How English Works | The FAQ

Frequently Asked Questions.

The phrase sounds reassuring.

Somebody has already anticipated what you might want to know.

There is a list.

Your answer is probably in there somewhere.

So you begin scrolling.

What is this service?
Who can use it?
When can I apply?
Why should I apply?
How do I apply?
What happens next?
What if I have another question?

Then a strange thing happens.

The format that was supposed to save time becomes a long interrogation.

Every heading begins with what, when, why or how.

The important noun arrives later.

The reader has to process each question before discovering whether it matters.

This is why the FAQ is interesting.

It can be useful.

It can also be a symptom that the information architecture has given up.

Quick Read

One-sentence answer: an FAQ works when real, recurring user questions are the clearest route to information; it fails when writers use invented questions to avoid designing better headings, tasks, pages or navigation.

A good FAQ therefore depends on several tests:

Are these genuinely frequent questions?
Do they represent user needs rather than organisational talking points?
Is question wording actually easier to scan than direct headings?
Does each answer solve one clear problem?
Is the answer authoritative and current?
Would the information belong more naturally inside the relevant task page?
Does the FAQ duplicate information already published elsewhere?
Can the reader reach the answer without reading irrelevant questions first?

The UK Government Digital Service has famously argued against default FAQ pages. Its content-design guidance explains that question headings can be slower to scan because the useful term is often buried behind repeated openings such as “how”, “what” and “when”. GOV.UK’s broader content standards begin instead with user needs: what people are trying to find out or get done, and what format best helps them do it.

That gives us the central principle:

The question format is a tool, not a default container for miscellaneous information.

Why FAQs Feel Natural

Human conversation is full of questions and answers.

“When does class start?”

“At 4 pm.”

“What should I bring?”

“Your marked paper and notebook.”

“Can parents wait inside?”

“Not during lessons.”

The FAQ copies this conversational pattern.

That can be powerful when the user genuinely arrives with a specific question and recognises their own wording in the list.

A parent asking:

What happens if my child misses a lesson?

may appreciate seeing exactly that question.

The format works because the question and the need align.

The problem begins when the writer invents a fake conversational shell around information that would be clearer as direct headings.

A Fake FAQ Reveals the Writer

Consider this question:

Why is our revolutionary integrated service architecture the best solution for modern families?

Did users frequently ask that?

Probably not.

The organisation wanted to say something promotional and dressed the statement as a user question.

This is writer-centred content disguised as reader-centred content.

A real user might ask:

Can I use the service if I already have an account?

Or:

How much does it cost?

Or:

What happens to my data?

GOV.UK’s user-needs guidance insists that needs should be based on what real users are trying to do and expressed in language users would recognise, not invented to justify existing content.

This is one of the best tests for FAQ English:

Would a real person naturally ask this question before reading our answer?

Questions Can Be Slower to Scan Than Headings

Compare:

How do I change my address?
How do I change my password?
How do I change my payment method?
How do I change my appointment?

The first three words are identical.

The discriminating term comes later.

Now compare direct headings:

Change your address
Change your password
Change your payment method
Change your appointment

The key object appears sooner.

GDS calls this frontloading: put the words people are looking for early so they can scan quickly.

This does not make questions bad.

It means question grammar has a cost.

Repeated question openings can hide the semantic difference between items.

When scanning matters, direct noun or verb phrases may outperform full questions.

The FAQ Often Becomes an Information Junk Drawer

Imagine a website team has five pieces of information that do not fit anywhere neatly.

Someone says:

Put them in the FAQ.

The FAQ becomes a storage area for unresolved architecture.

Over time it grows:

fees;
technical support;
refunds;
opening hours;
privacy;
history;
product features;
complaints;
career information;
delivery delays.

The questions are not one coherent domain.

They are simply everything the rest of the site failed to place.

Now the user must search a miscellaneous list.

This is why FAQ growth can be diagnostic.

A huge FAQ may mean the site has many questions.
It may also mean the site has weak information architecture.

Put the Answer Where the Need Happens

Suppose users repeatedly ask:

What file types can I upload?

One solution is to add it to an FAQ.

A better solution may be to place the answer beside the upload field:

Upload PDF, DOCX or JPG files up to 20 MB.

Now the user receives the information at the decision point.

The question may stop being frequently asked because the interface stopped hiding the answer.

This is a profound principle:

Some FAQs are evidence that another part of the system is failing to communicate early enough.

A question about refund eligibility may belong on the refund page.

A question about class times may belong beside the schedule.

A question about required documents may belong inside the application journey.

Good content design moves information toward the moment of need.

Frequently Asked Does Not Mean Important

Suppose 500 users ask:

Where is the login button?

That question is frequent.

The answer should probably not be a permanent FAQ explaining where the login button is.

The button should become easier to find.

Frequency can indicate content need.

It can also indicate design failure.

This changes how we interpret support data.

A repeated question is not automatically a request for another answer page.

It may be a sensor telling us where the system is unclear.

Some Questions Really Do Deserve Question Form

There are cases where a question is the best label.

Especially when the user’s mental state is genuinely question-shaped.

For example:

Am I eligible?

Or:

What if I do not know the exact date?

Or:

Can I change my answer after submitting?

These questions correspond to uncertainty that users may actually carry.

GOV.UK’s Design System itself uses question pages in transactional journeys when a service genuinely needs to ask the user something. Its guidance says teams should know why every question is being asked and only ask for information that is truly needed.

Question form is therefore not the enemy.

Unexamined question form is.

One Question Should Have One Clear Scope

Consider:

How do classes, fees, replacement lessons and school holidays work?

This is one sentence.

It contains four information needs.

The answer will probably become long and difficult to scan.

Separate:

When are classes held?
How are fees charged?
What happens if a student misses a lesson?
Are lessons held during school holidays?

Now each answer has a cleaner boundary.

This resembles good form design and good examination questions.

One clear task reduces interpretation cost.

Answers Should Frontload the Useful Part Too

Question:

Can I reschedule a lesson?

Weak answer:

We understand that from time to time families may experience unforeseen changes in their schedules and that flexibility is important in modern life. Our centre has always sought to provide a balanced arrangement for all students while maintaining class quality…

The reader still does not know.

Better:

Yes, if another suitable class has space. Contact us before the missed lesson where possible.

Then add necessary conditions.

The answer should meet the information need before giving background.

This is answer frontloading.

FAQ Answers Need Version Control

Frequently asked questions often concern changeable information.

Fees.

Schedules.

Eligibility.

Policies.

Deadlines.

If an FAQ answer becomes stale, the familiar format can make old information look reassuringly official.

GOV.UK’s content guidance emphasises that published content should either remain current and maintained or be clearly retained as a historical record.

That means FAQ maintenance is not optional.

Every answer needs ownership.

Who checks it?

When?

What event should trigger revision?

A tidy question list can become a stale knowledge trap if nobody owns the answers.

FAQ Duplication Creates Two Truths

Suppose the Fees page says:

Monthly fee: $320.

The FAQ says:

Fees start from $300.

Which is correct?

The duplication has created competing truths.

This is why GOV.UK content design explicitly considers reducing, splitting or removing content to avoid duplication and conflicting information.

A strong FAQ should not become a second shadow copy of the entire website.

Where possible, one canonical answer should live where the user needs it, and other surfaces should point to it.

Search Logs Can Produce Better Questions Than Meetings

How do organisations decide what is frequently asked?

One weak method is:

What questions do we think users probably have?

A stronger method uses evidence.

site-search terms;
support emails;
call-centre logs;
chat transcripts;
analytics;
user interviews;
failed form fields;
repeated teacher or parent questions.

GOV.UK’s user-needs guidance recommends using evidence such as analytics, call-centre data and previous research rather than relying on assumptions.

This is important because actual questions often differ from institutional assumptions.

The organisation may want to explain its philosophy.

The user may want to know:

Where do I park?

Real user language is humbling.

FAQ Order Is a Ranking System

What question comes first?

The most frequent?

The most important?

The first step in a process?

The question with the highest consequence if misunderstood?

An FAQ list silently ranks attention.

If the first question is:

What makes our company special?

while the refund deadline sits at question 27, the ordering reveals whose priorities shaped the page.

A user-centred ordering should reflect actual tasks and consequences.

Sometimes that means grouping by topic.

Sometimes by journey stage.

Sometimes by frequency.

Again, format follows need.

Accordion FAQs Can Hide Information Twice

Many modern FAQ pages show only questions until the reader expands one.

This can reduce visual clutter.

It also creates another layer of hidden content.

The reader must first find the right question.

Then activate it.

Then read the answer.

If the question wording is weak, the answer is effectively invisible.

This is not automatically bad.

It means the question label carries more responsibility because it becomes the only visible information scent for the hidden answer.

Accessibility, keyboard behaviour and state signalling also matter when content is expandable.

The general lesson is:

Every layer of hiding increases the importance of the words on the door.

Primary School: Turn Confusion into a Good Question

For younger learners, FAQ thinking can teach question quality.

A child says:

I don’t understand this.

Help the child refine:

What does this word mean?
Why did the character leave?
Which sentence tells us the answer?
How do I start this paragraph?

The more precise question narrows the help needed.

Then create a class FAQ from questions students actually ask.

Now the format is evidence-based.

Lower Secondary: Compare FAQ with Direct Structure

Give students the same information in two forms.

FAQ:

What is photosynthesis?
Where does photosynthesis happen?
What does photosynthesis need?
Why is photosynthesis important?

Direct headings:

Photosynthesis: Definition
Where Photosynthesis Happens
Inputs for Photosynthesis
Why Photosynthesis Matters

Ask which is easier to scan and why.

The goal is not to declare one universally superior.

The goal is to see how grammar changes retrieval cost.

Upper Secondary: Audit Whose Questions Are Being Asked

Older students can analyse an FAQ critically.

Ask:

Which questions sound like real users?
Which sound like marketing statements wearing question marks?
What important question is absent?
Which answer belongs elsewhere in the site or document?
Which question is frequent because the underlying interface is confusing?
Which answer is time-sensitive?
Who is responsible for keeping it current?
Does the ordering reflect user priority or institutional priority?

This is advanced discourse analysis applied to an everyday genre.

Ten Failure Modes of FAQ English

Invented question. The organisation writes what it wants to say and pretends users asked.
Slow scanning. Repeated “how/what/when” openings bury the discriminating term.
Junk-drawer architecture. Unrelated information accumulates in one list.
Late answer. Information appears in the FAQ instead of at the moment of need.
Frequency misread. A repeated question that signals design failure is answered forever instead of fixing the cause.
Multi-question item. One heading hides several separate needs.
Buried answer. Background comes before the useful response.
Duplicate truth. The FAQ repeats information that later diverges from the canonical page.
Stale answer. Time-sensitive information remains authoritative-looking after reality changes.
Hidden accordion. Weak question labels make the concealed answer difficult to discover.

How to Build a Better FAQ

First, ask whether you need an FAQ at all.

Collect evidence of real user questions.

support logs;
search terms;
emails;
interviews;
analytics;
classroom questions;
failed tasks.

Then examine each question.

Could the answer be placed earlier in the relevant journey?

Could a direct heading be easier to scan?

Does the answer duplicate another page?

Is the question actually one question?

Does the answer begin with what the reader needs?

Who owns updates?

Then group the remaining questions by user need, not organisational department.

Use the user’s vocabulary where possible.

Keep answers bounded.

Link to deeper material rather than rebuilding entire pages inside each answer.

And perform the strongest test:

If this FAQ disappeared tomorrow, where should each answer naturally live?

If the answer is obvious for most items, the FAQ may not be the best architecture.

The Deeper Idea: Questions Are Routes Through Ignorance

A question has a special structure.

It contains something known and something missing.

Known frame + unknown slot.

“When does class start?”

We know there is a class.

We do not know the time.

“Can I reschedule?”

We know there is an appointment.

We do not know whether the system permits a change.

A good FAQ works because it predicts those missing slots accurately.

A bad FAQ fails because it predicts the organisation’s preferred questions instead of the user’s real unknowns.

That is why the deepest work happens before the writing.

We must discover what people genuinely do not know.

Then decide where the answer belongs.

The question mark comes last.

Reader Checklist

Is this a real question users ask?
What user need sits beneath it?
Would a direct heading be easier to scan?
Does the question contain one clear task?
Does the answer begin with the useful part?
Should this information appear earlier in the user journey?
Does the answer duplicate a canonical page?
Is the information time-sensitive and owned?
Does the ordering reflect user priorities?
Would removing the FAQ reveal a better natural home for the answer?

Research and Further Reading

Final idea: the best FAQ is not the page with the most answers. It is the system that understands why the question arose and puts the answer where the reader actually needs it.