🚀 Getting Started

A public FAQ page built from your WhatsApp answers

You already wrote the answers. Publishing them costs one setting and deflects the questions before they arrive.

The question-and-answer pairs your automation uses can also be published as a public help page on a URL of your own. Same answers, second audience — the customers who would rather read than ask.

It is close to free, because the work is already done. If your bot can answer a question, that answer is a page.

What it is

A plain public page listing your published questions and answers, grouped into the categories you gave them, with a link into a WhatsApp chat for anyone whose question is not there.

Nothing is authenticated and nothing is personalised. It is a help page, and its value is exactly that it can be linked to, read and shared without anybody starting a conversation.

What is shown, and what never is

ContentPublic?
Q&A pairs you published, in collections you publishedYes
Q&A pairs you wrote but did not publishNo
Collections you did not publishNo
Content crawled from your websiteNo
Anything from a real customer conversationNo

Two deliberate gates — the collection has to be published and the individual answer has to be published — which means nothing reaches the page by accident. That is the right design for something anybody can read.

Why bother, when you have a bot

Four reasons, and the first two are the substantial ones:

  1. Some customers will not message you. A meaningful share of people prefer to read the answer than start a conversation, and today they leave instead.
  2. It is linkable. Your team can send it, you can put it in your delivery notes, and it can sit in your website navigation.
  3. It is a discipline. Writing an answer that has to stand alone publicly tends to produce a better answer than one written for a chat.
  4. It is one place to fix things. Change the answer once and both the bot and the page change.

What to put on it

The questions your team answers most, which you already know. If you are not sure, look at what customers actually asked over the last month rather than at what you assume they ask.

A workable starting set for a store:

Six answers is a genuinely useful page. Sixty is a page nobody reads.

Writing answers that work in both places

The same answer has to read well in a chat bubble and on a page, which is less of a constraint than it sounds — it mostly means keeping answers short and specific.

  1. Answer in the first sentence. No preamble.
  2. Be specific. "3 to 5 working days to major cities" beats "we deliver quickly".
  3. Avoid pointing at the page. "See our website" is useless in a chat and circular on the page.
  4. Keep it current. An outdated public answer is worse than none.

That last point deserves a calendar entry rather than good intentions, particularly around peak season when delivery times change.

Where it fits with the bot

They serve different moments. The page catches somebody researching; the bot catches somebody who has already decided to ask. The page has a chat link for the handover between the two.

The bot side — answering from your own material rather than from a general model — is covered in AI answers from your own website. The same answers feed both.

Set it up in fifteen minutes

  1. Pick your public address — a short lowercase name for your business.
  2. Write six answers, or publish six you already have.
  3. Give them categories so the page groups sensibly.
  4. Publish the collection, then publish the individual answers.
  5. Link it from your website footer, your delivery notes and your team's replies.

Step five is the one that decides whether it was worth doing. An unlinked help page helps nobody.

What it does not do

It is not a website and it is not a substitute for one. It is also not a marketing channel — nothing about publishing a page creates permission to message anybody, and the template rules in Meta's template documentation are unaffected by it.

What it does is remove a class of conversation from your inbox while making the answer easier to find. That is a narrow benefit, honestly stated, and it costs almost nothing to obtain.

Frequently asked

Can customers see my chat history on it?

No. Only answers you explicitly published, from collections you explicitly published.

Does it need a designer?

No. It is a plain page built from your answers.

Should it replace my website's FAQ?

Only if maintaining two is causing them to disagree — which it usually is. One source of truth is the point.

How do I know which questions to add?

From what customers actually ask. The mining of that is covered alongside the bot's own answering in AI answers from your own website.

The number of questions a small store really has

Fewer than you think. Most stores find that 6 to 10 questions cover the large majority of what customers ask, and that the same 3 account for most of the volume.

That is the whole argument for this page. If 3 questions produce most of your inbox, publishing 3 good answers where people can find them is the cheapest reduction in workload available to you.

Finding your real three

  1. Read a month of conversations, not from memory.
  2. Write down the question behind each one, not the wording.
  3. Count.
  4. Write the top three properly, and publish them.
  5. Add the next three a month later, once you know whether the first three helped.

Step two is where the value is. "Where is my order" and "has it shipped yet" and "any update?" are one question, and treating them as three is how FAQ pages end up long and useless.

Keeping it honest

AnswerProblemBetter
"We deliver quickly"Says nothing"3–5 working days to major cities, 5–8 elsewhere"
"Returns accepted"Under what conditions?"Unworn, within 14 days, we cover return delivery"
"Contact us for details"The page exists to avoid thatThe detail
"See our website"CircularThe answer itself

A public answer is a commitment, which is a reason to write it carefully rather than a reason to be vague. Vague answers do not reduce questions — they produce the question again, with less patience.

What it costs to run

Nothing, in messaging terms. The page sends no messages, so none of Meta's pricing applies to it — service conversations remain free and unlimited under Meta's pricing documentation, and the page simply reduces how many of them start.

The only ongoing cost is keeping it current, which is a diary entry rather than a budget line — and worth pairing with the quarterly template review described in keeping your integrations working.

The short version

Publish the answers you have already written. Start with the three questions that actually produce most of your inbox, give them categories, publish the collection and the individual answers, and then link the page from your website footer and your delivery notes.

Nothing reaches the page by accident — a collection has to be published and each answer has to be published — so it is safe to build up gradually rather than all at once.

Frequently asked, part two

Will search engines index it?

It is a public page, so it can be found. Treat it as a help page for your customers rather than as a search strategy — the answers should be written for somebody who already buys from you.

Can I have more than one page?

One address per business. What varies is which collections and which answers you publish to it, which is where the control sits.

What if an answer changes?

Change it once. Both the bot and the page read the same material, which is the main practical reason to keep them together rather than maintaining a separate FAQ on your website.

Is it a replacement for a website?

No. It has no products, no checkout and no branding to speak of. It is a help page, and businesses that already have a website should link to it rather than treat it as one.

How often should I review it?

Quarterly, and again before any peak trading period when your delivery times change. A stale public answer about delivery is the one most likely to cause a complaint.

What is the smallest useful version?

Three answers — delivery, returns, and how to reach a person — published and linked from your website footer. That takes about twenty minutes and it covers most of what a customer wants to check before they buy.

Does it work on a phone?

Yes. Most people who scan a QR or tap a link from a delivery note are on one, so the page is built to be read that way rather than on a desktop.

Need help? Message us