An owner told me last year that he’d collected eleven growth frameworks and his agency hadn’t moved in eighteen months. He could recite all of them. He’d bought two of them. When I asked which one he was running that week, he went quiet for a second and then said he supposed he was running bits of several, which is the honest answer and also the reason nothing was happening.
I’ve been on the other side of that. I’ve built models that turned out to be a nicer way of describing what I already believed, which is a comfortable thing to do and produces no change at all.
A growth framework earns its keep when it tells you what to stop doing, survives being taught to a competitor, names where it doesn’t apply, and changes your Monday rather than your vocabulary. Most of what gets sold as proprietary methodology fails at least two of those.
Quick Take
- The test that matters isn’t whether a framework is true, because most of them are true, and true is a low bar when you’re choosing between eleven of them.
- A framework that only adds work is a checklist wearing better clothes. The useful ones subtract, and subtraction is the part that feels expensive.
- If teaching your whole method to a competitor would sink you, the method was never the advantage, so the fear itself is diagnostic.
- Collecting frameworks feels like progress because it produces the sensation of understanding without requiring you to give anything up.
What this covers: what a growth framework is supposed to do, the four questions I run on any model including my own, how to spot renamed best practices, why owners collect frameworks instead of running one, and which of mine I’d start with.
What is an agency growth framework supposed to do?
A growth framework is a repeatable way of deciding what to work on next, and that word “deciding” is doing all the work. Most models sold to agency owners describe a business rather than deciding anything about it, so you finish the diagram feeling clearer and Monday looks the same as it did before.
I wrote a while back, in the context of differentiation, that a methodology is scaffolding for how you work and communicate rather than a moat, and that somebody can screenshot your framework, rebuild it in Figma, and have it on their site by Tuesday. That’s still my position, and it cuts both ways. If your own framework isn’t a moat, then the framework you’re evaluating isn’t one either, so you can stop shopping for the model that will make you uncopyable and start asking the smaller, more useful question of whether this particular one will change a decision you’re currently getting wrong.
The four questions I run on any growth framework
These apply to models I built and models I didn’t. I’ve had my own fail them, which is how the list got written.
Does it tell you what to stop doing? A framework that only adds is a checklist, and checklists don’t create growth because agency owners are not suffering from a shortage of things to do. The Constraint Map exists to answer this one: it names seven places an agency can be stuck, four on the demand side and three on the supply side, and its output is permission to ignore six of them this quarter. Owners find that harder than the diagnosis, because ignoring six things you know are imperfect requires believing the seventh matters more.
Would it survive being taught to a competitor? Take the whole model, hand it to the agency across town, and ask whether you’d be in trouble. If yes, you were relying on secrecy, and secrecy has a short shelf life in a market where everyone reads the same posts. I publish the 11 Differentiation Levers in full for this reason. The levers are easy to read and hard to run, because choosing one means giving up the others, and most agencies won’t make that trade no matter how clearly you explain it.
Does it name where it doesn’t apply? A model that claims to work at every size and stage is describing something too general to act on. Delivery Drag is a sub-$1M pattern, and I say so, because above that line the same symptoms usually have a different cause and the fix I’d prescribe would be wrong. When a framework has no stated boundary, that’s usually a marketing decision rather than a modeling one.
Does it change Monday, or only your vocabulary? This is the one that catches the most models, mine included. After you apply it, something on your calendar should be different by the end of the week. If the only change is that you now have better words for the situation you were already in, you bought a vocabulary. Better words help you brief a team and explain yourself to a client, and they don’t move revenue on their own.
How do you spot renamed best practices?
Most “proprietary frameworks” are relabeled versions of things that already exist, and the tell is that you can strip the branded names out and the advice underneath is advice you’ve read before. Try it on anything you’re considering. Replace every capitalized term with a plain description, then read what’s left. If the remainder is “get clear on who you serve, charge more, follow up,” you’ve found a naming exercise.
That’s not a reason to distrust everyone selling a model. Naming a pattern that nobody had named is useful work, because a named thing can be discussed, delegated, and noticed earlier next time. I coin terms constantly for exactly that reason. The distinction worth holding is between a name attached to a mechanism and a name attached to a familiar instruction, and the mechanism is the part you’re paying for.
The second tell is the absence of a cost. Every framework that’s doing work asks you to give something up, whether that’s a service line, a client segment, a channel you’re comfortable in, or the belief that you can serve everyone competently. A model that promises addition with no subtraction has removed the hard part, which is also the part that would have produced the result.
Why do owners collect frameworks instead of running one?
Because collecting produces the feeling of progress at none of the price. Reading a new model gives you a genuine hit of understanding, your business makes more sense at 9pm than it did at 8, and nothing has been risked. Running one costs you optionality, and optionality is the thing owners guard hardest when revenue feels thin.
There’s a version of this I see in coaching that has nothing to do with intelligence. An owner will understand a model completely, agree with it out loud, and then not run it, because running it means telling a client no, or dropping the service line that pays for August. The framework was never the blocker, and buying a twelfth one won’t touch the actual blocker, which is a decision sitting unmade. If you’ve got three of those sitting in your business right now, you’re in good company and you’re also the reason the eleven frameworks aren’t working.
I’d rather you ran one imperfect model all the way through than understood four perfectly. The one you run teaches you things the diagram can’t, because the micro-decisions that make a method work are the ones nobody writes down.
Which of my frameworks would I start with?
Depends on what’s broken, and the honest sequence starts with diagnosis instead of a favorite. Run the Constraint Map first, because everything downstream is guesswork until you know whether your problem lives on the demand side or the supply side, and most owners guess wrong about which one they’re in.
If it lands on demand, the sequence runs through positioning before channels. The four layers of positioning, differentiation, branding and messaging sort out which layer is the weak one, since agencies rewrite messaging to fix problems that live two layers down. Then the Brand-Demand Spectrum picks channels by stage rather than by preference, and the Five-Room Map finds the rooms your buyers are already sitting in so you’re not building an audience from zero.
If it lands on supply, you’re in Delivery Drag territory, and the work is protecting marketing time from client work rather than finding a better channel.
Each of those has a boundary, and I’ve tried to say so inside each one. None of them is a moat. They’re decision aids, and a decision aid is worth exactly what it changes.
The Monday version
Take the framework you’re most attracted to right now, including any of mine, and run the four questions on it. Write down what it tells you to stop doing. If that line is blank, the model is describing your business rather than deciding anything about it, and you can put it down without guilt.
Then pick one and give it a quarter of running, with a decision made in week one that you’d have to reverse publicly if you changed your mind. That public reversibility is what separates adopting a model from admiring one, because a decision nobody saw you make is one you can walk back on a slow Thursday and never notice you did.
If you’re sitting on a pile of frameworks and an agency that hasn’t moved, that’s a conversation we have regularly inside the Dynamic Agency Community, and you’d be surprised how many owners in there arrived with the same pile.
Where to go next
And if the goal is a business that runs without you in every seat, how to build a scalable digital agency covers the structural side.
FAQ
What is the best growth framework for a marketing agency?
There isn’t one, and the question usually means “which model will make the decision for me,” which no model does. Start by diagnosing whether your constraint is demand or supply, because a positioning framework applied to a capacity problem produces a very clear plan for the wrong thing. Once you know which side you’re on, almost any competent model for that side will outperform the perfect model for the other one.
How can I tell if an agency framework is just repackaged advice?
Strip out every capitalized term and read what’s left in plain language. If the remainder is advice you’ve seen before, the value was in the naming. Also check whether it asks you to give anything up, since a framework that’s all addition has removed the expensive part, which is usually the part that works.
Are proprietary frameworks a real competitive advantage?
No. Someone can copy the diagram in an afternoon. What they can’t copy is the judgment about when to deviate from it, which is why you can publish a method in full and still outperform the people reading it. Treat your framework as a credibility and alignment tool rather than protection.
How many growth frameworks should an agency be running at once?
One, with a diagnosis in front of it. Running several at once produces the thing I see most often in coaching, where an owner is executing fragments of four models and none of them long enough to know whether it worked. Give one a full quarter before you judge it.
Why do I understand these frameworks and still not grow?
Usually because the blocker is a decision rather than a concept. Write down the three decisions you’ve been carrying for more than a quarter, and check whether you genuinely know the right answer on all three. If you do, the missing piece is willingness rather than clarity, and no additional framework will supply it.