Chapter 7 · Part Two: The Comfort Check
I gave myself a rule that made no sense at the time.
Whatever we say has to fit in two words.
Not a tagline. Not a positioning statement. Not a headline with a subhead underneath doing the actual work. Two words that a person could hear once, at a conference, half paying attention, and repeat correctly to somebody else at lunch.
There was no research behind that. I didn’t test it. It was arbitrary, and I want to be honest that it was arbitrary, because the useful thing about this chapter isn’t the rule. It’s what an arbitrary constraint does to a room.
The answer we landed on was Controlling Complexity.
Not asset management, which is what we sold. Asset management is a category, and a category is a shelf, and nobody has ever felt anything about a shelf. Complexity was the thing our customers were actually drowning in, and complexity has a property that categories don’t: it never goes away. It only accumulates. Every acquisition, every new tool, every cloud migration, every remote worker, every year.
So we weren’t selling an inventory. We were selling control over something that gets worse whether or not you do anything.
Two words got us there. A brief wouldn’t have.
Why designing backward is rare
Question three sounds obvious. Throw out the process, build from the outcome. Everybody nods at that, and almost nobody does it.
The reason isn’t discipline. It’s that the existing process is the path of least fluency resistance, and by now you know what that means. The current approach is the version everybody in the building can already picture. It requires no explanation, no new vendors, no unfamiliar line items, and no meeting where somebody has to say a sentence they’ve never said before.
A redesigned approach requires all of those. Which means the redesign isn’t competing with the old process on merit. It’s competing against the old process plus the enormous hidden subsidy of already existing.
You can’t argue your way past that subsidy. You have to remove the old process from the table before the thinking starts, which is what a constraint is for.
Constraint as an instrument
Here’s the mechanism, stated plainly.
An open brief produces the category mean. Ask a talented team for “a campaign that communicates our value,” and they’ll produce something competent, defensible, and recognizable, because with no constraint the most available answer wins, and the most available answer is the one the whole category has already converged on.
A constraint makes the available answer illegal. That’s all it does. It isn’t a creativity technique, it’s an elimination technique. Two words made every sentence we had ever written about ourselves unusable, which forced a search, and the search is where the work happens.
The constraint doesn’t have to be correct. It has to be strict. I’ve never been able to defend “two words” as the right number on any grounds other than that it worked, and I’ve stopped trying.
A menu that works
Pick one. Don’t pick three, because a work with three constraints produces nothing and a team that resents you.
- Two words. The whole value, in two. Brutal, and the one I reach for most.
- No product nouns. Nothing in the output may name the thing you sell, the category, or a feature. Forces you to describe what changes for the person rather than what you ship.
- No comparatives. No faster, smarter, more accurate, more integrated, better. Whatever survives is your actual position, and for most companies this constraint alone deletes eighty percent of the messaging document.
- Sound off. It has to be understood with no audio and no copy. This is the constraint that produced the vehicle in the next section.
- One sentence, said by the buyer. Write the sentence a customer would use to describe you to a peer. You aren’t allowed to use any sentence you’d say about yourself.
- The default channel is banned. Whatever you’d obviously do, you may not do. If your answer to every objective is a webinar, ban webinars for one cycle and watch what the team finds.
- One tenth. Design it for a tenth of the budget. Not as an austerity measure, as a search. The tenth-scale version is frequently better and almost always more distinctive, because expensive is a substitute for interesting and removing the money removes the substitute.
- It has to survive a bar. Somebody has to be able to repeat it, drunk, at a conference, three days later, to somebody who wasn’t there.
Vehicle selection
Once you have the idea, you have the harder problem, which is getting it into a large number of heads without twenty million dollars and four years.
At Axonius the answer didn’t come from cybersecurity. It came from a video I couldn’t stop thinking about.
Amy Bream is a CrossFit athlete, born with one leg. The video is her attempting a lift. She fails. She fails again. She gets emotional. She goes back. She makes it.
No voiceover. No explanation. Nothing telling you how to feel.
And every person who watched it understood it instantly.
That’s when it clicked, and it’s the most transferable thing in this chapter:
The vehicle doesn’t have to explain the idea. It has to embody it.
Our idea was that in the face of enormous complexity, the best don’t just overcome it, they find a way to give back. That sentence came before the campaign and it generated the campaign. Then we needed something that was that sentence rather than something that described it.
Which is how we ended up reaching out to Amy, and then to Simone Biles, in a category that at that moment was buying stock photographs of men in hoodies.
The agency that made it with us, Ruckus, has the whole campaign on its case study page. Two of the films are worth your time.

How to actually find one
Three rules, in order of how often they get broken.
1. Look outside your category. Always. Every vehicle available inside cybersecurity has been used by cybersecurity, which means it arrives pre-loaded with fluency and can’t produce the reaction you need. The hooded figure was a vehicle once. Sport, craft, music, food, architecture, nature documentaries, competitive anything: these are unexploited in our market for no reason other than that nobody here looks there.
2. Test it with the sound off. If the vehicle needs a voiceover to land, it isn’t a vehicle, it’s a container for your copy. The Amy Bream video works with the sound off. That isn’t a coincidence, it’s the qualification.
3. Test whether it embodies or decorates. Ask what happens if you remove the idea and keep the vehicle. If it still works fine, the vehicle was decoration and you have bought an association rather than built a story. A celebrity who could be swapped for any other celebrity is decoration. A person whose actual life is the argument you’re making is a vehicle.
The visibility test
Here’s the one I didn’t have when I needed it, and it’s the reason the next chapter exists.
Before you commit, ask this:
Is the different thing visible to the audience, or only to me?
The failure mode is specific and it isn’t the one people expect. You don’t usually get punished for doing something too strange. You get punished for doing something conventional that contains, somewhere inside it, one genuinely original element that nobody outside the planning room will ever perceive.
That project will feel different to you the entire time you’re building it. You’ll be able to explain, at length, exactly what makes it unlike anything else in the category. Every word of that explanation will be true. And the market will experience the conventional wrapper, because the conventional wrapper is the only part it can see.
The test is mechanical. Describe your plan to somebody outside your company in thirty seconds, the way a buyer would encounter it rather than the way you built it. Then ask them what’s unusual about it.
If they can’t name the thing you’re proud of, the thing you’re proud of isn’t in the plan. It’s in the deck.
I didn’t run that test once, on a project that cost real money, and Chapter 8 is what happened.
The same objective at three budget levels
The objection to everything in this chapter is that it’s easy to talk about vehicles when somebody has already given you a budget for one. So here’s the same objective, redesigned three ways.
Objective: a security buyer who has encountered us once can describe what we do to a colleague a month later, in their own words, correctly.
With no money
- Pick the two words and enforce them everywhere for two quarters. Every asset, every deck, every rep’s LinkedIn headline, every conference bio, every email signature. Consistency is free and almost nobody does it, because internal boredom sets in roughly nine months before external recognition does.
- Build one analogy and use it relentlessly. Chapter 10 is the method. The Camry cost nothing and outlived every paid campaign of that era.
- Change who speaks. Put the person practitioners already trust in front of the audience instead of the person whose title says they should be. Free, and frequently prohibited by nothing except habit.
- Run the repeat-back test monthly. Five people outside the company, one week later, what do we do. That’s your measurement and it costs an hour.
With a real but ordinary budget
- Everything above, plus:
- One vehicle, sized to what you have. Not a celebrity. A person, a place, a made object, a piece of research nobody else would bother to do. The qualification is the sound-off test, not the price.
- Concentrate rather than distribute. The comfortable move with a moderate budget is to spread it across six channels so that no single line item looks reckless. That’s how you buy invisibility in six places. One channel at a depth that registers beats six at a depth that doesn’t.
- Fund the measurement. A baseline study isn’t the expensive part of anything in this list and it’s what converts next year’s conversation from an argument into a report.
With a large budget
- Everything above, plus:
- A vehicle at a scale that makes people ask what it is. That’s Chapter 11.
- And a warning that belongs here rather than there: a large budget makes it possible to be expensive instead of interesting, and expensive is much easier. Most of the worst marketing I’ve seen in this industry was extremely well funded. The constraint list above doesn’t stop being useful when you can afford to ignore it. It becomes the only thing keeping you honest.
The Comfort Check
Chapters 5 through 7: Axonius, complete
1. The comfortable assumption. Bigger company, more devices, bigger problem, bigger budget. Sell enterprise.
2. The true objective. Get known, at a scale enterprise sales can’t reach.
3. The redesigned approach. Two words, and a vehicle nobody in security would buy.