How many calls is my business missing?

Why the honest answer starts with "measure it"

This is the question every phone system gets sold on, and it is almost always answered with a statistic about somebody else's business. Resist that. Your missed-call rate depends on how many people answer phones, when your calls arrive, what your customers do when they cannot reach you, and how well your system is set up. None of that is captured by an industry average, and a figure quoted to you without anyone looking at your records tells you nothing except that someone wants to sell you something.

The good news is that this is one of the few business questions with a genuinely available answer. Your own call records already contain it. You need about two weeks of them and roughly an hour of attention.

A missed call is five different things

Most businesses count one of these and think they are done.

1. It rang and nobody answered. The obvious one, and usually the smallest of the five.

2. It never rang at all. Every line was busy, or the phone had quietly stopped being reachable. A device keeps its place in the system by re-registering periodically, and incoming calls are routed against that binding [1]; when the binding lapses, the handset still lights up and shows the right time but cannot be reached. Nobody in the building notices, because nothing rang.

3. It reached voicemail and died there. The message exists. Nobody actioned it. To the customer this is identical to no answer, except they think they are waiting on you.

4. It was answered, badly. A call answered for six seconds is a hang-up, an accidental transfer, or somebody picking up and putting the caller straight into a queue they abandoned. It appears in your records as answered, which is why raw answer rates flatter almost everyone.

5. It connected and could not be heard. Live voice is carried by a protocol that does not guarantee delivery, does not prevent out-of-order delivery, and does not assume the network underneath is reliable [2]. When that goes wrong, both parties give up. That is a lost call with a normal-looking record. Diagnosing it is a separate job, covered in why business calls keep dropping.

The two-week count

Export two weeks of call records from your phone service. You want the date and time, the calling number, which extension rang, whether the call was answered, and how long it lasted. On a hosted system you can usually pull this yourself. If you cannot, ask your provider for it in writing, and note carefully how hard that is, because a service that cannot show you a record of your own calls is telling you something.

Then produce five counts.

Unanswered. Rang, nobody picked up. Note the hour of day for each.

Straight to voicemail during business hours. These are your capacity and reachability failures, and they are the ones people never see.

Answered under ten seconds. Treat these as suspicious rather than successful, and check a few against what actually happened.

Repeat attempts. Any number that appears twice or more within about fifteen minutes. This is the highest-value line in the whole exercise: somebody actively trying to give you money.

Voicemails with no follow-up. Cross-check messages against your callbacks. This one is uncomfortable and it is usually where the biggest number hides.

The half that call records cannot see

Your records only contain calls that reached your system. They cannot show you a caller who dialled a wrong number from an old directory listing, hit an after-hours message an hour before you closed, or reached a menu they could not navigate and gave up.

So run the outside-in test alongside the count. Take a mobile that is not on your office network and call every number you publish — website, invoice, signage, online listings, directories — at your busiest hour, at lunch, and just before and just after closing. Write down what happened each time. Most businesses find at least one dead path the first time they try it. The full version of that audit is in why customers say they cannot reach you.

Turning the count into a decision

Now do arithmetic on your own numbers, not on anyone's marketing.

Take your repeat-attempt count for two weeks and annualise it. Multiply by the share of enquiries that normally become customers, then by what an average customer is worth to you over the time they stay. Be conservative at every step; a deliberately pessimistic figure that you believe is far more useful than an optimistic one you will discount.

Then compare that against what fixing the cause would cost. Our pricing is published on the pricing page so you can do this comparison without a sales call. Sometimes the answer is that you are missing three calls a month for small jobs and should do nothing. That is a legitimate outcome of measuring, and it is a better outcome than buying something on a feeling. The longer-form version of this arithmetic is in what missed calls cost a Canadian small business.

What the pattern tells you to fix

The shape of your count usually names the cause without further investigation.

Missed calls clustered at one hour means capacity or coverage. Start with how many phone lines your business needs and letting two people answer the same number.

Missed calls spread evenly through the day usually means nobody is specifically responsible for answering. See how to stop missing calls when you are with a customer.

Lots of voicemails, few callbacks, means a message-handling problem rather than a ringing problem: voicemail to email.

Calls landing after hours that you would have taken means your time rules are wrong. See after-hours and holiday call routing.

Calls going nowhere at random times, with no pattern, usually means registration or network. That is the same fault family as callers getting a busy signal.

One obligation to be aware of before you start

Call records are personal information about identifiable people, and PIPEDA's Schedule 1 applies to them. Identify the purpose you are collecting for, obtain knowledge and consent, keep your policies readily available to individuals, and destroy, erase or anonymise records once they are no longer required, under documented procedures [4].

Counting call outcomes for your own operational purposes is ordinary business practice. What is not ordinary is drifting from counting into recording calls, which is a separate activity with separate obligations and is dealt with in call recording under PIPEDA. Keep the two apart deliberately.

Where we sit

We run hosted phone systems for Canadian businesses, and we would rather you measured this before you bought anything from us. Most of Canada's 1.08 million small businesses [5] have no one on staff whose job is to pull call records and count outcomes, which is exactly why the question usually gets answered with a borrowed statistic instead of a real one.

What we provide is the records, in a form you can actually read, plus the fixes when the count points at something: capacity that does not run out at your busiest hour, registrations that recover on their own, ringing that reaches more than one person, correct hours, and messages that arrive as email rather than sitting in a box. If any provider, including us, tells you they will reduce your missed calls, ask what they tested and what the result was — a performance claim has to rest on an adequate and proper test and the burden of proving it falls on whoever made the claim [3]. Dependable telephone service is treated in Canadian law as infrastructure, which is the plain reading of the objectives in section 7 of the Telecommunications Act [6], and knowing whether yours is working should not require buying anything first.