Individual HTML example · Social proof

Randomised recent-purchase notification

A fictional store displays “12 people bought this in the last hour” using a random number unrelated to completed orders.

Activity messagesProduct pagePricing

This is an original fictional comparison. It depicts no real business and is not a regulator finding, statutory category or legal safe harbour.

Product pagePricing

Randomised recent-purchase notification

A fictional store displays “12 people bought this in the last hour” using a random number unrelated to completed orders.

Potentially problematicPopular right now

Popular right now. A generated number is presented as recent customer activity.. 12 bought this hour. Claim basis, collapsed

Neutral alternativeVerified activity information

Verified activity information. The message states its period and is shown only when live order data supports it.. 8 purchases in the past 24 hours. Claim basis, expanded: The relevant population, period, inventory or offer state is presented with the message.. View the source and period for “8 purchases in the past 24 hours”

Why the first version can mislead: The interface asks the user to rely on other people’s behaviour or approval. The cue “12 bought this hour” has persuasive meaning only if its population, period, metric and relationship match the represented claim. The panel instead says: “A generated number is presented as recent customer activity.” Unsupported activity can create false confidence and urgency around the purchase.

What a fairer design does: Use activity messages only when they are accurate, timely, relevant and supported by a defined data source.

Show annotated differences (4)
Potentially problematic version
  • A generated number is presented as recent customer activity.Why this state matters: The interface asks the user to rely on other people’s behaviour or approval. The cue “12 bought this hour” has persuasive meaning only if its population, period, metric and relationship match the represented claim. The panel instead says: “A generated number is presented as recent customer activity.” Unsupported activity can create false confidence and urgency around the purchase.
  • Claim basis, collapsedReview prompt: Which source record substantiates “12 bought this hour”, including its population, period, metric and any commercial relationship?
Neutral alternative
  • The message states its period and is shown only when live order data supports it.Fairer design: Use activity messages only when they are accurate, timely, relevant and supported by a defined data source.
  • View the source and period for “8 purchases in the past 24 hours”Review prompt: Record the claim, timestamp, data source and post-claim state; which evidence supports “The message describes actions by other people as current or recent”?
Review questions (3)
  • Which source record substantiates “12 bought this hour”, including its population, period, metric and any commercial relationship?
  • Record the claim, timestamp, data source and post-claim state; which evidence supports “The message describes actions by other people as current or recent”?
  • Can the claim be reproduced from a defined population and period, or does the evidence instead fit “Static aggregate popularity claim”?
Side-by-side product page interface showing “Popular right now” and the clearer alternative “Verified activity information”.Provenance: Original hypothetical product page example created for this library; no real business, product or interface is depicted.

Complete practice context

How this example relates to Activity messages

Claims about recent or current actions by other users are used to steer the decision, particularly where authenticity, timing or relevance is unsupported. This is a design and research taxonomy for learning and evidence review, not a statutory offence label or an automatic finding that an interface is unlawful.

The full guide explains inclusion and exclusion criteria, potential harms, fairer design, evidence questions and separate Australian legal information layers. The example page isolates one scenario so it can be shared and discussed without reproducing the complete library on one page.

Continue comparing

Activity messages

Fabricated named signup notification

Course landing pagePricing

A fictional course page rotates messages such as “Mia from Berlin just enrolled” using invented names and locations rather than completed enrolments.

Open this example

Popularity claims

“Most popular” has no defined metric

Subscription plansPricing

A fictional service labels its highest-margin plan “Most popular” without recording sales, selection period or the population being measured.

Open this example

Popularity claims

Bestseller badge follows margin rather than sales

Marketplace resultsPricing

A fictional marketplace assigns “Bestseller” to whichever item yields the highest commission, regardless of units sold.

Open this example
Sources and provenance for this example

Evidence base

Sources

  1. Dark commercial patternsOrganisation for Economic Co-operation and Development · Secondary · checked 2026-09-14 · OECD Digital Economy Papers No. 336
  2. Competition and Consumer Amendment (Unfair Trading Practices) Act 2026Federal Register of Legislation · Primary · checked 2026-09-14 · C2026A00064