"Systems-first." "A holistic growth system." Everyone uses the language now, which has drained it of meaning. For most companies, the "system" is a slide: a circle with five arrows, presented once, then abandoned the moment real work begins. The word is doing marketing for an approach that isn't actually a system at all.
A real system is narrower and more demanding than the slide suggests. It has properties you can check. If those properties aren't there, you have a strategy document, not a system, no matter what the deck calls it.
section 01connected, not just coordinated
The first property is connection. In a real system, each part's output is the next part's input. The audit produces the findings that feed the roadmap. The roadmap produces the priorities that direct the work. The work produces the results that get measured. The measurement feeds back into the next audit. Each stage hands something concrete to the next.
The counterfeit version is coordination: separate activities that a person manually carries between, copy-pasting from the audit doc into the strategy deck into the content calendar. That's a project manager doing the connecting by hand, and it breaks the moment they look away. Connection that depends on someone remembering to connect isn't a system. It's a to-do list.
section 02a loop, not a line
The second property is that it loops. A strategy has a beginning and an end: research, plan, execute, done, until next quarter when you start over from roughly zero. A system doesn't end. The output of the last stage becomes the input of the first, so it runs continuously, and the work compounds because nothing resets at the engagement boundary.
This is the difference between a program that gets stronger every month and one that relitigates the same starting point every quarter. Lines stop. Loops compound.
section 03a record you can check
The third property is accountability. In a real system, every decision carries a reason, and every prediction can be checked against what actually happened. Why this priority and not that one? What did we expect, and did it occur? A system that can't answer those is just opinion with extra steps.
This is the unglamorous part most "systems" skip, because keeping an honest record of what you decided and whether it worked is harder than producing a confident slide. But it's the part that lets the loop actually improve instead of just repeating.
section 04how to tell the difference
You don't need to audit the architecture to know which one you have. Ask three questions. Does each part's output actually feed the next, or does a person carry it across by hand? Does it keep running on its own, or stop the moment attention moves? Can it tell you why it made a decision and whether the decision was right?
Most "growth systems" fail all three and survive on the strength of the slide. The real ones are quieter about the language and stricter about the properties. The word is easy to claim. The loop is the thing that's hard to build, and it's the only part that matters.