The reachability of help paths determines whether harm is discovered
Aliases: help-seeking barriers · remedy path reachability · complaint accessibility
What it is
Remedy accessibility is the effective difficulty of the path between "I've been harmed" and "I get help." Whether harm becomes visible to the product side depends not on whether the harm occurred but on whether that path exists, can be found, and can be completed. For cognitively and economically vulnerable people, every added step of cost on the path — can't find the entry, can't parse the terms, can't afford the phone call, dare not complain — pushes the harm back into invisibility. A protection system's effective ceiling is set by the weakest user's ability to walk this path.
Why it happens
The help path is a funnel, and vulnerability amplifies attrition at every layer. Entry: help entries buried in nested menus or phrased as procedural terminology ("dispute resolution" rather than "I was charged wrongly") shed users under high cognitive load. Language: paths available only in the official language, via text hotlines, filter out entire linguistic minorities. Cost: processes requiring phone calls during working hours or printed postal materials filter out exactly the most time- and resource-constrained. Trust: fear of retaliation (account closure, service suspension) or shame ("admitting I was deceived") keeps many silent. After four filters, the arriving complaint sample skews toward the most articulate, and the product side structurally underestimates harm — quiet dissatisfaction is invisible.
Where it stops holding
Accessibility governs discovery, not handling: an open entrance with perfunctory processing fails just as badly, and the value of the former depends on the latter. The path also cannot rescue harms that leave no behavioral trace (users who simply leave), so it is a necessary, not sufficient, component of discovery — complementing proactive monitoring (abnormal churn, stratified billing anomalies). Overly generous entry has its own cost — abusive complaints consume handling capacity — but that is a capacity problem and must never become a reason to narrow the entrance.
Applying it
- Place help entries inside the highest-harm flows (billing, suspensions, transaction disputes), named in outcome language ("charged too much," "item never arrived"), not procedure language.
- Provide at least two channels (self-service form plus human contact) with asynchronous submission outside working hours; accept photos of documents instead of post only.
- State a no-retaliation commitment and a handling deadline in the complaint flow, with automatic escalation on timeout; keep the service active for complainants until adjudication.
- Verify: stratify by population the conversion from "harm signal → actual complaint" (for example, anomalous-charge events versus complaint volume); a significantly low ratio for a population means its path is filtered at some layer — walk entry, language, cost, and trust one by one.
Related
Cards in the same group
- P4.04.1Cognitively and economically vulnerable people are easier to manipulate
- P4.04.2Targeted delivery amplifies harm
- P4.04.3Protection must not take the form of exclusion
- P4.04.4Vulnerability is situational — anyone can enter it briefly
- P4.04.5Judging vulnerability itself requires extra sensitive data
- P4.04.6Restraint applied to everyone beats differential treatment after identification