After you’ve identified how deal success maps back to campaigns, select and switch one campaign to a value-based bidding strategy. Watch what happens to spend allocation over the next 30 days.

Our hypothesis: You'll almost certainly find that one of those three campaigns is propping up the dashboard and quietly losing you money, and another one — the one with the higher cost per lead you've been considering pausing — is actually your best.

The big picture

The companies closing this gap first aren't just spending more efficiently. They're competing on a different axis. They're letting the algorithms work for revenue while their competitors are still letting them work for clicks. McKinsey puts the gap between leaders and laggards at 85 percent in sales growth. Dataforest has 65 percent of businesses hitting higher quotas just by improving the flow of CRM data into their marketing stack.

That gap isn't going to narrow on its own. The teams optimizing toward closed revenue today are training their algorithms on a year of better data than the teams who start in 2027.

The dashboard isn't the score; the deal is the score. Every dollar you spend telling the platform otherwise is a dollar you're handing to a competitor who knows the difference.