Multiple simultaneous changes affect the same system, making it impossible to determine which change produced which result. The cap exists because clean learning requires isolated variables, not because the team lacks capacity for more activity. When two initiatives are running in parallel, both must affect different metrics and different areas of the funnel so their results can be read independently. The discipline of choosing two forces the fCMO and founder to be genuinely selective about what matters most rather than optimising everything at a level of intensity that produces no clear signal.

Ideas that do not meet the intake criteria, because they are not grounded in a system signal, because the area has not been active long enough to evaluate fairly, or because they do not align with current strategy, are parked in an optimisation backlog. They are not rejected permanently; they are deferred until there is signal evidence to support them or a quarter in which they become a genuine priority. The backlog is reviewed at the start of every Quarterly Review. Many ideas that were parked in one quarter become valid candidates in the next, once the relevant metric has had time to produce a meaningful pattern.

Inconclusive results are documented and treated as a learning signal in themselves. The fCMO reviews why the result was inconclusive. whether the test ran for insufficient time, whether a confounding variable was present, whether the success metric was too ambitious for the sample size available, or whether the hypothesis itself needs refinement. The documentation captures all of this so the next iteration of the test, if one is warranted, starts from a better design rather than repeating the same conditions. Declaring an inconclusive result a "loss" and abandoning the question is one of the most common failure modes in optimisation; the Optimisation Loop is designed to prevent it.

No, and this boundary is explicit and enforced. Core strategy, positioning, brand identity, and primary offers are not in scope for the Optimisation Loop. The loop improves execution, messaging refinement, conversion efficiency, and system process, working within the strategic direction that Strategy First Plan established. Changing strategy through the guise of optimisation is one of the common failure modes specifically listed in the Facilitator's Guide and the fCMO's role includes identifying and stopping it when it surfaces. Strategic changes require their own review process, which typically begins with a revisit of Strategy First Plan.

The Optimisation Loop feeds its outputs directly back into the system. Performance improvements update the Campaign Builder inputs for the next quarter, campaigns are redesigned on top of proven learning rather than rebuilt from assumptions. Process changes update the Workstream Engine, SOPs and ownership structures are refined based on what the quarter revealed about how work actually flows. Structural changes update the funnel architecture and conversion paths. And the learning log from each completed initiative strengthens the Scorecard thresholds and signal interpretation rules. Over time, the loop makes every other step of the Marketing Operating System more precise, not just the specific initiatives it runs.

The Optimisation Loop requires at least Steps 1–5, Strategy First, the Workstream Engine, Campaign Building, and Scorecard and Signals, to be operational before it can function meaningfully. Without a strategy, there is no stable foundation to optimise against. Without a scorecard, there is no signal evidence to trigger intake. Without workstreams, there are no defined processes to update. The Momentum Meeting (Step 6) is highly recommended before the first Optimisation Loop runs, as the Monthly Insights Summaries from those meetings are core inputs to the Quarterly Review. The Productive Chat will confirm exactly which steps are already in place and which need to be installed before the loop begins.