How AI cuts customer waiting times from hours to seconds

An engaged tone on Monday morning, a queue with hold music, an email that only gets answered two days later: if you want to cut customer waiting times, you first need to understand where they arise. This article shows where AI really does bring waiting times down to seconds, where it only speeds up the first response, and works through what this can mean day to day for a doctor's practice.

Where customers wait today

Waiting time almost always arises at the same point: more enquiries come in at once than the people available can handle. The typical situations are quickly listed.

• Doctor's practice: from eight o'clock the lines are jammed, while patients are already queuing at the reception desk.
• Car repair shop: the master mechanic is in the workshop, and one person in the office handles the phone, invoices and parts orders all at once.
• Property management: after a storm, dozens of tenants report damage at the same time, and each wants to know when someone will come.
• Online shop: questions by email land in an inbox that gets worked through once a day.

Then there are the times when nobody is reachable at all: evenings, weekends, the lunch break. For the customer, that is waiting time too, just without the hold music.

What long waiting times cost

Every business owner knows the obvious consequences. Some callers give up and try a competitor. Those who finally get through are already annoyed, and the conversation takes longer. A third effect is less obvious: people who don't get through call again, sometimes several times. That makes the line even busier without any more requests being dealt with. So long waiting times generate extra call volume themselves.

The team suffers too. Anyone who spends the whole morning on the phone feeling that ten more people are waiting outside works under pressure and makes more mistakes. The article Improving phone availability with AI describes how strongly phone availability ultimately affects revenue.

How AI cuts customer waiting times

AI has a structural advantage over a human receptionist: it can hold many conversations at once. An AI phone assistant answers the tenth call just as quickly as the first, and a chat or WhatsApp assistant replies to a hundred messages in parallel. It is worth keeping two things apart here.

Response time is the time to the first reply. With AI it really does drop to seconds, around the clock. Resolution time is the time until the request is dealt with. It only drops to seconds where the AI is allowed to act itself, for example booking or moving an appointment or giving information from the system. Everything else is taken in faster and handed over cleanly, but it is still dealt with by the team.

Request	Without AI	With AI
Booking or moving an appointment	Hold queue or call-back	handled immediately if the calendar is connected
Opening hours, directions, procedure	Phone or searching the website	answered immediately
Status enquiry (order, repair, delivery)	Checking with the team, call-back	immediately, if the AI can access the system
Damage report, complaint	Note on a scrap of paper, call-back later	taken in immediately, handed over to the team in full
Individual advice	Appointment with a specialist	Appointment with a specialist, arranged faster

Unlike a classic phone menu with “Press 2”, an AI voice assistant understands freely worded sentences. The article AI voice assistants beyond simple IVR menus describes the difference in more detail.

Worked example: Monday morning at a GP practice

The following figures are assumptions for illustration. Suppose a GP practice receives 150 calls on Mondays between eight and ten o'clock. Two medical assistants answer the phone alongside their other work, and each call takes four minutes on average. On paper, the two of them can therefore handle about 60 calls in those two hours. The remaining 90 callers hear the engaged tone, wait in the queue or try again later.

Let us further assume that 80 of the 150 calls are purely about appointments: a new appointment, a change, a cancellation. An AI phone assistant that books directly in the practice calendar takes these over. That leaves the team with 70 calls about matters that need medical staff. Of these, the two of them again handle about 60 in the two hours. Instead of 90 callers, only about ten are left waiting, and they too get an immediate response and are noted down for a call-back.

Whether the numbers work out like this for you depends on the share of appointment requests. If it is small, because most calls are about test results or acute complaints, the AI takes far less pressure off. The article on AI in doctors' practices covers how practices also organise reminders and follow-up care. A similar Monday morning at a small-animal practice is described in AI for veterinary practices.

Implementation in five steps

1. Measure the peaks. Ask your phone provider for call statistics, or keep a tally for two weeks: when how many calls come in and how many go unanswered.
2. Sort the requests. For a few days, note what the calls are about. This shows what the AI can take over.
3. Set limits. In practices and law firms, the AI only handles organisational matters. It gives no medical or legal information, and in emergencies it immediately refers callers to 112 or to the out-of-hours medical service on 116 117.
4. Clarify data protection. If you are bound by professional confidentiality, you need a vendor contractually obliged to maintain confidentiality, a data processing agreement and, ideally, servers in the EU.
5. Communicate openly. The assistant introduces itself as an AI, as the EU AI Act requires. A short note on the website and in the greeting helps so that regular customers aren't surprised.

Frequently asked questions

Does AI really cut every waiting time to seconds?

The first response, yes; the resolution, not always. Appointments, information and status enquiries are dealt with immediately if the systems are connected. Requests that need a specialist are taken in faster but are still handled by people.

Do customers get annoyed when an AI answers?

Some do, especially older regular customers. Most, however, are more annoyed by a hold queue. What matters is that the AI speaks clearly, actually resolves the request and passes it to a person on request.

What about calls outside opening hours?

This is often where the effect is greatest, because until now nobody was reachable at all. Appointment requests made in the evening are already in the diary the next morning. More on this in the article on appointment booking around the clock.

Is it worth it for small businesses too?

If the phone regularly can't be staffed, yes. If, on the other hand, you rarely get more than one call at a time and reliably answer it, you don't have a waiting-time problem that an AI needs to solve.

Conclusion

AI cuts waiting times mainly because it holds many conversations at once and never takes a break. The effect is strongest for appointments, standard information and status enquiries. Complex requests are taken in faster but remain the job of your team. Measure your peak times and the share of routine requests, and you will know how much relief is realistic.

The page AI practice assistant for doctors' practices shows how this can be put into practice for practices with phone peaks.