Should we hire a marketing leader, an agency, or a fractional CMO?

The scene I'm in this week

Last week I sat with the CEO of a $23M B2B cybersecurity company. It's June 2026. He slides his laptop across the table and there's a spreadsheet open. Three columns. Full-time VP of Marketing, with a salary band and a six-month ramp. An agency, with a monthly retainer and a list of deliverables. A fractional CMO, somewhere in the middle on cost. He's been comparing them for a month. He says, I just need you to tell me which one of these is the right call. We've got a marketer doing a lot of marketing and none of it is turning into deals.

That last sentence is the whole post, by the way. A marketer doing a lot of marketing and none of it turning into deals. Hold onto it.

I didn't look at his spreadsheet. I asked him one thing instead. I said, whoever you hire out of those three columns, what's the first thing they're going to do on day one? He thought about it and said, well, figure out our positioning, get the message tight, then build a plan around it. So I asked the follow-up. The message they're going to get tight, does it exist anywhere right now? Is it written down? Could you hand it to me on a single page? He went quiet. Same quiet I get almost every time.

Here's what he was actually doing with that spreadsheet. He was shopping for someone to make a decision he hadn't made yet, and pricing the three different ways to outsource it. That's not a hiring problem. That's a founder trying to buy his way out of the one piece of work nobody else can do for him. Let me show you why all three columns fail the exact same way.

What's actually broken when you reach for the org chart?

When marketing isn't producing, the founder's reflex is to change who's running it. New leader. New agency. New fractional. The instinct feels like progress because hiring feels like building. But you're changing the driver and never checking whether there's a road.

Here's the thing all three options have in common. They're delivery vehicles. A full-time marketing leader, an agency, and a fractional CMO are three different ways to run a message. None of them is a way to create one. They all assume the message already exists, that someone has already decided who your buyer really is, what villain you're ending, what one outcome a customer would be furious to lose. When that work is done, any of the three can carry it. When it isn't, all three default to the same thing: they manage channels, they ship content, they report on metrics, and they produce a lot of motion that never becomes pipeline.

That motion has a name. It's Solution-Focused Marketing, copy and campaigns that describe what your product does instead of enrolling a buyer in a problem worth solving now. It's the default setting of every vehicle you can hire, because describing features is safe and survives an internal review, while naming a villain and staking a claim feels risky. So you get a tidy homepage, a content calendar, a nurture sequence, all of it explaining, none of it enrolling. The deals don't move and you blame the hire.

You've maybe met the archetype already. Call her Sarah. She's the in-house marketing leader who's very good at the playbook she was trained on, the MQLs, the webinars, the attribution dashboard. She runs it beautifully. And in about three months the founder starts to feel that none of it is enrolling anyone, because Sarah was handed a vehicle and no message, and the playbook she knows can't manufacture a point of view the founder never gave her. Sarah isn't the problem. Sarah is what happens when you hire a driver and forget you never built the road.

Why is this a worse mistake in 2026 than it used to be?

Hiring the wrong vehicle used to cost you time and some money. You'd burn six months, part ways, try the next column, and eventually stumble into someone senior enough to half-reconstruct your message on their own. The market gave you slack. AI just took the slack away.

AI collapsed the cost of producing marketing deliverables to near zero. Any marketer, any agency, any fractional you hire can now generate a quarter's worth of content in an afternoon. Sounds like a win. It's a trap. Untrained AI produces trendslop, generic averaged-out copy that sounds confident and says nothing, because the model has nothing specific about you to work from. So the vehicle you hired doesn't just fail to create your message. It now mass-produces the generic version of it at a speed no human could match. The cost of being undifferentiated didn't go up. It went to zero, and the volume went to infinity. Garbage in, garbage out, at scale.

Which is why the variable that matters flipped. When deliverables were expensive, hiring capacity to produce them made sense. Now production is free and the scarce thing is a point of view worth producing. Gartner found that 74% of B2B buyers say unclear messaging is their top friction point in a purchase. Read that next to your spreadsheet. The thing costing you deals isn't a shortage of marketing output. It's that the output has nothing clear to say. No new vehicle fixes that. It just produces more of the unclear thing.

There's a sharper way to ask the question. Are you hiring someone to run a rebellion, or to manage your options? You close deals yourself, when you close them, because you sell a rebellion, the named enemy, the stake in the ground, the world that should exist. The vehicle you're about to hire, in every column, is going to sell options, the feature set, the comparison, the demo, because that's all the brief you'd give them contains. You can't hand someone a rebellion you've never written down. You can only hand them your options, and options lose.

The diagnostic ... run this before you pick a column

You don't need to interview a single candidate to find out whether you're about to waste this hire. Three tests. Run them this week, and the answer is usually obvious by the second one.

1. 1The Job-Description Test. Open the req or the agency brief you're about to send. Read it honestly. Does it ask this person to FIND your message, the buyer, the villain, the point of view, or does it ask them to RUN a message and hit channel numbers? Almost every founder writes the second kind of brief while believing they need the first kind of person. If your brief is a list of channels and KPIs, you're hiring a driver and assuming the road exists. Whatever the brief doesn't ask them to create, you'd better already have.
2. 2The Strategy-or-Execution Test. For each column, ask what you're actually buying. A full-time leader and a fractional CMO are sold as strategy. An agency is sold as execution. Now ask the harder question: which one are you actually missing? If you have no documented message, you have a strategy gap, and an agency will execute your way deeper into the wrong message faster. If you have a clear message and no hands, you have an execution gap, and a fractional setting more strategy is redundant. Most founders buy the opposite of what they're short on.
3. 3The Single-Page Test. Try to write your message on one page right now. Who is this for, specifically. The old way they're stuck in. The villain you're ending. The one outcome they'd be furious to lose. Give yourself fifteen minutes. If you can't fill the page, or what comes out is a feature list, you've just learned what every vehicle you hire will inherit. You can't delegate the page. You can only delegate what's on it once it exists.

What I see across 300+ founder engagements

I've run the Magnetic Messaging Framework work across more than 300 founder-led B2B companies in the $5M-$75M range. The hire-a-marketer question walks through the door constantly, and nine times out of ten the founder is debating the wrong variable. They're agonizing over which vehicle when the thing missing isn't a vehicle at all. It's the page. The message has never left the founder's head, so every column on the spreadsheet is just a different way to hand nothing to someone and pay them to make it look like something.

The fractional CMO column is where this gets most expensive, because it looks the most like a real fix. Fractional Freedom Friday put a number on it.

“7 in 10 fractional CMO engagements in 2026 end in disappointment within six months. The founder thought they were hiring a strategist who would extract the company's narrative and name the villain. What they got was a senior marketer who audited the funnel and reported on click-through rates. Those are not the same job.”

... Fractional Freedom Friday, May 2026

Read that closely, because it's the whole pattern in three sentences. The founder wanted extraction. They bought execution. The fractional, who is usually genuinely good, did the job the engagement was actually scoped for, ran the funnel, reported the metrics, and the founder ended up disappointed because the one thing they needed, the narrative pulled out of their own head and written down, was never in the contract. It's not a talent problem on either side. It's that nobody named what the job actually was before they filled it. The same disappointment shows up with a new VP of Marketing and with an agency. The fractional just fails faster and costs enough to sting.

The deeper tell is that these founders have usually changed everything except the page. New marketer, new agency, new fractional, new tool, new dashboard. The one move nobody made is sitting the founder down and extracting the message onto paper, because the founder is busy running the company, which keeps the message in his head, which guarantees the next hire inherits nothing, which produces the next disappointment. It's a loop, and swapping vehicles just makes the loop more expensive. I wrote about the in-house version of this, the structural reasons your team is set up to miss, in "67% of Sales Reps Are Missing Quota. Your 2022 Marketing Structure Is Why."

A real example

One of those 300 was that cybersecurity company, roughly $23M, PE-backed. Before they found me, they'd already run the play in order. They hired an in-house marketing manager first. Lots of activity, flat pipeline. They decided she was too junior and brought in a fractional CMO, six months, a sharp operator. He cleaned up the funnel, fixed the attribution, professionalized the reporting. The dashboards got beautiful. Pipeline stayed flat. The board read it as another miss and the founder, when I met him, was a week away from signing an agency to, quote, finally get the content engine going. Third vehicle, same empty road.

We didn't hire a fourth anything. We ran the Magnetic Messaging Framework discovery and pulled the message out of the founder's head in an afternoon. His real story was sharp and specific: the whole category sold dashboards and alerts while the buyer's actual nightmare was being the person who has to explain a breach to the board. Their old vendors sold visibility. He sold the end of that specific 2am phone call. None of that was on the homepage. None of it was in the fractional's brief. The deck was a feature tour. We documented the buyer, the villain, the old way they were ending, and the one outcome that buyer would be furious to lose, put it on the page, then built an AI Brand Twin on top of the framework so the team could draft in his point of view instead of the internet's average.

Here's the part that matters for your spreadsheet. We kept the fractional CMO. Same person the board had written off. Once he had the documented message to run instead of a funnel to audit, his work started landing, because the strategy gap was filled and his execution finally had something true to carry. Over the next two quarters pipeline moved for the first time in a year. The vehicle was never the problem. The founder had been buying drivers for a road that didn't exist. We built the road. The driver he already had turned out to be fine. If you want the AI half of how that story scaled, I unpack it in "You Don't Need a Marketing Team. You Need a Story. AI Should Be Doing the Rest."

What this means for you

Close the spreadsheet for a minute. The three columns aren't really different choices. They're three prices for the same assumption, that a message exists for someone to run. Until that assumption is true, picking a column just decides how you'll spend the next six months being disappointed. Here's the comparison that actually matters, the one nobody puts in the spreadsheet:

Notice the bottom row isn't a fourth vendor. It's the work you do before you choose, so the vehicle you pick has something real to carry. Three things you can do this week, none of which require committing to a column yet:

1. 1Run the Single-Page Test on yourself before you post any req. Fifteen minutes, one page: the buyer, the old way, the villain, the one outcome they can't lose. Whatever you can't write is what every hire will inherit empty. Most founders are shaken by how hard the page is, and that's the most useful thing they learn all quarter.
2. 2Rewrite the brief before you rewrite the budget. Take the req or agency SOW and read it against the Job-Description Test. If it's all channels and KPIs, you're about to buy execution for a strategy gap. Either rewrite the role to be about finding and documenting the message, or do that work first and then hire someone to run it. Don't hand a vehicle an empty road.
3. 3Pause the hire for thirty days. Not forever. Just long enough to break the reflex that the next vehicle is the fix. Use the thirty days to get the message onto paper, by yourself or with help. If at the end of it you've got a documented message and you still want a fractional or an agency to run it, hire with confidence. Now you're handing them a road, not asking them to find one.