After-Hours Voicebot: Handling Night and Weekend Questions
An after-hours voicebot answers the questions people ask when your office is closed: opening times, prices, how to book, what happens next. Its real job at night is narrower than it looks. It should solve what it can from your own information, promise only what your team can deliver the next morning, and leave a clear trail for whoever picks up the follow-up. This guide covers which questions arrive after hours, what the bot may and may not promise, and how to organise the next-day follow-up so nobody is left waiting.
What questions reach an after-hours voicebot?
Most evening and weekend questions are simple and repetitive: when are you open, how much does it cost, do you have it in stock, how do I get there, can I change my booking. People ask them late because that is when they finally have time, not because the matter is urgent.
A smaller group needs a person: a complaint, a refund dispute, a question about a specific order or contract, anything that depends on data the bot does not have. A third group is genuinely urgent, such as a failed delivery on the day of an event or a safety problem. Sorting questions into these three groups before launch decides almost everything else in the setup.
- Answer now: hours, prices from the public price list, directions, product details, how booking works.
- Answer and hand over: anything about one customer’s account, order or case. The bot explains what will happen and when.
- Escalate: urgent cases. The bot gives the emergency contact you really staff, or says plainly that there is none outside working hours.
Phone line or website: where does the voicebot work?
An after-hours voicebot can sit on a phone line or on your website, and the two are different projects. Phone voicebots replace or extend an answering service and need telephony, call routing and often an IVR menu.
A website voice assistant works in the browser: the visitor presses a button, speaks, and hears the answer. It helps people who are already on your site at 11 pm and would rather talk than type, for example on a phone while cooking or driving as a passenger. If your evening traffic is mostly website visits, start there. The article on when visitors prefer speaking explains how voice and text chat split the work.
To be clear about our own product: Botino adds a chatbot and a voice assistant to a website with one line of code, and the voice part works in the browser. It does not run phone lines or call centres. The rest of this guide applies to both channels.
What should an after-hours voicebot promise?
An after-hours voicebot should promise only what the morning team can keep. A promise made at midnight is a promise made by your company, and a bot that says “someone will call you first thing” creates a commitment your staff may not know exists.
Write the promises down before launch, as plain sentences the bot can use:
- “Our office opens at 8 am on weekdays. Messages sent through the contact form are answered on the next working day.”
- “I cannot see your order details. Please use the contact form and add your order number, and the team will reply on the next working day.”
- “For urgent problems with heating outside office hours, call the emergency line listed on our contact page.”
Avoid soft promises the bot cannot back up, such as “we will sort it out”, “you will get a discount” or exact callback times. If your team works in shifts or closes for holidays, the promise has to change with the calendar. The guide to updating holiday opening hours shows how to keep those dates from going stale.
Say “I do not know” instead of guessing
At night there is no colleague to correct a wrong answer, so the cost of a guess is higher. A bot that answers only from your own documents and says when it has no answer is safer than one that fills gaps. Configure the scope so off-topic or unknown questions get a short, honest reply and a pointer to the next step. The post on when a bot should admit uncertainty covers wording that does not frustrate callers.
How to prepare the knowledge base for nights and weekends
Prepare the knowledge base by writing down the answers your staff give most often during the day, then checking which of them still make sense when nobody is in the office. Many daytime answers end with “I will check with a colleague”, which does not work at 2 am.
- List the repeat questions. Ask the people who answer the phone or inbox which questions they hear every week. Read recent emails received outside working hours.
- Write self-contained answers. Each answer should work without a follow-up from a person: the price, the condition, the link or the step.
- Add an “outside hours” sentence wherever a case needs staff, saying when the team will look at it.
- Check spoken length. Voice answers should be short. Long lists of conditions belong on a page the bot can point to.
- Set the languages. Night visitors may come from other countries. Decide which languages the bot uses and what it says in others.
Next-day follow-up: who picks up what
The next-day follow-up works when every handed-over case reaches a named person through a channel they already check. The bot’s part is to collect the minimum needed and to tell the customer exactly what to expect; the team’s part is to respond within the promised time.
A simple routine for a small team looks like this:
- The bot directs anything account-specific to the contact form or a dedicated email address, so it lands in the existing inbox, not in a separate tool nobody opens.
- One person starts the day by clearing messages that arrived overnight, before new calls begin.
- Recurring night questions that the bot could not answer are added to the knowledge base the same week.
- Anything the bot promised that staff disagree with is corrected in the bot’s sources, not explained away to customers.
When a conversation moves from bot to person, the customer should not have to repeat themselves. If your setup lets staff read the transcript, use it. If not, ask the bot to suggest that the customer pastes key details into the form. More on this in handing over without losing context.
A worked example with assumptions
Take a hypothetical bicycle repair shop open on weekdays from 9 to 5. Assume most evening questions are about service prices, whether a part is in stock and how long a repair takes. The bot answers prices and typical repair times from the published price list and says that stock is confirmed by phone during opening hours.
A visitor on Saturday evening asks to move a Monday appointment. The bot cannot see the calendar, so it explains that bookings are changed through the contact form or by phone from Monday at 9, and that the shop replies on the same day. Nothing is promised that the owner cannot deliver, and on Monday morning the request is already waiting in the inbox.
FAQ
Can an after-hours voicebot take bookings?
It can if it is connected to a booking system or calendar. Without such a connection it should explain how to book and link to the booking page, rather than pretend to reserve a slot.
Should the bot say that the office is closed?
Yes, in a short sentence with the next opening time. People accept a closed office much better than a bot that hides it and then leaves them without an answer.
Do I need a separate bot for nights?
Usually not. The same bot works around the clock; what changes is a handful of answers that depend on staff being available. Keep those answers marked so you can review them when hours change.
Before you switch on an after-hours voicebot, write the three lists: what it answers, what it hands over, and what it escalates, with the exact promise for each. You can try how a voice assistant answers on a website on the live demo bots page, then test your own night questions against your list.
Related posts
Chatbot for SaaS Support: Plan, Billing and How-To Questions
A chatbot for SaaS reads your help center, pricing page and docs. Then it answers the how-to, plan and billing questions customers…
Custom GPT Retirement: Moving a Customer GPT to Your Website
Under the custom GPT retirement, custom GPTs are scheduled to stop running on December 11, 2026. If you shared a GPT with…
Chatbot GDPR Checklist: What to Settle Before Launch
Before you launch a website chatbot, settle a short list of questions: what personal data the chat collects, on what lawful basis,…