How Dispatch Actually Gets Decided at a Small Field-Service Company — and What It Costs When It's Wrong

At most small plumbing, HVAC and electrical companies the order jobs get worked in is decided by the order the phone rang. That is not a policy anyone chose. It is what happens when the intake record is a voicemail box or a shared inbox and the only sort available is chronological. It costs money in a specific and calculable way: when vans are the scarce resource, the sequence you send them out in determines how many jobs fit into a day and which ones you never get to. This page prices a single truck roll from published federal data, works through the sequencing arithmetic, lists what the dispatch platforms charge as of 26 August 2026, and is explicit about which commonly quoted industry numbers could not be traced to a primary source.

//What one truck roll costs, built from published inputs

A figure circulates constantly in field-service marketing: a truck roll costs somewhere between $150 and $500. It appears across vendor blogs, glossaries and sales decks. Traced in August 2026, every instance found cites another vendor page or nothing at all — no study, no survey, no described methodology. The figure is not necessarily wrong. It is simply unsourced, so it is not used here. What follows is built instead from numbers that do have a source, and the arithmetic is shown so it can be checked or corrected with your own inputs.

For the vehicle, the Internal Revenue Service sets the standard business mileage rate at 72.5 cents per mile for 1 January through 30 June 2026 and 76 cents per mile for 1 July through 31 December 2026, a mid-year increase the agency published rather than waiting for the annual revision. That rate is constructed to cover fuel, depreciation, insurance and maintenance, which makes it a defensible proxy for the vehicle side of a service call. For context on the fuel component, the US Energy Information Administration put the national average retail price at $4.085 a gallon for regular gasoline and $5.652 for on-highway diesel in the week ending 24 August 2026.

For the labour, the Bureau of Labor Statistics puts the median wage for plumbers, pipefitters and steamfitters at $30.27 an hour, or $62,970 a year, as of May 2024, across 504,500 jobs. Wage is not what an employee costs an employer. The BLS Employer Costs for Employee Compensation release for March 2026 puts private-industry total compensation at $46.60 per hour worked, of which $32.60 is wages and salaries and $14.01 is benefits — benefits being 30.1% of the total. That implies a burden multiplier of about 1.43 on wage, before any allowance for overhead, non-billable hours or the office staff who booked the job.

That $94 is a floor, and it is the cost of the visit happening at all — before parts, before the office time to book and route it, before the share of rent, insurance and licensing the job has to carry. It is also, and this is the part worth sitting with, the cost of the visit happening twice. A return trip because the part was not on the van is a second $94 that no customer pays for, and it consumes a slot that a billable job could have used.

//Why first-come-first-served loses money once the day is full

Sequencing only matters when capacity binds. On a slow Tuesday with four vans and six jobs the order is irrelevant, because everything gets done. The order starts costing money the moment the day is full, and a full day is the normal condition at a company that is doing well. Once every slot is spoken for, each job accepted is a job declined, and the question stops being whether the sequence is fair and becomes whether it is the most valuable one available.

Here is that stated as a worked example, with assumptions on the table rather than buried. A three-van shop has twelve slots in a day. Fourteen requests arrive. Two are no-heat calls in January that the customer will pay an after-hours premium for and will call a competitor about within the hour. Ten are routine — a dripping tap, an annual service — where the customer is content to take Thursday. Two are estimate visits with no revenue attached that day. Worked chronologically, the two no-heat calls land wherever they happened to arrive in the queue, which on average is the middle. If they arrived seventh and eleventh, ten lower-value jobs get dispatched ahead of them and one of the two emergencies falls off the end of the day. Worked by urgency, both emergencies go out first, the routine work slides to Thursday without anyone minding, and the same twelve slots produce a materially different day. Nothing about the crew, the vans, the pricing or the demand changed. Only the sort order did.

The second effect is quieter and harder to notice from inside. A queue with no memory cannot tell you that the call which arrived at 9pm on Saturday has now been waiting thirty-four hours while a call from 7:01 on Monday is being handled. Chronological order inside the staffed window is not chronological order overall. It is chronological order among the calls that happened to arrive while somebody was listening, which is a different and much less defensible thing. Most shops that believe they operate first-come-first-served are in fact operating first-to-reach-a-human, and those two rules produce different queues.

//What is actually known about response time, and what is not

The most frequently cited evidence on response speed is a 2011 Harvard Business Review study by James Oldroyd, Kristina McElheran and David Elkington. The authors audited first-response times at 2,241 US companies and separately analysed 1.25 million sales leads received by 29 business-to-consumer and 13 business-to-business firms. Of the audited companies, 37% responded within an hour, 16% responded within one to 24 hours, 24% took longer than 24 hours, and 23% never responded at all. Among companies that responded within thirty days, the average response time was 42 hours. Firms that made contact within an hour were nearly seven times as likely to qualify the lead as those that waited even one hour longer, and more than sixty times as likely as those that waited a day or more.

That study is worth reading with its limits attached, because it is routinely quoted without them. It measured web-form enquiries to companies across a range of industries, not emergency service calls to plumbers, and it was published fifteen years ago. It establishes that response speed has a large effect on whether an inbound enquiry converts at all. It does not establish the size of that effect for a homeowner standing in front of a failed water heater at 9pm, and no primary, methodologically described study specific to the residential trades was found when searching in August 2026. The direction of the finding is well supported. The magnitude, for this industry, is not published.

//What dispatch software costs

The number that matters in that table is the slope, not the entry price. Jobber counts as a user anyone who opens the account to view or manage the schedule, office or field alike, so a six-person shop on Core pays $49 plus five times $29, which is $194 a month. Housecall Pro includes five users on Essentials and charges $100 for the sixth, putting the same six-person shop at $289. Service Fusion charges nothing per user on any tier and states plainly that a shop with one technician and one dispatcher pays the same as one with twenty technicians and a full office, which makes its $245 entry price the most expensive of the three at two users and the cheapest well before you reach eleven. None of these is mispriced. They are priced for differently shaped businesses, and the shape of yours decides which one is expensive.

Two large vendors in this category, ServiceTitan and Workiz, publish no price at all. Their pricing pages carry a form. That is a legitimate commercial choice and it is also a fact worth recording, because it means any figure you find for them online came from a review site rather than the vendor, and the review sites do not agree with each other.

//What could not be verified

The other number that circulates as settled fact in this industry is the first-time fix rate, usually quoted somewhere between 70% and 80% as an industry average. Searching in August 2026 for the primary source produced software vendor blogs and glossary pages, several of them citing Aberdeen Group research that is no longer publicly retrievable and whose sample, date and definition are therefore uncheckable. If a current, published, methodologically described benchmark for first-time fix rates in the residential trades exists, it was not found. Anyone quoting one — including anyone selling dispatch software — is worth asking where it came from.

The truck-roll cost figure has the same problem, which is why this page built its own from federal data instead. And no published dataset was found giving conversion rates by response time for emergency residential service calls specifically, as distinct from online sales leads generally. Three widely repeated numbers in one industry, none of them traceable, is a reasonable prompt to be sceptical of the fourth.

What is verifiable is narrower and more useful. A single visit has a direct cost that can be constructed from federal wage and mileage data and lands near $94 under the stated assumptions. Capacity binds on exactly the days when the stakes are highest. Sequence determines which jobs fit inside that capacity, and a chronological queue does not sort by value, urgency or even by true arrival time once the phone stops being answered. The software question follows from that arithmetic rather than the other way round, which is the reverse of how it is usually sold.