There is no single magic number for how slow is too slow, because the answer depends on what you sell and who else is answering the phone. But the shape is clear: every extra hour of delay gives a customer another chance to contact someone who will reply. The practical target for a South African small business is an instant automatic acknowledgement at any hour, and a first human response the same working day.

What is a realistic response time target for a small business?

Set two numbers, and only two.

TargetWhat it coversSuggested starting point
AcknowledgementAny automated or templated response, at any hourWithin seconds, 24 hours a day
First human responseA real person, with contextSame working day, ideally within two hours

The acknowledgement exists so nobody is left staring at a blank chat. The human response exists so the job can actually start. They are not the same thing, and businesses that confuse the two end up either chasing every message themselves at 22:00 or replying with a form letter and calling it done.

Pick a target, write it down, and put it where the team can see it. A target nobody knows about is not a target.

Why does speed matter more than the message you send?

Because a reply is not judged on its own. It is judged against the alternative, which is another business that answered. A perfectly worded answer that arrives on Tuesday afternoon has already lost to a mediocre one that arrived on Sunday night.

The other half of the problem is internal. A fast reply to a messy enquiry creates work, not a customer. If you cannot triage quickly, you will start avoiding the channel, and the response time will get worse rather than better.

How do you measure your actual response time?

Do not guess. Measure it once, properly, and write the number down.

  1. Open WhatsApp Business, email and your web form enquiries.
  2. For each new enquiry, note the time it arrived and the time a human first replied with something specific.
  3. Do this for twenty enquiries. Twenty is enough to see the pattern.
  4. Calculate the average delay, and separately the average delay for enquiries that arrived after hours.
  5. Work out the ratio: enquiries answered within your target, divided by total enquiries.

Then run this every month:

Monthly cost of slow replies
= (enquiries × average delay in hours ÷ 8) × cost of an hour of your team's time
  + enquiries lost to a faster reply × average job value × gross margin %

The first line is what slowness costs in staff time. The second is what it costs in work you never got. Most owners who run this calculation for the first time immediately know which problem is bigger for them.

What should you do when a lead arrives outside business hours?

Acknowledge it, capture it, and make sure it is impossible to miss in the morning. That is the whole job.

  • Turn on the WhatsApp away message so the customer gets a reply at 19:00 on a Sunday.
  • Make sure overnight messages stay unread, pinned, and labelled.
  • Set the phone to ring for starred contacts and business numbers, or use a second device that is not silenced.
  • Put a five-minute check into the morning routine: overnight list, then today’s expected arrivals.
  • Keep a list of genuinely urgent triggers, so the team knows what "urgent" means instead of guessing.

An out-of-hours acknowledgement is not a commitment to work overnight. It is a promise that the message was seen. Those are very different promises, and only one of them keeps the customer.

Does the channel matter when it comes to response time?

Yes, because the channel decides who has to be awake.

ChannelTypical reality in a small SA businessFix
WhatsAppThe one people actually use, especially on mobile dataAway message, shared devices, morning triage
Phone callHigh intent, but only reaches someone who is freeVoicemail with a callback promise and a named owner
Web formSlowest, and nobody is watching it liveInstant confirmation, routed to one inbox
EmailSlowest to be noticed, buried under newslettersNotification to a shared inbox, same-day sweep

Do not try to win every channel at once. Find the one your customers already use, get it to a same-day human response, then widen.

Who should reply when a lead arrives at 18:00 on a Sunday?

One named person, on a rota, with permission to send a holding message rather than a full answer. That is enough. The failure mode is not that nobody replied on Sunday. The failure mode is that the message was read by three people and actioned by none.

A rota with two or three names is more robust than a rota with one name and no backup. A rota with nobody on it is a decision, not an oversight, and it is fine as long as it is deliberate and the customer still gets an acknowledgement.

What slows a reply down more than anything else?

In practice, it is almost never the writing. It is the decisions a person has to make before they can reply.

  • Nobody knows the price, or the price lives in one person's head.
  • Nobody knows whether the work is possible this week.
  • The enquiry needs a site visit before an honest answer, and there is no rule about saying so.
  • Nobody can see the other enquiries, so they do not know what is already being handled.
  • The reply needs a manager's approval and the manager is asleep.

Fix the decisions, not the typing. An approved price list, a "we can be there on X" rule, and a named backup will take more enquiries off your desk than any template will.

Where do you start this week?

  1. Measure twenty enquiries by hand and write down the average delay.
  2. Publish one response time target where the team can see it.
  3. Switch on the after-hours acknowledgement for every channel you use.
  4. Name a primary and a backup for out-of-hours messages.
  5. Put a daily sweep in the diary — same time, ten minutes, every working day.

If you want to know where your response time actually stands before you change anything, book a free strategy session with Agentcy.