To handle repetitive customer questions, log the ones you get most often and answer each once in a short, clear help page. Save reusable reply templates, then automate the routine ones with a chatbot trained on your own content. Keep a person ready for anything unusual or sensitive, and review the list regularly.

At a glance

  • Intercom’s guide advises responding immediately even when you do not yet have the solution, and personalising every reply.
  • Intercom says a chatbot grounded in a knowledge base reduces hallucinations because answers come from your own content.
  • Typical repetitive questions include opening hours, delivery times, refund policy, account changes and how to contact support.
  • Chati’s chatbot answers from a business’s own website, shows its sources and hands over to a human team when unsure.

How do you handle repetitive customer questions?

Work through six moves: log the questions, answer each once, build templates, automate the routine ones, escalate the tricky ones to a person, then review it all regularly. The aim is consistent answers and less time lost, not fewer conversations with your customers.

Here are the six steps in order:

  1. Log and rank the questions you get most.
  2. Write one clear FAQ or help page answer for each.
  3. Create reply templates that still sound human.
  4. Let a chatbot trained on your own content answer the routine questions.
  5. Set rules for when a person takes over.
  6. Review the list monthly and keep answers accurate.

A question that arrives again and again is rarely a sign of a difficult customer. More often it tells you that something on your website, in your emails or on your price list is unclear or hard to find. Fixing that once helps everyone who comes after.

This workflow suits small teams who spend part of every day answering the same emails, calls and messages. You do not need special software to start. A shared spreadsheet and an hour of tidy writing will get you most of the way.

Step 1: How do you find which questions are repeated most?

Tally the questions that arrive by email, phone, social messages and chat for two weeks, then rank them by how often they appear. The top ten will usually account for a large share of the interruptions you feel.

Messy heap of orange cards beside four neat stacks of falling height, each held by a blue peg, showing questions sorted by
Sorting a jumble of questions into ranked stacks shows which ones come up most often.

A simple way to log them

Use a spreadsheet or a shared note with three columns: the question in the customer’s own words, the channel it came through and a tick each time it appears. Everyone on the team adds to the same file, so nothing is counted twice or missed.

Where to look

  • Your email inbox, including sent items
  • Call notes or a quick tick-sheet by the phone
  • Web form enquiries
  • Messages on social media and any live chat

Group them by type

Once you have a list, sort it into groups: pricing, availability, process, policies and location. Patterns show up quickly. A dental practice might find that new patient fees come up daily. A letting agent may be asked about viewings and referencing again and again. A trades business could keep fielding questions on which areas it covers for call-outs. Each of those is a perfect candidate for a single, clear answer.

Step 2: What should a good FAQ or help page include?

Each top question gets one short, direct answer, written once and placed where customers actually look. Start with the answer itself, then add detail only if it helps.

  • Lead with the answer. If someone asks “Do you cover Leeds?”, the first words should be “Yes, we cover Leeds and the surrounding area” (or a plain no).
  • Keep the wording plain. Write as you would speak to a customer at the counter.
  • Put key answers on the website. Avoid burying them in PDFs that are hard to search and easy to forget when policies change.
  • Show a date or a review reminder. A line such as “Last checked November 2026” tells readers the information is current and reminds you to check it.

Atlassian has a useful write-up of self-service knowledge base best practices if you want to go further. For most small businesses, though, the simple rules above are enough to begin.

Step 3: How do you write reply templates that still sound human?

Treat a template as a starting point: pre-written, then personalised with the customer’s name and their actual situation. A reply that reads like a form letter will frustrate people more than a slow one.

Blank green card with a folded yellow tab and a red pencil beside it on a blue windowsill, ready for a reply to be drafted
A blank card and a pencil stand in for a reply template: a starting point to be personalised before it is sent.

Respond quickly, even if only to acknowledge

Intercom’s guide suggests responding immediately even when you do not have the solution yet. A short “Thanks, we have your message and will come back to you by end of day” settles a worried customer.

Avoid over-promising

Say only what you can deliver. If you are unsure about a date or a price, say so. Explain things in more than one way too, because a customer who did not follow the first explanation will not be helped by the same words repeated.

End with a clear next step

Finish each reply with one action: book here, send us a photo, call this number. Do not leave the customer wondering what happens next.

Keep templates in one shared place

Store them in a shared document or your email tool so the whole team gives the same answer. Add a short note to each one on when to use it and what to personalise.

Step 4: Can a chatbot answer repetitive customer questions?

Yes, provided it answers from your own verified content rather than guessing. A well-set-up chatbot handles the routine questions at any hour and leaves your team free for conversations that need a person.

You can see the details on our pricing page.

Rule-based bots versus AI trained on your website

Older rule-based bots follow fixed menus. They break as soon as a customer words something unexpectedly. An AI chatbot trained on your website can understand a question phrased in many ways and find the relevant answer in your content. Intercom notes that grounding a bot in a knowledge base reduces hallucinations, because the answers come from your own material. Atlassian’s overview of customer service chatbots covers the wider benefits.

Why showing sources matters

When a bot shows where an answer came from, customers can trust it and you can check it. If a source is out of date, you spot it straight away and fix the page.

How Chati does it

Chati answers from your own website, shows its sources and hands the chat to your human team when it is unsure. It is trained on your data, with custom integration so it fits how your business works. You can find the detail on how it works.

Extras that save more time

Chati also offers branding, lead capture and bookings. That helps with out-of-hours questions in particular. A customer who messages at 9pm asking about a first appointment gets an answer at once, and can leave their details or book, rather than waiting until the morning.

Step 5: When should a person take over from the bot?

Complaints, sensitive matters, complex quotes and anything the bot is unsure about should go to a human. A bot is for routine questions, and knowing its limits is part of making it work.

One hand passes a red paper speech bubble to another hand reaching in, a handover, with blue cards on a line behind.
Passing a conversation from one pair of hands to another is what a good handover from bot to person should feel like.

Set clear handover rules

Write down which topics always go to a person and who picks them up. Make sure the chat history passes across, so the customer does not have to repeat themselves. Chati hands chats to your human team when it is unsure, which gives you a safety net for the questions you did not plan for.

Take care in regulated sectors

Solicitors and accountants should keep the bot to general information about their services, opening hours and how to get in touch. Advice on a person’s particular circumstances belongs with a qualified member of your team.

Tell customers they are talking to a bot

Say so at the start of the chat and make it easy to ask for a person. Customers are generally relaxed about it when you are open, and less so when they feel misled.

Which tool suits which type of question?

Match each type of question to the simplest tool that answers it well. Facts and policies suit an FAQ and a chatbot, while personal or emotional queries need a person.

Question type Best tool Example Review frequency
Facts and policies FAQ and chatbot Opening hours, refund policy Quarterly
Process questions FAQ plus reply template How referencing works for a tenancy Quarterly
Bookings and enquiries Chatbot with lead capture Asking for a first appointment Monthly
Quotes with detail Template, then a person A call-out to an unusual job Monthly
Personal or emotional queries Human A complaint or a worried pet owner As needed

Step 6: How do you keep answers accurate and measure results?

Review your question log monthly, fix any gaps and track a few simple measures. An FAQ that was right in spring can be wrong by autumn if a price or policy has changed.

What to track

  • How many repeat questions arrive each month
  • How long it takes you to reply
  • How many chats were handed to a person

If repeat questions fall and response times shorten, the workflow is doing its job. Read the unanswered and escalated chats too. They show you which new topics deserve an FAQ entry.

A short monthly checklist

  • Check the top ten questions and make sure each answer is still right
  • Read escalated and unanswered chats and add new FAQ topics
  • Check templates for outdated details
  • Update the review date on the pages you checked

Update straight away, not at the next review, after any change to prices, hours or policies.

What should you do if you cannot answer a question?

Be honest, say you will find out, give a timeframe, then add the answer to your FAQ afterwards. Customers forgive “I will check and come back to you by Thursday” far more readily than a wrong answer.

  • Never guess, especially on prices or legal and health matters.
  • Route it to the right colleague and tell the customer who will be in touch.
  • Close the loop by adding the answer to your FAQ or the content your chatbot uses, so the next person gets it straight away.

We’d be happy to help you set up a chatbot that answers from your own website and passes the tricky ones to your team. Get in touch with Chati and we can talk it through.

Frequently asked questions

How do you respond politely to the same question again and again?

Use a template as your base and add the customer’s name and a line about their situation. Thank them for asking, give the answer first and offer a next step. They do not know it is the tenth time today.

How many FAQs should a small business website have?

Begin with the ten or so questions you get most often, based on your log. Add more only when real customers ask them. A short, accurate list beats a long one that nobody keeps up to date.

Will a chatbot replace my customer service team?

No. A chatbot takes the routine questions so your team can spend time on complaints, quotes and personal queries. With Chati, chats go to your human team when the bot is unsure.

How do I stop customers asking questions that are already on my website?

Make the answers easier to find. Put them on the pages where customers look, lead with the answer and avoid PDFs. A chatbot that answers from your website and shows its sources also brings the answer to the customer.

What is the best way to train a chatbot on my business information?

Start with clear, current content on your own website, because the bot answers from it. Chati is trained on your data and offers custom integration. Afterwards, read through the chats to spot gaps in your content.

How often should I update my FAQ page?

Check the top questions monthly and the full page at least quarterly. Update straight away after any change to prices, opening hours or policies, and show a review date on the page.

Is it okay to use a chatbot under UK data protection rules?

Often yes, but you must handle personal information properly. Tell customers they are talking to a bot, collect only the details you need and be clear on how you use them. For advice on your own situation, speak to a qualified adviser.