Last week, Jason Beddall was the guest on the Moneyball CMO Podcast. He heads growth marketing at NContracts, and he’s one of those rare marketers who actually knows what he’s doing with data. We’ve worked together for a while. He’s a RevSure customer, and honestly, sometimes I think he understands the product better than I do.

But we didn’t talk much about ourselves on this pod. Instead, we discussed something more interesting: how AI is making his team more productive and less effective at the same time.

Here’s what he said that stuck with me:

Buyers could smell the automation resulting in collapsing response rates and stalled conversions.

And I think this is the pattern we’re all seeing but nobody wants to talk about. Everyone’s rushing to automate everything. And most of us are automating the wrong things.

Jason’s a builder by nature. He has spent years running his own content production company before moving into marketing leadership. When NContracts needed better attribution, his instinct was: we’ll build it ourselves.

They knew their requirements and had great engineers.

So they built. 18 months. Solid effort.

They got to 65%.

And that’s where they stayed.

I’ve seen this pattern dozens of times now. Here’s what most CMOs don’t realize:

The last 35% is where all the value lives.

At 65% you have dashboards. You have reports. You can show charts in meetings.

But you can’t make decisions. You can’t see what’s actually driving pipeline.

When Jason’s team finally bought RevSure and got the full picture, they found $2M in pipeline sitting in their data. Not new pipeline, but what they couldn’t see before because everything was fragmented.

It is also important to understand the human capital and the time cost of team building the tooling that can be bought.

Most marketing leaders keep building because buying feels like admitting defeat. It’s not, it’s better to spend your team’s time driving pipeline.

I felt this.

Because that’s what attribution has become. A credit assignment machine. Marketing wants credit. Sales wants credit. Everyone’s arguing about methodology.

This is backwards.

Attribution should answer:

You can’t answer those when your team spends more time defending touch models than acting on insights.

But here’s the uncomfortable truth I’m seeing: Most marketing teams don’t want attribution. They want validation.

They’ve already decided LinkedIn is strategic. They’ve already committed to the big event. Attribution is just the tool they use to justify decisions already made.

And when the data contradicts those decisions? They question the methodology.

I’ve watched CMOs get data showing their flagship campaign has negative lift. Instead of killing it, they ask: “Can we rerun the analysis?”

The data’s not wrong. The courage is missing.

Jason’s team had data everywhere - salesforce, marketing automation, ad platforms, analytics tools. All of it technically correct. None of it talking to each other.

This sounds obvious, right? But most marketing teams don’t have this.

They have:

All showing different numbers. All technically defensible, but none of them the same.

The result? Endless meetings about whose numbers are right instead of decisions about where to spend.

When Jason’s team got RevSure, they had a single source of truth, which helped them find that $2M in pipeline they couldn’t see before. But more importantly, they could finally make decisions.

Should we spend more on LinkedIn? Let’s look at the data.

Is this campaign working? Here’s what the numbers say.

Which channels should we kill? Here’s the evidence.

No more gut feeling or spreadsheet archaeology. Just: here’s what’s happening, here’s what we should do.

You probably have pipeline sitting in your data right now that you can’t see. The question is whether you’re willing to invest in seeing it.

Jason’s content team is way more productive than they were a year ago. They’ve built agents, automated research, 5-10X’d their output.

But when they tried to take that same automation approach and apply it to customer outreach, everything broke.

I think this is the pattern a lot of CMOs are missing right now.

AI is incredible for internal stuff. Content production, research, data analysis, all the things your team does that don’t face customers.

But the moment you try to use it to fake human connection, prospects smell it. And they ghost you.

Your AI-generated emails aren’t landing in spam. They’re landing in inbox and getting ignored. Because they feel fake.

Simply, because they are fake.

You can generate 1000 personalised emails in an hour, the efficiency is incredible. But efficiency doesn’t matter if nobody responds.

Jason’s approach now: Use AI to make the team faster on everything internal - content, ops, analysis, all of it.

But when it comes to reaching out to prospects? Human touch and real personalisation. Actual understanding of what they’re dealing with is most important for actual conversions.

Jason’s been in marketing for a decade. But he’s not coasting on experience.

This is the practical reality for marketers.

The CMOs who win won’t have the best AI tools. They’ll know where to use AI and where to use humans.

They’ll understand when to build, when to invest in buying the right tool and what it takes to have the judgement, taste and courage to act on data that contradicts their instincts.

When I asked Jason what channels always get funded, what never gets cut. He said, “Content. We’re big believers in knowledge-as-a-service.”

NContracts operates in financial compliance. Heavily regulated space with high trust requirements. You can’t growth hack your way into these accounts.

So they create blogs, case studies, whitepapers that actually educate buyers how to think about vendor risk and regulatory compliance.

The best marketing doesn’t pitch. It teaches.

When you educate prospects on how to think about their problem, not your product - they start trusting you. And when they trust you, they buy.

I asked Jason if he’s seeing buyer agents yet. AI bots filling forms, downloading content, researching vendors.

This is the thing that keeps me up at night.

Your MQL metrics are about to become meaningless. Your form fills might be bots. Your content downloads might be AI research agents doing comp analysis.

The first human you talk to won’t be at the top of the funnel anymore. It’ll be at the bottom, after their AI has already researched you, compared you to competitors, and built a shortlist.

All your brand work, all your content, all your thought leadership - it gets consumed by machines before humans see it.

I don’t know what the answer is yet. I know most CMOs aren’t thinking about this, but it is going to hit us sooner than we think.

This conversation with Jason reminded me why I started Moneyball CMO.

The best marketing leaders aren’t the ones with perfect execution. They’re the ones honest about what’s working and what isn’t. Willing to kill things. Willing to buy when building doesn’t make sense.

That’s what Jason’s doing. That’s what the CMOs I respect are doing.

They don’t wait for perfect data. They make decisions with the best information available, measure what happens, and adapt.

And they remember that marketing is still about humans connecting with humans.

He pulled back on AI outreach when everyone’s rushing toward automation. That took conviction.

The technology is table stakes now. Everyone has access to the same tools.

What separates the CMOs who thrive from the ones who survive is simple:

Watch the full conversation here:

Know a CMO doing something different? Reply to this email. I want to hear their story.

P.S. Jason’s team found $2M in pipeline they couldn’t see before. How much are you missing? If you want to find out, visit RevSure.