Research · Safe to Say, Safe to Reach, Safe to Use

The 4:47pm Test

Does expertise really work when life gets messy?

Part of Safe to Say, Safe to Reach, Safe to Use · Draft v0.1 · 1 October 2026

On this page: The idea · What it covers · Checklist · The professional path · The whole system · Putting it into practice

The idea

4:47pm on a regular Tuesday

This idea sits at the heart of Safe to Say, Safe to Reach, Safe to Use.

What are we actually testing? Not whether regulated expertise helps when everything’s neat: someone’s calm, has time and confidence, strong Wi-Fi, all their paperwork, and a professional sitting right beside them.

We care about whether it helps at 4:47pm on a regular Tuesday.

“4:47pm” isn’t a literal time. It names a state: you’re stretched thin, you’re missing crucial info, there are worries piling up, and the clock’s running out. Tuesday’s just an ordinary day, nothing fancy, nothing catastrophic. It’s real life.

At 4:47pm, you, or the person who needs help, might be wiped out, scared, angry, grieving, or just overwhelmed. Maybe you don’t have all the facts. Maybe there’s a tight deadline, and the service shuts at five. Maybe your phone’s dying. You might have to pick up your child, might not have privacy, or English might not be your first language. You might not even know if the source you’re talking to is trustworthy.

And the professional you call? They’re under the gun too. Incomplete info, limited authority, someone else already waiting. No promises about the next step. No time to piece things together if the process is broken.

This isn’t some rare stress test. It’s the everyday reality for lots of people who need to use law, research, policy, or any kind of expertise.

The 4:47pm Test asks a simple question: does this expertise hold up when people don’t have all the time, knowledge, privacy, confidence, documents, or emotional distance the system quietly assumes?

If your thing only works for the perfect user, the gap between “on paper” and “in real life” is wide open.

What it covers

What does the 4:47pm Test cover?

The Test checks three things:

Safe to Say

Does the information hold the truth, its boundaries and its limits, even under pressure?

Safe to Reach

Can someone find it, get to it safely, and see why it matters before their time or focus runs out?

Safe to Use

Can they follow it? Spot what they don’t know? Tell what they can rely on? Take the next step, get help, push back if there’s a mistake, or bounce back if things go sideways?

The test isn’t about every possible worst case hitting all at once. Instead, teams try combinations of likely constraints that real users run into.

A few examples:

Phone’s nearly dead, it’s almost closing time, deadline tomorrow. Can the person find help fast, see the urgency, use their phone, and know a fallback plan?
Needs an interpreter, missing documents, system’s unfamiliar. Is help accessible? Are the real questions clear? What does support actually do?
Monitored communications, unsafe mail, and family violence. Can the person reach help safely? Does the pathway keep privacy?
Regional Victoria, patchy internet, no car. Does the system offer realistic other routes? Can the person act, recover, and get help?
Frontline worker has gaps in info, too many people waiting. Do they have authority and a clear script? Can they safely hand things over, or are they expected to fix a broken design?

The scenarios should come from lived experience or actual evidence, not just generic personas.

Your setting

How does the test fit your setting?

The test bends to match where it’s being used. Imagine:

Court: someone gets a notice right before a big filing deadline.
Hospital: discharge instructions at night, when the main clinic’s closed.
Government service: an online application, but the person’s got a wobbly phone connection.
Financial services: hardship info for someone who can’t call during business hours.
AI tools: a user faces an unusual or risky situation and needs a human to step in.
Regulator: a safety warning sent to someone who can’t read English or get messages safely at home.

The goal isn’t to magically solve every problem instantly. It’s about shining a light on the traps before people who can’t afford mistakes fall into them.

A pathway holds up better when someone can figure out:

Is this urgent?
What can I actually trust here?
What do I need: info, papers, access?
Which way can I reach someone right now?
What if that way’s blocked?
Any backup or escalation routes?
What do I do if nobody gets back to me, or something turns out wrong?

Checklist

Checklist: questions to test with

Can the person find it?

Do they spot the right resource, even if they don’t know official terms?
Is it in the logical place?
Is it accessible by phone?
Is it available when needed, or locked behind a narrow window?
Can they use it safely, in their language, with privacy?
Are interpreters, accessible formats, or alternatives on hand?
If the obvious way doesn’t work, can they find a trustworthy backup?

Can they see why it matters?

Do the first words or screens tell them what’s up?
Is it clear if it applies to them?
Are rights, risks and deadlines obvious and real?
Is the urgency shown, but not exaggerated?
Does the resource give a reason to keep reading?
Does it avoid fake fear or pressure?
Is there a clear next step?

Can they understand what’s happening?

Could they explain it in regular words?
Can they tell who’s in charge, who’ll explain, and who decides?
Do they see what facts or documents matter?
Can they separate knowns from unknowns?
Is it clear what this resource covers, and what it can’t decide?

Can they tell what they can rely on?

Can they tell a formal decision apart from general guidance?
Can they tell whether it has assessed their situation, or only explained the general rules?
Is it clear what’s established, what’s uncertain, and what’s outside scope?
Do they know when an expert or formal decision-maker has to step in?
Is the resource honest about its own authority?

Can they act realistically?

Is the main point, deadline and next step easy to find?
Do they know what to do, when, how, and with what?
Can they pull it off with their actual time, money, privacy and access?
Can they decide when and how they’ll do it?
Are common obstacles called out?
Where the resource is allowed to, does it show what really needs escalating and what can wait?

Can they reach help?

Is a real, named service available, not just “seek advice”?
Is there a number, website, counter, or other contact route?
Are opening hours, costs, eligibility and accessibility clear?
Can they actually reach the service before closing?
Is the help real? Can the service do what it says?
If help isn’t there, is that limitation clear and is a backup suggested?

Can they question and recover?

Is there a way to correct errors, ask for review, complain, or escalate?
Do they know what to do if they can’t get through?
Is there a Plan B if the first try fails?
Can they pick up again if a referral fails or a step is missed?
Does the resource explain its limits and who takes over when needed?

Testing the professional path

The 4:47pm Test isn’t just for users. It asks:

Can an ordinary professional help, even under tough conditions?

Frontline workers often have incomplete info, limited authority, maybe three other people waiting, and not enough time to untangle a process.

The system shouldn’t demand they:

Guess what the expert meant.
Quietly give advice beyond their authority.
Make up a referral route.
Promise impossible outcomes.
Explain deadlines they can’t see.
Patch up handover problems.
Become the glue that holds a wonky system together.

If your process only “works” because exceptional staff fill the cracks, the process itself is broken.

Testing the whole system

Sometimes, the 4:47pm Test shows flaws that better wording won’t fix, like:

Phones with two-hour wait times.
Services closed before people get a chance to call.
Online forms that crash on mobile.
No safe way to get messages.
Services that have no capacity anyway.
Forms demanding documents nobody has.
No interpreter pathway.
No way to record safe contact info.
Pass-the-parcel between services, repeating your story.
“Review” routes that exist on paper, but are impossible to find or use.
No real options: rules make help structurally unavailable.

The problem isn’t with users. The findings are about the system, and fixing them really means changing policies, staffing, agreements, technology, handover arrangements, or even rules, not just fixing what’s written.

Putting the test into practice

You should use the 4:47pm Test in three places:

1
Before launch: while designing.
2
During testing: with real users and frontline staff.
3
After launch: by monitoring what happens.

Before launch

Ask your team:

What scenario are we testing?
What’s likely to go wrong for these users?
Is the main info, deadline and first step clear, with any crucial fine print right beside it?
Does it assume too much knowledge, privacy, money, or confidence?
Can the person reach a real human before closing?
Is there a backup?
Does the person on the line have the real tools and authority to deliver?

During testing

Don’t just check “Was that clear?” Give users the resource and ask:

What happened for you?
Does this apply to you (or someone like you)?
What would you do now? By when?
Who calls the shots here?
What can you rely on?
What if the first contact point fails?
What’s the missing question here?
What stops you from moving forward?
What would help get you unstuck?

After launch

Track:

Missed deadlines.
Repeat questions.
Half-finished forms.
Referrals gone nowhere.
Unsafe assumptions.
Long wait times.
Missed contacts.
Escalations.
Workarounds from the frontline.
Whether people are using backups.
The stuff they ask only after stumbling or failing.
Whether some groups are consistently falling through the cracks.

What the test gives you

What does the test give you?

Not a stamp that says “Works!” or “Doesn’t work.” It gives you evidence, reveals gaps, clarifies who’s responsible, and triggers a fix.

If meaning or authority is muddled: revise, clarify, and get it checked by the right person.
If people can’t reach or access the info: change how, when, or in what format it’s sent.
If the next step fails: repair it, check ownership, and retest.
If there’s a structural barrier: document it, name the owner, and escalate it for policy, technology or service change.
If there’s a big risk: make a call: restrict, warn, redesign, or sometimes don’t release.
If the person carrying it is being asked to go beyond their authority: redesign the boundary, script and handover so it doesn’t depend on heroics.
If help isn’t available: flag it as unfinished, be upfront, and don’t pretend it’s fine.

It passes the test only when you can show what you tested, what failed, what you changed, who owns any remaining blocks, and when you’ll try again.

The final question

Could a real person use this when it mattered, with its meaning and limits intact, understand what they can trust and do, and move forward without tripping on unseen cracks?

If not, the 4:47pm Test tells you exactly what to fix: in the information, the delivery, or the system behind it.

The framework Handovers

Chami Rupasinghe · Admitted Australian lawyer (non-practising) · Founder, Lejacie · info@lejacie.com

The 4:47pm Test is part of Safe to Say, Safe to Reach, Safe to Use, a framework still in development (Draft v0.1). It’s a design and review tool, not a certification, a compliance standard or legal advice. Passing the test doesn’t mean your resource is lawful, accessible or safe in every scenario. It means you’ve checked whether it works in real conditions, and recorded what still doesn’t.