Demo: fictional airlines, illustrative data.Back to the site
Campaign UpliftReleased· Released to you

New route: London to Los Angeles.

Did the launch campaign for the new non-stop Heathrow to Los Angeles route shift consideration among long-haul travellers?

Audience
UK adults 25 to 64 who have taken a long-haul return flight in the last three years
Fieldwork
21 Sept to 9 Oct 2026
Sample
300 per cell · detects ±10.5 pts
Primary KPI
Consideration, fixed at Gate 2
Step 7 of 7 · Released3 of 3 gates signed

Method lock

Locked before fieldwork.

The spec, survey version, primary KPI, sub-groups and analysis plan are hashed at Gate 2. If any of it changed afterwards, the hash would not match.

Locked 18 Sept, 11:05nm-london-la-survey-v1.2

SHA-256 20ca02bf37d3cf7f4668eec02dba5fc392b3ccaa24bb0e760e2df57c07cf14c2

Show the locked plan
{
  "design": "Campaign Uplift",
  "audience": "UK adults 25 to 64 who have taken a long-haul return flight in the last three years",
  "perCell": 300,
  "cells": [
    "exposed",
    "control"
  ],
  "primaryKpi": "consideration",
  "secondaryKpis": [
    "awareness",
    "recall",
    "favourability",
    "intent",
    "message"
  ],
  "subgroups": [
    "age band",
    "Flies long-haul twice a year or more"
  ],
  "test": "two-proportion z (pooled), Wald 95% interval, Holm adjustment across sub-groups and creatives",
  "minimumDetectableLift": 0.105,
  "exclusions": [
    "decoy-ad over-claim",
    "attention check",
    "speeder (< one third of median time)",
    "duplicate or fraud flag",
    "exposure not confirmed by media data"
  ]
}

Human gates

Who signed what, and when.

  1. Gate 1 · Researcher approves spec and survey

    Signed by SpinProof Research Desk, lead researcher, 2 Sept, 15:20

  2. Gate 2 · Analysis plan locked and hashed

    Signed by SpinProof Research Desk, lead researcher, 18 Sept, 11:05

  3. Gate 3 · Researcher signs the report

    Signed by SpinProof Research Desk, second reviewer, 14 Oct, 16:40

Agent run log

Every action, with the model or code version used.

Agents draft, recruit, check and write. Deterministic code does the statistics. Real people give every answer.

  1. Intake agent27 Aug, 09:12

    Turned the booking into study spec v1: Campaign Uplift, audience, KPIs and cells.

    claude-sonnet-5-5 · intake prompt v1.4

  2. Intelligence agent27 Aug, 09:31

    Built the context brief: competitor ads from Meta Ad Library, share of search baseline, AI answer visibility baseline, news and social scan.

    claude-sonnet-5-5 · intel prompt v1.2

  3. Questionnaire agent27 Aug, 10:02

    Built the survey from the Campaign Uplift template, with decoy ad, attention checks and consent screen.

    claude-sonnet-5-5 · questionnaire prompt v2.0

  4. Lead researcher2 Sept, 15:20

    Gate 1 signed: spec and survey approved with two wording edits.

    SpinProof Research Desk

  5. Channels agent3 Sept, 10:15

    Set up recruitment: Meta ad sets with screener, QR codes for placements, panel quotas, SpinProof panel invites.

    claude-sonnet-5-5 · channels prompt v1.3

  6. Analysis agent18 Sept, 11:05

    Locked the analysis plan. SHA-256 20ca02bf37d3…

    stats-engine 0.9.2, deterministic

  7. Lead researcher18 Sept, 11:06

    Gate 2 signed: plan locked before fieldwork.

    SpinProof Research Desk

  8. Fieldwork agent21 Sept, 08:00

    Fieldwork opened. Invites live on all channels; rewards queue active.

    claude-sonnet-5-5 · fieldwork prompt v1.1

  9. Verification agent22 Sept, 18:00

    Daily quality pass: fraud, speeders, attention and decoy checks run; exclusions logged with reasons.

    claude-sonnet-5-5 · verification prompt v1.6

  10. Verification agent26 Sept, 18:00

    Daily quality pass: fraud, speeders, attention and decoy checks run; exclusions logged with reasons.

    claude-sonnet-5-5 · verification prompt v1.6

  11. Verification agent30 Sept, 18:00

    Daily quality pass: fraud, speeders, attention and decoy checks run; exclusions logged with reasons.

    claude-sonnet-5-5 · verification prompt v1.6

  12. Verification agent4 Oct, 18:00

    Daily quality pass: fraud, speeders, attention and decoy checks run; exclusions logged with reasons.

    claude-sonnet-5-5 · verification prompt v1.6

  13. Verification agent8 Oct, 18:00

    Daily quality pass: fraud, speeders, attention and decoy checks run; exclusions logged with reasons.

    claude-sonnet-5-5 · verification prompt v1.6

  14. Verification agent9 Oct, 18:30

    77 completes excluded in total: decoy-ad over-claim 29, failed attention check 20, speeder 15, duplicate or fraud flag 5, exposure not confirmed 8.

    claude-sonnet-5-5 · verification prompt v1.6

  15. Fieldwork agent9 Oct, 20:00

    Quotas met in every cell. Fieldwork closed; rewards paid.

    claude-sonnet-5-5 · fieldwork prompt v1.1

  16. Analysis agent12 Oct, 09:00

    Ran the locked plan: every KPI, pre-registered sub-groups, Holm adjustment. No deviations from plan.

    stats-engine 0.9.2, deterministic

  17. Reporting agent12 Oct, 11:30

    Drafted the report from computed tables only. Every number linked to an analysis output.

    claude-sonnet-5-5 · reporting prompt v1.8

  18. Evidence agent12 Oct, 11:45

    Assembled the Evidence File. 0 untraced numbers. Charts re-checked against tested values.

    claude-sonnet-5-5 · evidence prompt v1.0

  19. Second reviewer14 Oct, 16:40

    Gate 3 signed: report checked against the Evidence File.

    SpinProof Research Desk

  20. Norms agent15 Oct, 09:00

    Added the study to the travel benchmark set.

    claude-sonnet-5-5 · norms prompt v0.9

  21. Portal15 Oct, 09:05

    Released to client portal with raw data, crosstabs and Evidence File.

    portal

Results trace

Every reported number, traced to raw answers.

Each number in the report links to the rows it came from and the analysis output that produced it.

ReportedSource rowsOutput
R1Consideration lift +9.7 pts (+2.2 to +17.1)
responses.csv: valid=1, consideration=1 by cell → 113/300 exposed, 84/300 control
analysis/kpis.json#consideration
R2600 verified people
responses.csv: count of valid=1
analysis/sample.json#valid
R3Decoy false-claim rate 4.3%
responses.csv: decoy=1 over all completes
analysis/quality.json#decoy
R4Lift at 3+ exposures +13.9 pts
responses.csv: cell=exposed, freq=3 against all control
analysis/frequency.json#3
R5Persons lifted 227,167
absolute lift × campaign reach of 2,350,000 (media plan)
analysis/impact.json#persons
S1Age: 18–34 +1.7 pts, Holm p 0.81
responses.csv filtered to sub-group
analysis/subgroups.json#0
S2Age: 35–54 +17.1 pts, Holm p 0.01
responses.csv filtered to sub-group
analysis/subgroups.json#1
S3Age: 55+ +7.0 pts, Holm p 0.68
responses.csv filtered to sub-group
analysis/subgroups.json#2
S4Segment: Flies long-haul twice a year or more +9.9 pts, Holm p 0.43
responses.csv filtered to sub-group
analysis/subgroups.json#3
S5Segment: Everyone else +9.3 pts, Holm p 0.19
responses.csv filtered to sub-group
analysis/subgroups.json#4

Downloads

Always included, never on request.

Demo data. Fictional airlines, illustrative results generated for this demo.