Not Every Framework Carries the Same Weight
Two agencies can run the exact same framework and get very different results when it turns out to be wrong.
Picture two shops that both design client work around the same idea: get the strategy right and execution takes care of itself. For the first, that’s one principle among several. They pressure-test the strategy, but they also have delivery systems, QA, and account managers catching problems downstream. For the second, the strategy doc is the work. Everything after it is just execution, so a flawed strategy ships straight through to the client. Same framework. Same conviction in it. But when it misfires, one agency catches the problem and the other delivers the wrong solution at full speed.
That difference is what I want to name. It comes down to how much weight you’ve put on the framework, not whether the framework is any good.
Three kinds of frameworks
Every agency runs on a stack of frameworks. How you diagnose a client’s problem. How you scope the work. How you decide what “good” looks like. What you believe actually makes the work effective. Most owners never sort these by consequence, so they treat all of them as roughly equal. They aren’t.
There are three tiers.
Load-bearing frameworks. Remove this one and the way you design and deliver work falls apart. Other decisions are bolted on top of it. How you scope, what you price the engagement at, what you promise in the pitch, how the team executes. When a load-bearing framework breaks, several things crack at once because they were all resting on the same beam.
Secondary frameworks. These carry real weight, but they aren’t the only thing holding the work up. If one fails, you feel it and things start to sag, but the engagement stays on track while you fix it because something else is also taking the load.
Decorative frameworks. These are there, and they might even help you think or explain your approach to a client, but nothing in the actual work depends on them. Pull one out and the only thing that changes is how the pitch deck looks. The frameworks you name-drop in a proposal but don’t actually use to make decisions usually live here.
A decorative framework isn’t useless, by the way. Some of them are good for communicating your thinking to a client. They just aren’t doing structural work, and it helps to be honest about which job a framework is actually doing.
How to tell which is which
You can’t sort your frameworks by how often you talk about them. Plenty of agencies reference their signature “methodology” constantly while it changes nothing about how they actually scope or build the work. That’s a decorative framework everyone treats as load-bearing.
The real test is counterfactual. Ask what breaks if this is wrong.
Take any framework you design work around and finish the sentence: “If this turned out to be false tomorrow, then ___.” If the answer is “then the way we scope and deliver every engagement stops working and we’d have to rethink the whole process,” it’s load-bearing. If the answer is “then we’d lose one input but the rest of the process would hold,” it’s secondary. If the answer is “then nothing about the work really changes,” it’s decorative.
The second thing to look at is coupling. A framework becomes load-bearing not because it’s impressive in the abstract but because other decisions are attached to it. So trace what’s downstream. If your scoping, your staffing, and your client promises all sit on top of one assumption about how the work should be done, that assumption is carrying three things, not one. When it shifts, all three move.
The combination that actually hurts you
A framework can carry a lot of weight and still be safe, as long as it’s been tested. The risky pairing is high load plus untested.
That’s the framework you lean on hardest and have never actually checked under different conditions. It worked on the kind of clients you started with, so you stopped questioning it, and now it’s load-bearing precisely because it’s invisible. You don’t treat it as a decision anymore, it’s just how you do the work.
For a lot of agencies, “our discovery process always surfaces the real problem” is exactly this. It was never really chosen as a method, it just happened to work early on, and the whole delivery model quietly arranged itself around it. That part of the process holds up every engagement, and it’s the last thing anyone thinks to pressure-test. You find out it was load-bearing the first time it gives you a wrong read and the entire project is already built on top of it.
What to do once you’ve sorted them
The fix usually isn’t a better framework, which surprises people.
When you find a load-bearing framework that worries you, you have three real options. You can reinforce it, meaning you pressure-test it on purpose, run it against the clients and situations where it might fail, and make sure it actually holds. You can take weight off it, meaning you stop hanging every scoping and delivery decision on the same single assumption. Or you can add a second support, meaning you build a secondary framework that can carry some of the same load so no single failure sinks the work.
Most struggling agencies don’t have a framework problem so much as a distribution problem. Everything they run on is either load-bearing, so one method holds up every engagement, or decorative, so the proposal looks rigorous but the work underneath isn’t actually supported by it. The healthy middle, a few secondary frameworks quietly sharing the weight, is thin or missing. (The same imbalance shows up on the business side, by the way. When “referrals will keep coming” is the one wall holding up the whole pipeline, you’ve got the exact same problem in a different room.)
So the exercise is simple, even if the answers aren’t. List the frameworks you actually design and deliver work with. Sort each one into load-bearing, secondary, or decorative. Then look hard at anything that landed in load-bearing and ask when you last tested it.
The framework you never think about is usually the one holding the most weight.
