Evidence Ledger

Each global claim links here. Rows include domain, measurement, effect size, window, sources, and confidence level.

How to read a row

  • Effect size always in percentage points for the target behavior, not proxies
  • Window is explicit (for example D0–D7)
  • Sources link to a case write‑up and raw data when possible
ID Claim Domain Effect size Window Confidence Sources
BS-0001 Behavior-first validation during planning improves initiative success compared with assumption-first approaches. Cross-industry To be filled per case 6–18 months Working Case
BS-0002 Shorter time to first target behavior correlates with higher medium-term retention. Consumer and SaaS To be filled per case Baseline to 90 days Working Method
BS-0003 Real-world ‘nudge’ interventions typically have small average effects (~1–2%), far below academic claims (~8–9%). Public sector and enterprise ~1.4% average (vs ~8.7% published) Varies by program High Study
BS-0004 Opt-out (default) organ donation regimes do not reliably increase transplantation; complementary system investments drive outcomes. Healthcare policy No significant difference opt-out vs opt-in across countries; context-dependent Cross-national, multi-year comparisons High Review, Study
BS-0005 Pivoting from low-fit behaviors to high-fit behaviors drives rapid adoption (Instagram: check-ins → photo sharing). Technology Rapid user growth (e.g., 1M in 67 days) following behavior-focused redesign Launch window; first 60–90 days High Interview
BS-0006 Behavior-first internal tool pivots (Slack) yield high early adoption and strong cohort retention when they solve real team behaviors. Enterprise software 8k users day-1; 93% team retention after 2+ weeks (company-reported) First 30–90 days of public beta Working Company post
BS-0007 Mobile money infrastructure (M‑PESA) enables financial behaviors at population scale when environmental bottlenecks are removed. Finance / Emerging markets Majority adoption within ~3 years; RCTs show 80%+ adoption under agent access; 2% poverty reduction 3–5 years post launch Working Field/RCT
BS-0008 Personalized automation that removes choice overload (Spotify Discover Weekly) increases sustained discovery behaviors. Technology / Media >20% → >30% of listening from recommendations after feature launch (reported) First 12–24 months Working Engineering blog
BS-0009 Behavior scaffolding (micro-lessons, progress visibility, immediate feedback) enables durable learning behaviors (Duolingo). Education / EdTech Significant MAU growth; improved retention post-2020; next-day retention improved from ~2012 baseline 12–36 months Working Company/PM
BS-0010 Frictionless access at scale (Zoom) drives massive increases in meeting participation when environmental constraints spike. Enterprise / Communication 10M → 300M daily participants (Dec 2019–Apr 2020); TTFB near-instant Q1–Q2 2020 Working Company
BS-0011 At-scale ‘nudge’ interventions show small real-world effects (~1–2%), far below published averages in many domains. Public sector / General ~1–2 pp at scale (vs ~8.7 pp in journals) Varies by program High Meta-analysis, Study
BS-0012 Commuter carpooling RCTs show context-bound effects; system design (safety, coordination, incentives) matters more than default prompts. Transportation / Public policy Varies; contingent on study design and system supports Trial-specific Working RCT
BS-0013 Home energy reports (Opower) generate small, persistent behavior changes; effects are modest and context-dependent. Energy / Residential Small average reductions; persistence with decay 12–24 months Working Quasi/RCT
BS-0014 Eliminating transaction fees (Robinhood) removes economic friction and enables first-time investing behaviors for new segments. Finance Large platform growth; industry-wide fee elimination followed 2013–2019 Working Industry
BS-0015 Reputation and trust systems (Airbnb) enable high-stakes sharing behaviors; conversion improves when trust cues are salient. Technology / Marketplaces Conversion deltas vary by study; generally positive 12–36 months Working Observational/Experiment
BS-0016 Behavioral public strategy connects micro-foundations (individuals, teams, tools) to meso-level performance; simple nudges are brittle without system design. Public policy Not applicable; conceptual synthesis Not applicable High Article
BS-0017 Temporal design of measurement (windows, cadence) materially affects whether change is detected; precise temporal hypotheses improve detection. Methods / Measurement Not applicable; meta-analysis of temporal aspects Not applicable High Meta-analysis
BS-0018 Self-identity constructs add predictive power for behavior beyond attitudes and norms; identity fit improves outcomes. Psychology / Theory Not applicable; incremental predictive validity Not applicable Working Meta-analysis
BS-0019 Minimum-component (bottleneck) logic: the lowest-scoring determinant constrains behavior; addressing bottleneck components is necessary for reliable change. Methods / Theory Not applicable Not applicable High Model
BS-0020 Values-affirmation (identity-aligned) interventions yield durable academic benefits versus reminders alone. Education Context-dependent; durable effects reported in field studies Multi-year Working Field
BS-0021 Psychological targeting (personality-tailored messaging) outperforms generic messaging for persuasion and conversion. Marketing / Persuasion Positive treatment effects; varies by context Campaign-specific Working Experiment
BS-0022 Product failures frequently cite ‘no market need’ as the primary reason (post-mortems). Product / Strategy ~42% of post-mortems (CB Insights) Various; cross-company High Report

How to add a row

  1. Add an entry to /_data/evidence_ledger.yml.
  2. Use {% include evidence-ref.html id="BS-XXXX" %} next to the claim in your page.
  3. Keep denominators explicit and link to case write‑ups.

How to read a row (example)

ID Claim Domain Effect size Window Confidence Sources
BS‑0001 TTFB < 5 min increased completion of [behavior] among new users B2C mobile Δ‑B +14 pp (22→36) D0‑D7 post‑exposure Medium Case write‑up, Raw data