Alternative entrances need to be as prominent as the main one, not hidden in a corner

What it is

The alternative channel exists, but its entrance cannot be found: the staffed counter tucked round a corner on the third floor, the hotline number in the page footer five screens down, the offline route whose instructions begin with "download the app to view the guide". For people forced onto the alternative path this is a second exclusion — the first denied them the digital door, the second hides the non-digital one.

"Prominent" here has two layers: spatial prominence (on-site signage, wayfinding, sightlines) and informational prominence (how deep the offline path sits in the online page). Both must meet "find it in passing", not "find it with patience".

Why it happens

Why do entrances get hidden? Two forces push the same way.

One is the migration incentive. An institution driving online adoption has no motive to make the alternative conspicuous — a visible entrance slows migration. Behavioural science calls designs that deter a choice by adding friction sludge: you are not forbidden the staffed counter; the cost of finding it is raised until you give up voluntarily. It shares DNA with dark patterns, except the beneficiary is the institution's KPI rather than the user.

Two is the mismatch of search costs. Which level the entrance hides at is zero-cost internal knowledge to the designer and a fully-paid search burden to the user — and the very populations forced onto alternative channels (older, low-income, disabled) have the least search skill, confidence and slack. The people least able to afford search costs are assigned all of them.

The combined result is an entrance that exists objectively and is unreachable subjectively: in the annual report it is a service; at the point of use it is a riddle.

Studying it

• Friction and choice-architecture research: the experimental paradigm around sludge — abandonment rates of the same choice under different friction levels — supplies magnitude references for "bury the entrance one level deeper, lose N percent of seekers".
• Service-journey walkthroughs: observing real halls and websites, recording where people lose their way, at which level they abandon the search, and whom they ask.
• Discoverability testing: give testers the task "complete this without the app" and measure steps, time and help requests to locate the alternative entrance.

One methodological caution: walkthrough and test samples must include the populations actually forced onto alternative channels; running usability tests with young research assistants systematically underestimates the difficulty and yields the false negative "actually quite easy to find".

Where it stops holding

• Prominent is not equal-weighted. Leading with the digital entrance on the first screen, with the alternative one level down, is a legitimate ordering — most users, most of the time, go digital. The line in question lies between "hard to find" and "at hand": reachable within two steps is the floor; a fifth-screen footer is not an ordering but a concealment.
• Prominence requirements scale with the necessity of the service. Emergency, social-security and basic-financial entrances must be unconditionally prominent; entertainment services may rank theirs lower. Forcing every entrance onto the first screen dilutes the critical ones.
• Online prominence ≠ offline prominence. A contact number on the webpage that no one answers merely relocates unreachability from the information layer to the physical one — both layers must pass acceptance together.

Applying it

• Present both paths side by side on the service pages of the site and app: "do it online" and "do it in person / by phone" at the same level, with the offline path unfolded, no interstitial confirmation, no login required to read it.
• On site, follow two-step reachability: from the main entrance, within two steps either arrive at the alternative counter or meet explicit signage (a physical sign that does not depend on scanning a QR code).
• Hold the signage itself to accessibility standards: large type, high contrast, no colour-only distinctions — and never "scan to reach a human", which stuffs the digital precondition back into the alternative channel.
• How to check: anonymous walkthroughs — send testers unfamiliar with the institution to complete one task by non-digital means only, timing the whole way and logging sticking points; write "steps and minutes to find the entrance" into acceptance criteria, and remediate on the spot whatever exceeds them.

Related

• Same group: Z8.05.1 Services that presuppose a smartphone exclude non-users by design · Z8.05.2 Alternative channels need functional parity, not token retention · Z8.05.3 Exclusion concentrates among older, low-income and disabled people
• Nearby: Z8.06 Accessibility requirements for public facilities · Z8.02 Touchless services and public terminals
• Search terms: sludge · friction design · choice architecture · service discoverability