From lesson 1, readable now without an account: Translating a business outcome into a digital marketing task
The task begins by separating a business outcome from a marketing activity the team can actually perform. “Increase revenue” is a business outcome, but it does not identify whom the specialist should address or which behavior must change. The specialist turns it into an actionable problem through four questions: who is the intended audience, which observable action precedes the outcome, which barrier can marketing influence, and what window can reveal change? When documented answers are unavailable, the statement remains a working hypothesis rather than a performance promise.
In the quote-request example, increasing all submissions is the wrong unit because management wants requests meeting three qualification conditions. The analysis unit is a qualified request, not a click or form submission alone. The specialist writes an operational definition: a request from a person in scope who states a compatible need and supplies the required contact details. The baseline is then fixed at 45 qualified requests out of 120 monthly requests. Total request volume becomes a guardrail because qualification rate could rise merely by collapsing overall volume, producing an attractive ratio that fails the business purpose.
Once the definition is stable, the team writes the proposed causal chain. The problem may be unsuitable reach, a vague promise attracting unqualified interest, or a form that hides service conditions. Each explanation implies a different intervention: targeting changes who sees the message, message changes expectations, and form design changes what a person knows before submitting. The team does not change all three simultaneously. It selects the explanation with the strongest available support and records what will change and what will remain constant. This makes interpretation possible without claiming more causality than the design supports.
A useful task card contains the business outcome, audience behavior, campaign deliverable, primary metric, guardrail and authority limit. Here, the deliverable could be a page and message that state fit conditions before submission; the primary measure is qualified requests during eight weeks; guardrails are total requests and cost per qualified request. The specialist does not use “raise awareness” unless the observable behavior representing awareness and its relationship to qualification are defined. Nor is the metric replaced after results appear, because retrospective substitution turns a decision rule into a justification.
Before handover, the specialist runs three cases. The ordinary case checks whether all three qualification conditions are collected consistently. The boundary case handles a partial request or an unknown condition without forcing it into pass or fail. The negative case changes the qualification definition midway through the window; the break must be marked, and the two periods cannot be compared as homogeneous. The test record preserves input, expected behavior, actual behavior and action, together with denominator, window and source so another practitioner can reproduce the calculation.
The specialist closes with a bounded recommendation: problem, definition, baseline, hypothesis, proposed change, measurement, guardrail and the decision required from the authorized owner. Price is outside authority in this case, so a discount is not presented as an implementation detail. If the outcome depends on a commercial change or operating capacity beyond the campaign, that dependency is stated and escalated rather than hidden inside a marketing objective. Task quality is therefore demonstrated by the team’s ability to execute one intervention, interpret its result within limits and name the fact that would change the decision.