Missed calls are the cheapest money in the business

An enquiry that rings out has already been paid for. The marketing, the reputation, the twenty years of referrals that made somebody pick up the phone, all of that cost is sunk before the first ring, which is what makes the unanswered call the cheapest revenue in a service business to recover. It is also the one nobody can see, because a call that was never answered is not in any report.

First, ignore the statistics you will be shown

Search for anything about missed calls and you will be buried in numbers. Sixty-two percent of business calls go unanswered. The average contractor loses a hundred and twenty-six thousand a year. Missed calls cost the industry a hundred billion. These are quoted everywhere, cited by each other in a loop, and almost all of them originate with companies that sell call-answering services.

Look for the methodology behind any of them and there generally is not one. No sample size, no definition of what counts as a missed call, no basis for the revenue-per-call figure that turns a percentage into a headline loss. Numbers built that way are not evidence, they are advertising with a decimal point, and repeating them is how a real problem gets treated as a marketing claim.

This matters beyond pedantry. An owner who is quoted an absurd number, checks it against their own business, and finds it obviously wrong will dismiss the whole subject, including the part that is true. The underlying problem is real and it is worth fixing. It just does not need inflating, and the smaller true number is more useful because you can act on it.

What the research actually shows

There is one piece of proper work here and it is fifteen years old. In March 2011 Harvard Business Review published an audit by James Oldroyd, Kristina McElheran and David Elkington covering 2,241 US companies. They sent web-generated test enquiries and measured what came back.

The distribution is the part worth sitting with, because it is not the shape most owners expect.

Read the bottom two rows together. Forty-seven percent of a national sample, given a genuine enquiry from somebody trying to buy, either took longer than a day or never replied. Among those that did eventually answer inside thirty days, the average time to first response was forty-two hours. Not forty-two minutes.

Nearly one company in four, audited across a national sample, never answered a genuine enquiry from somebody trying to give them money.

The same group of researchers had already published the speed finding, from a separate study of about 15,000 leads. Contacting somebody within five minutes rather than thirty made a firm twenty-one times more likely to qualify that lead. The HBR audit put the weaker version of the same effect at company level: respond inside an hour and you are close to seven times more likely to qualify a lead than a firm that responds later.

Two caveats belong here, because otherwise this becomes the same kind of claim I just objected to. Both studies are about web-form enquiries at mostly larger US firms, not phone calls to a fifteen-person trade business in Bavaria. And both are old enough that customer expectations have only tightened since. So do not treat the multipliers as your numbers. Treat them as the direction and the rough scale of the effect, which is all any external study can honestly give you.

Why good businesses miss calls

The reason this is worth fixing rather than complaining about is that missed calls in a trade business are structural. Nobody is being lazy. The arithmetic simply does not work.

1. Everybody capable of answering is on a job

   The people who can actually speak to a technical enquiry are the same people out doing the work. There is no shift where someone competent sits by a phone, because that person would be worth more in a van.

2. The calls arrive when a machine breaks

   Demand for a repair service is not distributed evenly across a working day. It clusters, badly, and it clusters at exactly the moments the crew is already committed. Weekends and early mornings are worse.

3. There is no queue, so there is no evidence

   A ring-out leaves no record. Not a lost enquiry, not a callback task, nothing. The company cannot know how many it missed because the missing is invisible by construction.

4. Voicemail is treated as a solution

   It is not. A caller with a broken shutter and a truck waiting does not leave a message, they call the next company. Voicemail converts a missed call into a missed call that feels handled.

5. Nobody owns the callback

   Even when a message does get left, returning it is nobody's specific job. It falls to whoever notices, which in practice means whoever is least busy, which in practice means it happens tomorrow.

Notice that none of the five is solved by trying harder or by hiring a receptionist, which is the answer most owners reach for and the reason most of them drop it. A receptionist covers office hours, costs a salary, and still cannot answer a technical question about a fire door. The problem is not a missing person. It is a missing record.

Measuring your own, honestly

The only number worth acting on is yours, and getting it is a week of work rather than a project. You need three things your phone provider can already tell you, and one you have to look up yourself.

Worked through, it looks like this. Every number below is a placeholder chosen to show the shape of the calculation. None of them is a finding, none of them is a benchmark, and all four have to come out of your own records before the result means anything.

That yearly figure is the floor, and it is the one to trust, because it rests on two inputs you can pull from records rather than estimate. The lifetime version is larger and softer. If one in four of those eight jobs would have become a maintenance customer, the annual loss is not 43,200 of revenue but two dozen service relationships a year that never started, each carrying a decade of inspections behind it. That number is bigger by an order of magnitude and it compounds three assumptions instead of two, so treat it as the direction rather than the amount.

Being explicit about which of those two you are quoting is the whole discipline. The floor is what you act on and what you would defend to a bank. The lifetime figure is what explains why the floor understates the damage, and it is also, precisely because it is soft, the number the vendors at the top of this page are quietly using when they tell you what an unanswered call costs.

Two more rules keep this honest. Use your own conversion rate from your own records, never a benchmark, because the benchmark is the thing being sold to you. And run the second calculation on lifetime value as well as first job, because in a service business with a maintenance cycle the difference between those two numbers is where the actual answer lives. A three hundred euro callout that becomes a twelve-year service relationship is not a three hundred euro loss.

You will almost certainly find the total is smaller than the industry claims and larger than you assumed. That is the useful place to land. It is also, in my experience, the first number that makes an owner take the rest of the operating layer seriously, because unlike most efficiency arguments it is unambiguous: the customer wanted to buy, and nobody picked up.

Fixing it, in order

The sequence matters, because the first step is worth more than the rest combined and costs almost nothing. Start at the top and stop when it is good enough.

• Make every unanswered call create a record. Not voicemail, a record, with a number attached and a status that starts as open
• Send an automatic message within seconds of the missed call, from a number the caller can reply to, saying when somebody will ring back
• Give the callback an owner and a time limit, and let the record stay open until it is closed by a human
• Route out-of-hours and overflow somewhere that captures details, before you consider anything that tries to answer questions
• Only then look at anything that handles the conversation itself, and only for triage, never for pricing or technical advice
• Review the open records weekly, because the list of who was missed is also the cheapest sales list in the business

The order is deliberate and the last two are where people go wrong. Buying a system that answers the phone in a natural voice is the exciting option and the wrong first step, because it solves the interesting problem instead of the expensive one. The expensive problem is that the enquiry vanished. Capture it and a human closes it. Skip capture and you have an impressive system that still loses the call whenever it cannot cope.

What it is actually worth

Recovered enquiries are the fastest of the four layers to move, which is why this is where I start. It is also bounded, and being clear about the bound is what separates this from the numbers at the top of the page. You are recovering a portion of the calls that currently go nowhere, at your existing conversion rate, on your existing prices. You are not doubling the business.

What makes it worth doing anyway is the cost side. There is no additional marketing spend, no new customer to win, no price increase to defend, and no capacity to add, because these people were already trying to buy from you. That is a very unusual shape for a revenue improvement, and it is why it goes first, ahead of anything more ambitious. The layer it belongs to, and why fixing it before the others is the right order, is in the four-layer growth engine. Why nobody inside the business has got to it yet is in the product cannot be disrupted, the operations can, and if you want the whole operating picture rather than one leak, that is the system.

Response-time findings from James B. Oldroyd, Kristina McElheran and David Elkington, "The Short Life of Online Sales Leads", Harvard Business Review, March 2011 (audit of 2,241 US companies), and the same authors' Lead Response Management study of approximately 15,000 leads.