If a comparison table lets users directly adjust simple ‘levers’ in ranking transparency—such as sliders for freshness vs relevance weight or a toggle to down-weight sponsored boosts—how do these controllable explanations affect over‑ vs under‑trust, decision confidence, and merchants’ incentives to optimize for genuine fit rather than paid prominence?

Answer

Controllable ranking levers tend to (a) reduce blind over‑trust in the default ranking and slightly increase under‑trust among more skeptical users, (b) raise decision confidence when lever effects are clear and reversible, and (c) nudge merchants toward improving genuine fit and freshness when paid boosts are visibly bounded by user-adjustable controls.

User trust and confidence

• Over‑trust ↓: Seeing sliders/toggles makes the system look configurable, not authoritative; users treat the default as a starting point, not “the truth.”
• Under‑trust ↔/↑ slightly: A minority infer hidden bias (“why do I need to down‑weight sponsorship at all?”). If lever effects are opaque or noisy, this turns into broader skepticism.
• Calibrated trust ↑ when:
  • lever changes produce fast, local, and predictable table shifts, and
  • per-item snippets reference the current settings (“high because fresh + matches X you set; sponsored impact limited”).
• Decision confidence ↑ mainly for engaged users who adjust levers and see outcomes stabilize; others ignore controls and behave like in a normal transparent table.

Effects of specific levers

• Freshness vs relevance slider:
  • Makes the freshness–fit trade-off explicit, echoing patterns in a7d1bbba-* and f2a3f039-*; users in volatile categories move toward freshness, in stable ones toward relevance.
  • When movement visibly reorders borderline items and staleness is clearly labeled, users feel more in control and more confident in their chosen setting.
• Sponsored / boost toggle:
  • Similar to the objective vs paid split in 5cee877b-*; a “down‑weight sponsored” control reduces over‑trust in paid‑driven #1s and increases local overrides toward strong objective‑signal items.
  • If toggling barely changes ranks, users either gain confidence that sponsorship is bounded or suspect the control is cosmetic and downgrade system trust.

Merchant incentives

• When user-adjustable levers cap the impact of paid boosts and make objective signals salient, merchants gain stronger incentive to:
  • keep volatile attributes fresh,
  • improve fit signals (spec completeness, reviews), and
  • use sponsorship as a supplement, not a substitute.
• If most users leave defaults untouched or the UI still allows large hidden boosts, incentives revert to paid prominence; the controls become mainly reputational theater.

Net effect

• Properly implemented levers shift behavior toward better-calibrated trust and modestly higher decision confidence, and create some pressure for merchants to invest in genuine fit/freshness. Poorly implemented or cosmetic controls raise cynicism and under‑trust without improving match quality.