The Recommendation Integrity Audit: What to Verify Before Referring a Reader
A due-diligence system for checking product facts, merchant operations, audience fit, evidence quality, and commercial incentives before recommending anything.
Pillar 12
Connect onboarding, support, refunds, disclosure, case-study standards, and long-term audience relationships.
Published guidance
A due-diligence system for checking product facts, merchant operations, audience fit, evidence quality, and commercial incentives before recommending anything.
Understand how funding, access, evidence, ranking, drafting, approval, disclosure, correction, and reader interests shape affiliate editorial independence.
Create a controlled path for topic origin, candidate set, evidence rules, access labels, conflict data, scoring, disclosure, approval, and correction.
Diagnose payout-shaped candidates, shifting criteria, merchant copy, asymmetric criticism, hidden paid placement, remote disclosure, and delayed correction.
Compare four formats by disclosure timing, proximity, repetition, truncation, visual hierarchy, relationship clarity, accessibility, reuse, and correction.
Understand how confirmation, delivery, prerequisites, effort, timing, progress, support, accessibility, recovery, consent, and feedback shape onboarding.
Map the promise, success state, transaction record, prerequisites, task order, time ranges, progress, help, recovery, feedback, privacy, and change triggers.
Diagnose ambiguous confirmation, surprise prerequisites, false progress, hidden help, inaccessible recovery, promise conflict, and stalled feedback.
Compare four onboarding models by complexity, risk, variability, volume, timing, accessibility, support capacity, handoff, recovery, and total burden.
Audit promise, state, record, next action, prerequisites, order, timing, progress, support, accessibility, privacy, recovery, ownership, and feedback.
Understand how intake, severity, ownership, status, channel fit, accessibility, handoff, escalation, resolution, and service learning create reliable support.
Design a case path for intake, consent, category, severity, channel, owner, target, status, handoff, escalation, resolution, appeal, and learning.
Diagnose blocked intake, repeated context, channel mismatch, ownerless handoffs, empty status, circular escalation, premature closure, and stalled repair.
Compare four support routes by urgency, complexity, evidence, privacy, accessibility, dialogue, authority, continuity, staffing, recovery, and failure cost.
Audit entry, reference, data, category, severity, channel, owner, target, status, handoff, escalation, accessibility, resolution, appeal, and learning.
Understand how policy, eligibility, timing, status, records, payment, access, return, exceptions, escalation, and correction shape post-purchase trust.
Map policy, jurisdiction, request, identity, eligibility, service stop, return, refund amount, payment status, records, exceptions, appeal, and learning.
Diagnose hidden entry, channel asymmetry, repeated retention, unclear states, moving eligibility, refund silence, inaccessible recovery, and premature closure.
Compare five resolutions by customer goal, service state, eligibility, timing, cost, continuity, data, evidence, reversibility, effort, and correction risk.
Audit policy, request access, identity, eligibility, service stop, return, calculation, payment, records, accessibility, exceptions, appeal, and learning.
See how purpose, selection, consent, baseline, intervention, outcomes, attribution, editing, relationships, privacy, and correction shape credible cases.
Create a workflow for selection, permission, privacy, baseline, intervention, evidence, attribution, editing, disclosure, follow-up, and correction.
Diagnose proof-seeking, selected winners, missing baselines, blended causes, weak outcomes, causal leaps, hidden relationships, and frozen correction.
Compare four feedback methods by decision level, depth, population, timing, burden, selection, privacy, aggregation, interpretation, action, and continuity.
Audit selection, permission, privacy, baseline, intervention, evidence, outcomes, attribution, edits, relationships, disclosure, follow-up, and correction.