A weighted model for ranking the activity backlog. Score each candidate against the four factors below, then apply the operating constraints to land on the actual queue.
Expected revenue, conversion, or KBO-linked outcome at full traffic.
The most important factor. Estimate the lift × reach × value of a conversion at full traffic. A 1% lift on the checkout page usually beats a 10% lift on a low-traffic landing page — don't let percentage points fool you.
Effort to implement, including DOM stability and integration dependencies.
Heavy DOM manipulation, integration dependencies, or custom code should lower priority unless the impact justifies the cost. Easy wins beat hero projects when the backlog is long.
Are the audiences, traits, and tracking already in place?
Activities that need new audiences in RTCDP, new mbox parameters, or new analytics events take weeks longer than the build itself. Score down anything that depends on data work that hasn't started.
Compliance review surface area — disclaimers, consent, contact frequency.
In regulated industries, anything that changes messaging, contact cadence, or data usage needs early legal/compliance sign-off. High sensitivity adds calendar time even when the build is trivial.
Don't book the whole roster on new ideas.
Programs that always launch new activities and never iterate on winners leave most of the value on the table. Hold back roughly 30% of capacity for following up on activities that won, lost inconclusively, or hinted at a deeper insight.
A short checklist before an activity enters the build queue.
Audience: Confirm the audience exists (or can be built) in Adobe Target / RTCDP within the activity's timeline.
Implementation: Confirm AEM modular compatibility (or wherever the page lives), so the variant build doesn't hit avoidable DOM-stability risk.
Measurement: Confirm the analytics tagging plan covers primary, secondary, and guardrail metrics before launch — not after.