Developing strategic judgement

Previous posts outlined why marketing science faces structural limits. Complete information is unavailable. Precision often creates false certainty. Optimization occurs in systems that continuously adapt.

The implication is not to abandon analysis or prediction. It is to stop treating prediction as the primary objective.

In complex environments, prediction and decision understanding serve different roles. Prediction estimates what might happen if assumptions hold. Decision frameworks explain why people choose, how those assumptions form, and when they change. Used together, they provide a more complete picture than either can alone.

That is why many of marketing’s most useful tools are not narrow predictive models, but decision heuristics. They operate under uncertainty, incorporate factors that resist measurement, and support judgment rather than control. They do not replace forecasting. They shape how forecasts are interpreted and acted upon.

A practical example: There has been extensive discussion about rising CAC and payback periods in B2B SaaS. Most responses focus on how to return to prior performance levels. Viewed through historical numbers, that instinct makes sense.

But decision theory suggests that it is unlikely we will ever be able to return to that level of performance.

Value is reference-point dependent. Early SaaS buyers compared software to spreadsheets or manual processes, creating large perceived gains. Today, most buyers already have tools. The reference point has shifted. What once felt like a major improvement now feels incremental.

That shift rarely appears in dashboards. Reference points are not tracked. They cannot be directly measured. Yet they fundamentally change acquisition economics. Increasing spend to chase past benchmarks often means chasing a smaller perceived value pool.

When CAC rises, the question is often not how to optimize spend, but how perceived value and belief have changed.

Why heuristics matter: My own models failed not because the analysis was wrong, but because conditions changed. Tests worked, then stopped working. Success did not scale. The error was treating influence as control.

Decision heuristics exist because control is unavailable. They help operate where key variables resist measurement and adaptation outpaces models.
Hayek’s critique was not that mathematics is useless. It was that mistaking precision for complete knowledge leads to failure. Decision frameworks avoid that mistake by forcing explicit reasoning about belief, perceived costs, trust, and decision thresholds.

That is not less rigorous than prediction. In adaptive systems, it is more rigorous.

The objection that decision frameworks are not predictive is not a weakness. It is the point. When uncertainty cannot be eliminated, the goal is not control. It is judgment. Prediction estimates probabilities. Decision understanding guides action when reality diverges from the model.
Precision alone is not sufficient.