Most agencies lose because they try to win the wrong fight. They look at the holding company across town — the one with the bigger team, the global footprint, the marquee clients, the awards shelf — and they decide the path forward is to become a smaller, scrappier version of the same thing. Better hustle. Sharper work. Faster turnaround. More heart. And then they wonder why, three years in, they're exhausted, undercharging, and still pitching against the same incumbent who somehow always seems to get the final call.
The problem isn't effort. The problem is that being a smaller version of your competitor is just losing more efficiently. If you're playing their game with their pieces on their board, the structural advantages they've spent twenty years accumulating will grind you down no matter how talented you are. The only way out isn't to play harder. It's to play a game they cannot follow you into.
That's what asymmetry actually means, and it's the most misunderstood word in strategy.
Cheaper Is Not Asymmetric
The first mistake people make is conflating asymmetry with being lean. Lean is good. Lean is necessary. But lean by itself is just a cost advantage, and cost advantages get competed away the moment the incumbent decides you're worth taking seriously. A holding company can discount an account for two quarters to crush a challenger. They can stand up a "lean innovation studio" inside their own building and staff it with their own people. They can match your price for as long as it takes to make you go away. If the only thing you have is a lower number on the invoice, you're not asymmetric — you're a temporary inefficiency in their pricing model, and they will correct you.
True asymmetry requires something different. It requires a structural feature of how you operate that your competitor cannot copy without ceasing to be themselves. That's the test, and it's a hard one. If the incumbent could match your move next quarter by reorganising a department or hiring three new people, you don't have asymmetry. You have a head start. The clock is already running on it.
The question to ask isn't "what do we do better?" It's "what is structurally true about our competitor that they cannot change, and how do we make the things they cannot change the reason they lose?"
What Big Agencies Cannot Change
Holding companies are not slow because their people are lazy. They're slow because slowness is the cost of coordinating hundreds of stakeholders across multiple offices and a six-layer approval chain. They're not expensive because they're greedy. They're expensive because their margin model assumes a certain billable rate, a certain overhead ratio, and a certain bench utilisation, and the whole machine collapses if any of those drift. They're not creatively cautious because the people inside them lack imagination. They're cautious because they have a brand to protect, blue-chip clients who do not want to be associated with anything weird, and shareholders who punish surprises.
These aren't flaws. They're the load-bearing walls of the building. You can't renovate them without the building falling down. Which means anything you build whose entire premise is "we do the thing those walls forbid" is something the incumbent cannot follow you into. They will describe it. They will write thought leadership about it. They will not actually become it, because becoming it would destroy them.
That's where you want to live.
What This Looks Like in Practice
Speed is the most obvious asymmetric axis, and the most underused. Not "we're a bit faster," but "concept to live in 72 hours, every time, no exceptions." That promise is structurally impossible for an agency whose legal review takes a week and whose creative director is in three other meetings. They cannot match it without firing half their account team and rewriting their contracts. So they don't. And if you've built your entire shop around that promise — pricing, staffing, tooling, client expectations, contract terms — you own a category they cannot enter.
Risk tolerance is another. A founder-led shop with no shareholders and no global brand to protect can make work that is sharper, weirder, more politically pointed, or more formally experimental than anything a holding company can sign off on. The big shop's clients would flee. Their internal review would kill it. Their PR team would have a heart attack. So they produce safe work, and safe work is invisible work, and invisible work is the entire reason their clients are quietly looking for someone like you. The asymmetry isn't that you're braver. It's that bravery is structurally available to you and structurally forbidden to them.
Cost structure is a third. If your overhead is one-twentieth of theirs, you can take work at prices that would be suicidal for them and still be more profitable than they are. They can match your price for one account, maybe two, but they cannot live there. Every month they discount to defend an account is a month they're funding the defence out of reserves that exist for other purposes. Eventually they stop. Eventually they let the account go, because the account was no longer profitable on their cost base anyway. You didn't beat them. You made the fight uneconomic for them and they walked away from it.
Definitions of victory matter too, and this one is the most overlooked. The holding company needs to win the pitch, bill the hours, hit the quarter, and renew the contract. Those are the only outcomes their P&L recognises. A founder-led shop can take equity instead of fees. Can lose money on a project deliberately to build a case study that wins ten more. Can work for a brand that pays in attention and audience rather than cash. Can refuse the pitch entirely and pursue work the incumbent would never see as work. When your win condition is genuinely different from your competitor's win condition, you're not really competing with them. You're foraging in a part of the ecosystem they don't perceive as food.
The Productisation Trap and Its Opposite
A lot of agencies have figured out half of this and stopped halfway. They've productised — fixed-price packages, defined scopes, clear deliverables — and they think productisation alone is the asymmetry. It isn't. Productisation is just a delivery model, and the holding companies are quietly rolling out their own productised offerings as you read this. If your only differentiator is "we have a price list," you're going to be matched within a year.
Productisation becomes asymmetric only when it's combined with something the incumbent's structure forbids. Public, transparent pricing on your website is asymmetric — not because the prices themselves are special, but because publishing them forces a level of operational discipline a holding company cannot replicate without exposing the chaos inside its own pricing. Working in public, sharing the process, showing the rough drafts, naming the clients who said no — these are asymmetric not because they're clever marketing tactics but because they would be career-ending inside any large agency, which means no large agency will ever do them. That's the moat. The moat is that the move would kill them if they tried it.
The Question That Replaces All Other Questions
Once you start thinking this way, most of the questions agency leaders agonise over become uninteresting. Should we hire another senior creative? Should we open a second office? Should we chase that RFP? Should we pitch for free? These are all optimisation questions, and optimisation is what you do when you've already accepted the terms of the fight. They assume the game is fixed and you're trying to play it slightly better.
The asymmetry question is different. It asks: what would I have to build so that my competitor's greatest strengths become the precise reason they lose? Their scale, their process, their margin model, their client relationships, their cautious brand — what would I have to be so that each of those becomes a liability instead of an advantage? The answer is rarely comfortable, because it usually means giving up things you thought you wanted. Smaller, weirder, more focused, more public, more opinionated, more willing to walk away. It means choosing a shape that the incumbent's shape cannot become.
But that's the only fight worth picking. Every other fight is one you've already half-lost the moment you agree to show up. The shops that actually displace incumbents are not the ones who tried harder at the same game. They're the ones who quietly built a different game next door, and waited, and let the old game's own weight do most of the work.
Incumbents don't fall because a challenger out-executed them on their own terms. They fall because someone built an operating model whose entire premise is something the incumbent's existence forbids — and by the time the threat is obvious, the assumption the incumbent was built on has already cracked, and there's no way to retrofit the building while it's still standing.
Build the thing they cannot become. That's the whole strategy. Everything else is detail.
