Most of Your Enquiries Arrive After Hours. Here's What to Do About It

Look at the timestamps on your last month of enquiries. Use the actual log rather than memory, which means the call list on your mobile and the submission times on your contact form. Sort them by hour of the day and count how many after hours enquiries landed while nobody was at a desk.

For most service businesses I have looked at, the shape surprises the owner. There is a spike in the morning before work, another one from about five in the evening, and a weekend pattern that nobody accounts for because nobody is there to see it. The reason is not mysterious. People search for a plumber or a physio or a mortgage adviser once their own working day is finished, which is exactly when your working day is finished too.

After hours enquiries are not a small edge case you can ignore. They are a slice of your pipeline that arrives when your business is a closed door, and what happens in the next ten minutes decides whether that person becomes a job or becomes someone else's job.

Measure your own pattern before you spend a cent#

I am not going to give you a percentage, because the honest answer is that it depends heavily on what you sell and who buys it. An emergency plumber and a commercial accountant have almost nothing in common on this. What I can give you is the method for finding your own number, which takes about half an hour.

Export three things into a spreadsheet:

• Every inbound call from the last calendar month, with the timestamp and whether it was answered. Most VoIP systems export this directly. If you are running off mobiles, the call log on the handset will do, though you will have to type it in.
• Every website form submission and chat message, with its timestamp. Your form tool or CRM has this.
• Every DM and message on whatever social channels you use for enquiries. Instagram and Facebook show the time on each thread.

Put them all in one column of timestamps. Add a column that extracts the hour, and another that flags whether the enquiry landed inside your published opening hours. Then count.

Two numbers matter. The first is the share of total enquiries arriving outside hours, which is your baseline for after hours enquiries. The second, which almost nobody checks, is what happened to them. Of the after-hours calls, how many did you return? Of the ones you returned, how long did it take? A call you rang back at 9:15 the next morning after it came in at 6:40 the previous evening is nearly fifteen hours old. In a competitive trade, that lead has usually been served by someone else.

If you want the revenue version of this exercise, the arithmetic is laid out in full in what missed calls are really costing you. Multiply your after-hours count by your close rate and your average job value and you have the size of the prize, in your own numbers rather than a vendor's.

Why the evening enquiry is often the better lead#

There is a quality argument on top of the volume argument. Someone who calls you at 7pm on a Tuesday has usually made a decision. They have had the problem all day, they have thought about it, and now they are acting. Compare that to the 11am browser who is collecting quotes and will collect four more before lunch.

Urgency also compresses the buying window. A blocked drain, a broken tooth, a leaking roof during a storm: these people are not comparing your Google reviews for three days. They ring the first number, and if it does not answer they ring the second. Your position in that queue is decided by whether a voice picks up, not by your marketing.

Weekend enquiries for a small business behave the same way. Saturday morning is when people finally deal with the thing they noticed on Wednesday. If your answer to Saturday is a voicemail greeting recorded in 2019, you have effectively opted out of a chunk of the week.

Three ways to cover after hours enquiries#

There are three realistic ways to cover the hours you are not working, and after hours customer service options sit at very different price and quality points.

Voicemail costs nothing and catches the patient. It works acceptably when your customers have no alternative supplier and no urgency, which describes very few service businesses. Its real failure is silent: you never learn about the callers who hung up at the beep, so the option looks better than it is. If you keep voicemail, at least change the greeting to say when you will call back and give a text number as an alternative. Some people who will not leave a message will send a text.

A human answering service puts a real person on the line at any hour. They are good at tone and at handling a distressed caller. The limits are cost, which scales with call volume and after-hours loading, and knowledge. An operator who covers thirty businesses cannot tell your caller whether you service Titirangi, what a standard callout costs, or whether you can be there before lunch. They take a message, which is faster than voicemail but still a message.

An AI agent answers the phone, the web chat and the DMs with the same voice and the same knowledge, at any hour, without a per-call premium for 2am. It can quote your standard pricing bands, confirm whether an address is in your service area, and put a booking straight into your calendar. It does not get tired at the fortieth call of the night, and it does not get emotional with a rude caller.

The honest comparison between the human and the automated option, including the cases where the human is clearly the better buy, is set out in AI vs human receptionist. I would not pretend the AI wins every column.

What an after-hours AI agent should actually do#

Coverage on its own is worthless. An agent set up to handle after hours enquiries has to resolve something, not merely record it. An agent that answers at midnight and says "we will get back to you in business hours" has just built a more expensive voicemail. The value comes from what it resolves before you wake up.

A useful after-hours configuration handles four jobs. It answers the questions people actually ask at night, which for most trades is some version of "do you cover my area" and "roughly what will this cost". It collects the details you would have asked for anyway, including the address, the nature of the job and a callback number. It books directly into a real open slot where the job is standard enough to book. And it separates the genuine emergency from the thing that can wait, then follows your rule about which one justifies a phone call to you at 1am.

That last part is a business decision, not a technical one. Decide in advance what wakes you. For a plumber, a burst pipe with water running probably does. A dripping tap does not, and the agent should say so kindly and book the first morning slot.

Getting the agent to answer accurately depends entirely on what you feed it, which is covered in how to train an AI agent on your business. The overnight shift is where a poorly trained agent does the most damage, because there is no human watching to catch a wrong answer before the customer acts on it.

Rules that keep the overnight shift from embarrassing you#

A few things I would insist on before letting an agent run unsupervised at night.

Be plain that the customer is talking to an AI assistant. Pretending otherwise is the fastest way to lose the trust you just earned by answering, and it is one of the behaviours covered in why customers hate your chatbot.

Give it a hard boundary on price. Ranges and standard callout fees are fine. A firm quote for a job nobody has looked at is a commitment you may not want to honour at 8am.

Make the escalation path real. If someone says the word emergency, or the agent detects it is out of its depth, there must be a path to a human that actually rings. The mechanics of doing that without making the customer repeat themselves are in the handoff.

Read the transcripts. Every morning for the first fortnight, then weekly. You will find questions you did not know people asked, and answers you need to correct. This is the single highest-value half hour in the whole setup.

Where to start this week#

Do the timestamp export first. Half an hour in a spreadsheet tells you whether after hours enquiries are a real gap in your business or a rounding error, and it stops you buying a solution to a problem you do not have.

If the number is meaningful, the next step is deciding how much of the overnight conversation you want resolved rather than merely captured. That decision, more than the software, is what separates 24/7 customer response that earns money from an expensive out-of-office message. The full picture of how these systems work end to end, what they cost and where they break sits in the AI receptionist for small business guide.

And whatever you choose, change the voicemail greeting today. It takes two minutes and it is currently the only thing standing between an evening caller and your competitor.