One axis: your industry. The other: the behavioural mechanism. One pattern, ten reads. Every card is a DEMO - an illustration of the reading format, not a real run.
Mechanisms are hypotheses, not guarantees - behavioural effects vary by population and context. Atlas ranks which honest message to test first; it does not promise the effect, and it is not a manipulation kit: we test messages that help people do what they already intend to do.
Industry
Mechanism
SAAS · PRICING
Which of three pricing pages do we ship?
You have Basic/Pro/Team. Variant A: three tiers priced independently. Variant B: adds a deliberately dominated middle option. Variant C: annual framing first. Atlas runs all three on 1,000 simulated buyers in your segment.
WINDROSE ATLAS · DEMO RUN
EVIDENCE LABEL
SIM-RANK · DIRECTIONAL
READING
B > C > A · stability 76/100
TOP HYPOTHESIS
The dominated middle makes Pro feel like the obvious choice (decoy / asymmetric dominance).
VERIFY NEXT
A 2-week A/B on the live pricing page. Decision metric: paid conversions, not clicks.
Mechanism Anchoring + decoy
A compass, not a verdict.
FINTECH · IN-APP MESSAGE
Which push gets users to turn on auto-saving?
Variant A: “Save 200 zł monthly.” Variant B: “Don’t lose 2,400 zł a year.” Variant C: “People like you save 8% of income.” Atlas tests them on a simulated panel parameterised to your user demographics.
WINDROSE ATLAS · DEMO RUN
EVIDENCE LABEL
SIM-RANK · DIRECTIONAL
READING
B > C > A · stability 71/100
TOP HYPOTHESIS
Loss framing outweighs equivalent gain framing in this segment (loss aversion).
VERIFY NEXT
Ship top-2 as an in-app experiment; measure activated auto-saving after 30 days, not taps.
Mechanism Loss aversion / framing
A compass, not a verdict.
INSURANCE · OFFER ARCHITECTURE
Which cover should be the pre-selected default?
Three offer flows differ only in the default: minimal cover, recommended cover, or no default (forced choice). Atlas reads how each shifts choices and drop-off in your applicant population.
Split test in the quote funnel; watch completion AND downstream complaints - an honest default must survive both.
Mechanism Default / status quo
A compass, not a verdict.
RETAIL · CHECKOUT OFFER
“Free shipping” or “10% off” - which moves your cart?
Economically similar, behaviourally different. Variant A: 10% discount. Variant B: free shipping over threshold. Variant C: free returns badge. Atlas ranks them for your basket sizes and shopper profile.
WINDROSE ATLAS · DEMO RUN
EVIDENCE LABEL
SIM-RANK · DIRECTIONAL
READING
B > A > C · stability 79/100
TOP HYPOTHESIS
“Free” is processed as a category, not a number (zero-price effect); threshold also lifts average basket (goal gradient).
VERIFY NEXT
A 3-week storewide test; metric: contribution margin per order, not conversion alone.
Mechanism Zero-price effect · Goal gradient
A compass, not a verdict.
HEALTH · PUBLIC MESSAGE · 5 VARIANTS
Five invitation letters to a screening test - which goes to print?
A: gain frame (“early detection saves…”). B: loss frame. C: social norm (“7 in 10 neighbours attended”). D: implementation intention (“choose your day and hour now”). E: authority (local physician signature). Atlas reads all five across age and education segments - this is where multi-variant beats a coin toss.
WINDROSE ATLAS · DEMO RUN
EVIDENCE LABEL
SIM-RANK · DIRECTIONAL
READING
D > C > A > E > B · stability 66/100 · segment note: C leads in the 60+ segment
TOP HYPOTHESIS
Concrete planning beats persuasion for already-willing recipients (implementation intentions); norms matter more with age.
VERIFY NEXT
A regional pilot with the health provider; metric: booked appointments.
Mechanism Social proof · Commitment & consistency · Loss aversion / framing · Authority
A compass, not a verdict.
HEALTH · ADHERENCE
Which reminder keeps patients on their medication schedule?
Variant A: neutral reminder. Variant B: streak (“day 14 of your plan”). Variant C: reminder tied to an existing habit (“after brushing your teeth”). Atlas reads the variants for your patient demographics - as communication design, not as therapy.
WINDROSE ATLAS · DEMO RUN
EVIDENCE LABEL
SIM-RANK · DIRECTIONAL
READING
C > B > A · stability 72/100
TOP HYPOTHESIS
Anchoring the dose to an existing routine beats raw willpower (habit stacking / present bias).
VERIFY NEXT
An IRB-cleared pilot with the clinic; metric: refill and self-report adherence at 8 weeks.
Mechanism Present bias · Commitment & consistency
A compass, not a verdict.
FITNESS · RETENTION
What keeps January members past March?
Variant A: discount for a 6-month commitment. Variant B: visible streaks and milestones. Variant C: small refundable deposit on a weekly goal. Atlas ranks them for your member profile before you re-price anything.
Churn here is driven by confusion, not price; fewer choices lower the urge to re-shop (choice overload / status quo).
VERIFY NEXT
A save-desk script test on one region for 60 days; metric: 6-month survival, not same-day saves.
Mechanism Choice overload · Default / status quo
A compass, not a verdict.
PUBLIC SECTOR · CIVIC COMMUNICATION
Which waste-sorting message goes on 40,000 bins?
Variant A: instruction. Variant B: local norm (“83% of this district sorts correctly”). Variant C: consequence framing (costs of contamination). Atlas pre-tests on a panel parameterised to the city’s demographics - before the consultation, not instead of it.
WINDROSE ATLAS · DEMO RUN
EVIDENCE LABEL
SIM-RANK · DIRECTIONAL
READING
B > A > C · stability 73/100
TOP HYPOTHESIS
Local, specific norms beat generic instruction (social proof - the more local, the stronger).
VERIFY NEXT
A two-district field pilot with the city; metric: contamination rate at 8 weeks. The machine proposes. Residents decide.