Marketing is messy. Plan accordingly.

Hey there.

On a recent work trip, I stumbled onto a podcast called 2Bobs. It's David C. Baker and Blair Enns talking through the business of doing creative work. I don't get through many podcasts anymore now that I work from home (perhaps the only thing I miss about my old Los Angeles hour-plus commute), but this one immediately spoke to me. Over the past few months, I've listened through the entire back catalog and I'm working my way through all of Blair's books. If you make your living in the agency or creative world, I would highly, highly recommend both.

The episode that pulled this newsletter out of me is built on a post Blair wrote called "Embrace the Marketing Mess." The premise is simple: marketing is messy, not neat. Startup marketing most of all. We plan it like it's orderly, we build the calendar and the budget as if the inputs map cleanly to the outputs, and it almost never works that way.

Marketing success is fat-tailed (to use Blair and David's words). If your stats knowledge is as foggy as mine, here's the short version: picture the classic bell curve. Most outcomes cluster near the average, and the further you get from the middle, the rarer things become, until the extremes basically fall off the edge of the page. That's a thin-tailed world, and it's how most of us are trained to think.

Fat tails break that assumption. The extreme outcomes don't vanish. They show up often enough, and land hard enough, that a small handful of them define the entire average. The rare monster winner isn't a fluke you round off… it's the whole story. Nine swings do nothing. The tenth clears the fence and pays for the other nine. Build your plan around the sensible nine and you'll optimize your way right past the only one that mattered. This isn't all that different than the power laws that dictate venture capital success.

This concept is very relevant for startups trying to figure out marketing.

Let's get to it.

‍

First, stop fine-tuning your budget to the decimal.

Watch what happens in most planning meetings. The instinct is always to optimize, or to control what you feel like you can control. Move some budget from this channel to that one, build the attribution model that will finally explain everything. It feels like progress because it feels like control. It's mostly theater.

Kyle Coleman, the CMO at Copy.ai, put it well on Exit Five, borrowing from Christopher Lochhead's Play Bigger: stop "peanut butter spreading your marketing budget" across everything and expecting a tidy 15 percent more leads from October to November. That's not how any of this works. Fat tails don't show up in a spreadsheet, at least not right away.

(This can also be applied to the five-year financial projections in your Seed Round investment deck.)

Second, try a lot of things, but not all at once.

This is the part people botch when they hear "experiment more." They take it as a license to run ten things in parallel, which in practice means ten things done kind of well, none of them with enough behind it to read the result. That isn't a portfolio of experiments, it's just being busy for the sake of it.

A real "test" needs enough budget, focus, and runway to leave a mark you can measure. So sequence them, and give each one a legitimate swing (or shot now that we are all in World Cup mode). Doing five channels at 40 percent teaches you nothing. Doing one or two at full conviction teaches you what to do next.

Third, do more of what works and less of what doesn't.

When something hits, you won't need a dashboard to tell you. That's the moment discipline matters most. Blair's phrase from the article referenced above was "pour gasoline on it." Dave Gerhardt calls it replaying the hits, running the winner back until it stops working, because eventually it will. Every campaign has a shelf life. The harder half is… the hard part.

The problem isn't starting these experiments, it's stopping them in time. Stopping means admitting the bet you championed isn't paying off, so instead it limps along on life support, quietly eating budget because someone's ego is attached to it, or some vanity metrics are going OK. Decide what "not working" looks like before you launch. When you hit that line, move on without overthinking it.

Fourth, and most important: don't run somebody else's playbook.

This is the hard part, the actual strategy work, and it's the part most people rush past because it's slow and there's nothing to launch at the end of it. Before you borrow a single tactic, you need your own foundation. Your story, your narrative, your ICP, your competitive intel. April Dunford (arguably the GOAT of B2B marketing) built her entire case for positioning on one idea: it's context. The opening scene of a movie, set by your alternatives, your differentiators, your best-fit buyer. Lift a competitor's positioning wholesale and you've inherited a frame that has nothing to do with your story.

Once that foundation is in place, then you can go shopping. Take what's worked for companies you admire, whether it's channels, strategy, or creative, and hold each one up against your own situation to see how much of it actually fits. Sometimes it's a full eclipse, a clean overlap where their move maps almost perfectly onto your context. More often it's a Venn diagram, where a slice of it works and the rest doesn't. And sometimes, no matter how well it worked for them, it just won't work for you. The skill is knowing which of the three you're looking at before you spend more time or money.

Steal the instinct behind what worked, not the execution, then go run your own version.

‍

One guardrail sits under all of this. Run a high volume of small bets and a few genuinely big swings, but never the bet-the-company kind.

You have to survive long enough to get lucky, and luck tends to take about ten times longer to show up than you expect it to in the startup world. Staying alive is the whole game, and for most startups it is the hardest part of all. Quietly doing no real marketing at all is the same existential risk from the other side, just spread so thin you don't feel it until it's late.

I mentioned Blair's books up top. The three are The Win Without Pitching Manifesto, The Four Conversations, and Pricing Creativity. They look like they're about selling and pricing creative work, but what they are really about is conviction, and I think the message is an important one even if you are a tech founder and not a creative.

I see the irony in writing a four-point framework about embracing the mess that is startup marketing. The mess isn't an excuse to skip the thinking. It's permission to stop pretending you can forecast which swing or shot will connect. Do the foundational work first: your story, your narrative, your ICP, your research, your competitive intel. Then pick one or two things and run them all the way, or at least experiment with that intent. See if they land.

Push hard if they do, or walk away, fast, if they don't. Most days, how quickly you decide matters much more than how clever the plan was.

Stop doing ten things kind of well.

‍

Yours in marketing, Jeff

‍