Frontline Go-Along Interview¶

Method — instantiates Revealed-Use Path Alignment

Walk the real route beside the person who uses it and have them narrate each deviation as they make it, recovering the reason, the who, and the who's-missing that logs cannot show.

Frontline Go-Along Interview is a contextual, in-situ interview: you accompany a real user along their genuine workaround and have them narrate, in the moment and at the point of friction, why they leave the official path. Its distinctive move is turning a trace into an explained, testable hypothesis — and, at the same time, surfacing whose trace it is and who never appears in it. A clickstream can tell you a third of users search instead of navigate; only the go-along can tell you they search because the label means nothing to them, and that the newest staff can't do the workaround at all. It is the archetype's one instrument for cause and for absence.

Example¶

Dispatch software assigns delivery drivers a "recommended" stop sequence, and the ops team notices drivers constantly overriding it. A go-along rides with three of them across different shifts, asking each to talk through every time they ignore the app. One explains the app routes him down a street with no legal place to stop, so he loops the block; another front-loads a fragile-parcel building because the app's order would leave it baking in the van; a third — a relief driver new to the area — actually follows the app and gets lost, which reveals that the veterans' tidy workaround silently depends on local knowledge the app assumes everyone has. The output is three interpreted hypotheses (illegal stops, thermal risk, tacit route knowledge) plus a visibility finding: the official sequence quietly disadvantages new and relief drivers, who are invisible in the veterans' clean deviation trace.

How it works¶

Start from an observed deviation, not a blank slate. You go along a route the traces already flagged and ask about that departure — which is what keeps the method inside this archetype and out of open-ended needs discovery.
Narrate in the moment. Think-aloud at the point of friction surfaces tacit reasons a desk interview would never reach.
Probe for the cause type. Missing affordance, time pressure, coercion, inaccessibility, ignorance, safety — the interpretation rule sorts the deviation into a testable claim rather than a verdict about the user.
Ask who isn't here. Deliberately seek the users for whom the workaround fails, or who never deviate, so the redesign isn't built only for the fluent few.

Tuning parameters¶

Sampling for absence — interview only fluent deviators, or deliberately recruit new, struggling, and non-deviating users. Sampling for absence is what turns the method from anecdote into an equity check.
Narration timing — think-aloud in the moment versus retrospective recall. In-the-moment captures tacit friction; retrospective is easier to schedule but loses the detail that mattered.
Interviewer framing — neutral curiosity versus authority presence. A manager on the ride-along suppresses the very rule-breaking you came to understand.
Depth versus coverage — a few deep go-alongs recover rare causes; many shallow ones test how widespread each cause is.

When it helps, and when it misleads¶

Its strength is unique in this set: it is the only mechanism that recovers cause and context, and the only one positioned to find whom the trace omits — it turns "people deviate" into "people deviate because X, and here is who can't." Its failure mode is the Hawthorne effect: being accompanied and asked to narrate can change the very behaviour under study, and users readily rationalise a habit after the fact.^[1] Small samples invite over-generalising one vivid story, and the classic misuse is cherry-picking the single quote that justifies a decision already made. The discipline is to pair the recovered "why" with scale evidence — a heatmap or a clickstream scan — so a compelling account is never mistaken for a common one, and to always ask who is not on the ride.

How it implements the components¶

trace_interpretation_rule — its core: it ties each deviation to its context and converts it into a testable hypothesis about cause rather than a reified fact about the user.
equity_and_visibility_check — narrating with real, varied users reveals whose behaviour the trace represents and whom it silences.
counter_trace_search — it actively seeks the users who don't deviate, who struggle, or who are absent from the data, so an absence isn't mistaken for consent.

It does NOT establish the official baseline (that's Desire Path Walkthrough) or measure the friction and scale of a deviation (Friction Mapping Session and Clickstream Deviation Scan); it explains a trace, it does not size one.

Instantiates: Revealed-Use Path Alignment — it supplies the archetype's interpretation and equity readings, the context that keeps the redesign from acting on a trace it doesn't understand.
Sibling mechanisms: Desire Path Walkthrough · Friction Mapping Session · Clickstream Deviation Scan · Use-Trace Heatmap · Workaround Inventory · Before/After Trace Monitoring · Trace Decay Review · Safety and Accessibility Review · Route Closure with Alternative · Temporary Paving Pilot · Informal Route Legalization Patch

Notes¶

A go-along can slide into open-ended user research, which is a different archetype (the broader User Context Validation). What keeps it here is the tether: it starts from a specific, already-observed off-path trace and stays on it. Validating general fit or discovering fresh needs is not this mechanism's job.

References¶

[1] The Hawthorne effect — people alter their behaviour when they know they are being observed. On a narrated ride-along this cuts two ways: the deviation may be softened for the observer, and the reason offered may be a tidy rationalisation, so the account is corroborated against unobtrusive traces. ↩