Voice Systems
SyX Solutions · July 21, 2026 · 13 min read
A customer with a problem does not wait. They work down a list until somebody answers, and the person who answers gets the job. That is the entire mechanism, and it is why the calls you missed last week are not sitting in a queue waiting to be recovered. They were resolved by somebody else, usually within the hour.
This article makes that case using sources you can check. It also does something unusual for this category: it names three of the statistics most commonly used to sell you a solution to this problem, and reports that we could not trace any of them to a primary source. If an argument only works with unverifiable numbers, it is not an argument. The verifiable version turns out to be stronger.
A missed call in a service business is not a delayed sale. It is a completed sale that somebody else made.
The reason is that the customer is not shopping in the way a retail buyer shops. They have a broken thing, a deadline, or a leak. They open a map, a search result, or a neighbor's recommendation, and they start dialing. The first competent voice ends the search. There is no scenario where a homeowner with water coming through a ceiling collects three quotes and considers them over a weekend.
This creates an unusual property in the economics of a contracting business. Most marketing spend is designed to make the phone ring. The conversion step after the ring is the cheapest part of the funnel and it is the part most often left to chance, because it happens while everybody is busy doing the work the last call produced.
The rest of this article establishes three things: that response speed genuinely governs outcomes, that most of the specific numbers quoted about it are unreliable, and that the structural reason contractors miss calls is headcount rather than attitude.
The best-documented study in this area is a 2011 Harvard Business Review article, "The Short Life of Online Sales Leads," by James Oldroyd, Kristina McElheran, and David Elkington.
The authors audited 2,241 United States companies by sending each one a test inquiry and timing the response. The distribution was not close to good.
| Response time to a test inquiry | Share of 2,241 companies |
|---|---|
| Within one hour | 37% |
| One to 24 hours | 16% |
| More than 24 hours | 24% |
| Never responded | 23% |
Among the companies that did respond within 30 days, the average response time was 42 hours. Nearly a quarter never responded at all, to an inquiry they had presumably paid to generate.
A second analysis in the same article examined roughly 1.25 million leads across 29 business-to-consumer and 13 business-to-business firms. Companies that attempted contact within an hour of the inquiry were nearly seven times as likely to qualify the lead, meaning to reach a meaningful conversation with a decision maker, as those that waited a single additional hour. They were more than sixty times as likely as companies that waited 24 hours or longer.
Now the caveat that is almost always omitted when this study is quoted at you. Both analyses measured responses to inquiries submitted through web forms, in industries including financial services, automotive, education, software, and professional services. They did not measure inbound phone calls ringing at a contractor. The finding transfers as a strong general principle about response speed. It is not a measurement of your trade, and anyone presenting it as one is stretching it.
If you have encountered the claim that responding in five minutes rather than thirty makes you twenty-one times more likely to qualify a lead, and seen it attributed to Harvard, those two things came from different places.
That figure is not in the Harvard Business Review article. The Harvard numbers are hour-based: seven times at one hour versus two, sixty times at one hour versus a day. There is no five-minute comparison in the published text.
The twenty-one-fold figure comes from a separate study page published at leadresponsemanagement.org, associated with the same lead researcher and built on a dataset supplied by InsideSales.com, a sales software vendor. That page describes analysis of three years of data across six companies, covering more than fifteen thousand leads and over a hundred thousand call attempts, focused on leads obtained through web forms and validated in the mortgage and insurance sectors.
We are not saying that study is wrong. We are saying three things about it that you should know before it is used to sell you something. It is a vendor-supplied dataset rather than an independent one. It measures web form leads, like the Harvard research, not phone calls. And it is routinely presented under Harvard's name, which it does not carry.
Why this matters more than the number itself: when the most-quoted statistic in a category is misattributed this consistently, it tells you the category does not check its sources. That is worth remembering when you are being shown a slide of percentages by somebody who wants to run your phone.
We went looking for primary sources behind the three claims that appear most often in missed-call marketing. We found none of them.
That a specific large majority of customers who cannot reach a business immediately call a competitor. This appears everywhere, typically at eighty-five percent, usually with no attribution and occasionally attributed to a study that is never named. We could not locate an origin.
That a specific large majority of callers who reach voicemail leave no message. Same result. No traceable primary source. This one is uncomfortable for us to publish, because a version of it currently appears elsewhere on this website, written before we adopted this standard. It is flagged for removal, and we are naming it here rather than waiting until it is gone.
That service businesses miss a specific percentage of inbound calls, usually stated as thirty to fifty or as sixty-two percent. Also untraceable. This one appeared in an earlier version of one of our own articles and has been removed from it.
None of these is necessarily false. The first one in particular matches everything anybody who has run a service business has observed. But "obviously true" and "measured" are different categories, and a number presented as measured should be traceable to the measurement. When it is not, the honest move is to say so and make the argument another way.
The argument survives without them, which is the interesting part. Everything in the next two sections is checkable, and together it is a stronger case than the unsourced percentages ever were.
The fallback plan for a missed call is to call the person back. That plan is fighting a documented behavioral trend.
Pew Research Center surveyed 10,211 United States adults in July 2020 and found that 80 percent do not generally answer their cellphone when an unknown number calls. Sixty-seven percent described their practice as specifically declining unknown numbers while checking voicemail if one is left. Only 19 percent said they generally do answer.
Two honest limits on that finding. It is 2020 data, and it describes behavior toward unknown numbers generally rather than toward a business the person deliberately called an hour earlier. Somebody who dialed you is more likely to pick up your callback than a stranger. Your number is still unfamiliar on their screen, and the default is working against you.
The environment has grown more hostile since. Pew's 2025 survey of 9,397 adults found 31 percent of Americans receive scam calls at least daily, including 21 percent multiple times a day, and 68 percent at least weekly. The Federal Trade Commission recorded 2,085,133 Do Not Call complaints in fiscal year 2024, against roughly 254 million actively registered numbers. Ninety-one percent of American adults carry a smartphone, which means universal access to screening, blocking, and labeling.
Put plainly: the person you are calling back has been trained by years of unwanted calls to ignore numbers they do not recognize, and they have a device that helps them do it. The live call is worth substantially more than the returned one, and the gap grows every year.
Contractors do not miss calls because they are careless. They miss calls because the businesses are small and everyone is on a job site. This is measurable.
The Bureau of Labor Statistics publishes the Quarterly Census of Employment and Wages, counting establishments and employment by industry. The 2024 national figures for private specialty trade contracting:
| Industry, 2024 national | Establishments | Employees | Average per establishment |
|---|---|---|---|
| All specialty trade contractors | 588,440 | 5,151,234 | 8.8 |
| Residential plumbing, heating and AC | 92,844 | 672,777 | 7.2 |
| Residential electrical | 60,516 | 359,292 | 5.9 |
| Residential roofing | 25,307 | 135,211 | 5.3 |
Those averages include the owner and the field crews. A residential roofing establishment with 5.3 people has perhaps one person who is not on a roof, and that person is ordering material, chasing a supplier, or driving. The phone rings into that.
The picture is the same locally. In Harris County, Texas in 2024, residential roofing establishments averaged 4.8 employees across 225 establishments, residential electrical 9.1 across 404, and residential plumbing and heating and air-conditioning 10.0 across 815.
The Census Bureau's County Business Patterns program adds the distribution behind the average, which is the more damning view. In plumbing, heating, and air-conditioning contracting in 2022, 63,898 of 109,601 establishments nationally had fewer than five employees, and 84,448 of them, or 77.1 percent, had fewer than ten.
That is the real explanation. Three quarters of this industry is too small to dedicate a person to a phone, and the demand does not care.
Whatever number you have in your head for missed calls is probably low, because several categories of lost call never register as a missed call at all.
The most common is the call that reached a person and still went nowhere. Somebody answered from a truck, could not get to a calendar, promised to call back with a time, and the day swallowed it. That call shows in the log as answered. It converted no better than voicemail.
Then there is the caller who hangs up during the greeting. If your line plays a long recording before it does anything useful, a share of callers leave before the system counts them as a real call. Those people were ready to buy roughly fifteen seconds earlier.
There is the call that arrived while the line was already busy, which on many small business setups produces a busy tone or a silent roll to voicemail rather than a logged event. There is the call that went to a personal mobile because that number is what somebody put on a truck door or a yard sign three years ago, and it now sits in a personal voicemail box nobody treats as a business inbox. And there is the second call from the same person twenty minutes later, which looks like a duplicate in a report and is actually a customer running out of patience in real time.
None of this requires a statistic to check. Pull your own records for one busy week, count what you actually see, and then ask which of the categories above your system would even record. The gap between those two numbers is the part of the problem that has been invisible.
If you are paying to generate calls, every unanswered one is a loss you paid twice for.
The arithmetic is unforgiving. You spend money to put your number in front of somebody with intent. They call. Nobody answers. You have now bought a phone call for a competitor, because the caller continues down the list they were already working through, and the next business receives an intent-qualified lead at zero acquisition cost.
This is why the answering problem should usually be fixed before the lead generation problem, and it is the opposite of the order most businesses do it in. Turning up ad spend on an unanswered line increases the rate at which you fund other people's revenue. Fixing the line first raises the return on every marketing dollar already committed, including the free channels like search and referral that you are not measuring at all.
It also changes what your advertising data means. Campaigns that look unprofitable are sometimes campaigns whose calls arrived at the wrong hour. Until the phone is answered consistently, you cannot tell the difference between a channel that does not work and a channel you were not there for.
Answering the call is the entry requirement, not the win condition. The job is won when the appointment exists.
This is where most solutions to the missed-call problem stop one step short. A traditional answering service answers, takes a message, and passes it to you. That is genuinely better than voicemail, and the customer still hangs up without a time on the calendar, still holding an open decision, still with your competitor's number one tap away.
What actually closes the loop is narrower than it sounds.
Judge anything sold to you on this problem, ours included, on whether it produces appointments or messages. Everything else, including how convincing the voice is, is secondary to that one distinction. If you want to see the shape of a full build, the underground utility system shows how the intake and the site fit together, and the roofing comparison covers how the options in that trade differ.
Answering calls with an AI voice and placing them are governed differently, and the difference is written into the FCC's own ruling.
On February 8, 2024 the Federal Communications Commission released a Declaratory Ruling in CG Docket No. 23-362 confirming that the Telephone Consumer Protection Act's restrictions on artificial or prerecorded voice reach AI technologies that generate human voices. Callers must obtain prior express consent before making a call using such a voice, absent an emergency purpose or exemption.
The scoping sentence is the one that gets left out. The ruling states these requirements "are applicable to any AI technology that initiates any outbound telephone call using an artificial or prerecorded voice to consumers." A customer dialing your published number has initiated that call. Similarly, 47 CFR 64.1200(c)(1) bars telephone solicitations to residential subscribers before 8am or after 9pm local time, which restricts calls you place, not calls you receive.
This is a real distinction and not a blanket exemption, so three limits are worth stating.
You do not need an industry statistic to answer this. You need two of your own numbers.
How many calls did you not answer last week, and what is an average job worth to you. Multiply those, apply whatever close rate you honestly believe you would have hit, and you have the figure. It is your figure, drawn from your business, and it does not require you to trust a percentage from a vendor slide. The missed call calculator runs that arithmetic if you would rather not do it by hand.
If the answer is small, this is not your problem and you should go fix something else. A business whose phone does not ring does not need a better way to answer it, and we will tell you that on a call rather than sell you a system that sits silent until you cancel it. The businesses this works for already know their number is uncomfortable.
If the answer is uncomfortable, there are two next steps. The first costs you nothing: call (346) 514-5030. That is our own front desk, running the system we sell, at whatever hour you happen to read this. Interrupt it, argue with it, ask it something awkward. We publish it because a company selling answered phones should have to answer its own.
The second is a booked 30-minute discovery call. On it we work through your call volume, your average job value, your service area, and the jobs you refuse, and you leave knowing whether the math works. Terms are fixed and stated up front: no setup fee, a three month minimum, and a defined block of included minutes with overage at fifty cents per minute past it. We do not quote unlimited minutes on a metered channel, because nobody can honestly deliver that.
To read how the build works before talking to anyone, the voice agent overview covers it, and the contractors page covers scoping by trade.
Every figure above comes from one of the following. Government data files were downloaded and parsed directly rather than summarized.
Keep Reading
This article is part of the SyX Solutions blog. If you want the arithmetic instead of the argument, read what it actually costs you to miss a call, or compare the ways to get the phone covered in answering service vs AI receptionist. The build itself is the AI voice agent, backed by missed call automation.
Two of your own numbers
Call (346) 514-5030 and hear the system we run our own business on. When you want the math done on your call volume and your job value, book a 30-minute discovery call. No setup fee, three month minimum.