Evidence Plan

The plan that says how each reading will be believed: the design its value is established by, the rule that divides its credit where it is attributed, the way it must move, every parameter the plan has to fix and where its value will come from, and the job that pins one reading to a single definition two teams cannot report differently. Two tables and one job.

You leave with An evidence plan: every reading with the design that establishes it and the rule that divides its credit, every parameter with its unit and the design that calibrates it, and one definition per reading, so a number is trusted for a stated reason and not because it was reported.

Open on the board →

Why it matters

What goes wrong without it, as the commitments that license this deliverable say it, each standing on its sources.

• A deliverable is a method the industry already writesA plan is the thing a marketer is judged on, and a plan whose sections were chosen by whoever built the tool is an opinion with a table of contents. The industry has written these methods down: the measurement model that runs from objectives to KPIs to targets to segments, the client brief the agencies’ bodies agreed, the outcome standards that say what an instrument must disclose, the statechart a lifecycle is a machine of, the experiments that establish cause, the regulation that says on what ground a person may be reached. Citing them on the deliverable, and deriving its why from them the way a kind’s is derived, is what lets a reader see the method behind the page and what lets the composition of each plan be corrected against its source rather than argued.
  Digital Marketing and Measurement Model · The Goal Question Metric Approach · Aligning Organizations Through Measurement: The GQM+Strategies Approach · The Client Brief: A best practice guide to briefing communications agencies · Outcomes and Data Quality Standards · XDM Individual Profile Class · Statecharts: a visual formalism for complex systems · State Chart XML (SCXML): State Machine Notation for Control Abstraction · Trustworthy Online Controlled Experiments: A Practical Guide to A/B Testing · A Comparison of Approaches to Advertising Measurement: Evidence from Big Field Experiments at Facebook · Regulation (EU) 2016/679, Article 6 · Regulation (EU) 2016/679, Article 7: Conditions for consent · Consent Receipt Specification 1.1 · Transparency and Consent Framework v2.2

Tables— 2

Each is a view drawn across every term of the kind it starts from, on one page, with the file to take away.

• tableEvidence
  One line per reading, with the design its value is established by, the rule that divides its credit where it is attributed, and the way it must move.
• tableSettings
  One line per parameter, with the unit it is measured in, the basis its default rests on, what it is a parameter of, the tactic it constrains, and the evidence design that calibrates it.

Jobs— 1

Each is a view done from one term you already have, with a page for every one it can be done from.

• from a readingSpecify a reading
  “Two teams report the same number differently. I need one definition: what it is made of, what establishes it, which way is better, in what form it reads and in what unit.”

Other deliverables

• Measurement Plan
• Campaign Brief
• Permission Plan
• Instrumentation Plan
• Lifecycle Map