Most technician onboarding programs are a sentence: "Ride with Mike for two weeks." Two weeks later the new hire gets a truck, whether or not they can run a no-cool call alone, and Mike goes back to his own numbers having trained nobody in particular. A ride-along program works when it's built like a ladder instead: five rungs, each with a written exit test, signed off one job type at a time, and scheduled by dispatch rather than by the calendar.
This post lays out that ladder, a 30/60/90 plan built on it, how to pay the mentor so they want to do it, and a worked example of what cutting a few weeks off the ramp is worth.
Why "ride with Mike for two weeks" fails
The calendar version breaks in three predictable ways.
There's no definition of done. "Two weeks" says how long, not what the new tech should be able to do at the end. So the decision to cut them loose is made by the schedule, usually on a Monday when someone called out.
The mentor never fades. Without stages, the senior tech does the work and the new hire hands over tools. That's comfortable for both of them and teaches very little. The new tech's first truly solo call is also their first time leading a customer conversation.
The mentor pays for it. A senior tech on commission or performance pay who brings a trainee along runs slower, sells less and explains more. If the pay plan doesn't account for that, the best techs avoid mentoring, and the job falls to whoever has the lightest board.
The stakes are real. In the Federal Reserve's latest small business survey, 46% of employer firms named hiring or retaining qualified staff as an operational challenge, second only to reaching customers and growing sales1. BLS says new HVAC technicians typically need a lengthy period of on-the-job training to become competent, and it classifies the occupation's typical training as long-term on-the-job training, which it defines as more than 12 months2. You won't shortcut that year. You can stop wasting the first three months of it.
What the research says new hires actually need
There isn't a peer-reviewed study of HVAC ride-alongs. There is solid research on how new employees adjust, and it lines up well with how good trade mentors already work.
A 2007 meta-analysis in the Journal of Applied Psychology pulled together 70 samples of new employees and found that three signs of adjustment were the link between how people are onboarded and their later performance, satisfaction, commitment and turnover: role clarity (knowing what's expected), self-efficacy (confidence you can do it) and social acceptance (feeling part of the crew)3. The lead author's practical guide for the SHRM Foundation turns that into four building blocks, the "Four C's": compliance, clarification, culture and connection. By her estimate, only about 20% of organizations onboard proactively across all four; most cover the rules and the job and leave culture and connection to chance4.
The teaching method comes from an older idea. Educational researchers Allan Collins, John Seely Brown and colleagues looked at how traditional apprenticeships actually transfer skill and described it in four moves5:
There are four important aspects of traditional apprenticeship: modeling, scaffolding, fading, and coaching.
The word that matters for a ride-along program is fading. The mentor deliberately does less over time. A program that has no planned fading is just two people in one truck.
BLS describes the same progression in plain terms for HVAC: new techs start with basic tasks like insulating refrigerant lines or cleaning furnaces and move on to harder ones like cutting and soldering pipe and checking electrical circuits2. The ladder below just writes that progression down and attaches a test to each step.
The five-rung ride-along ladder
| Rung | Who does the work | Who talks to the customer | Mentor's job | Exit test |
|---|---|---|---|---|
| 1. Watch | Mentor | Mentor | Model the work and narrate the thinking out loud | New tech can explain each step back, and the safety checks for this job type |
| 2. Hands | New tech, on tasks the mentor assigns | Mentor | Hand over tasks; check each one | Completes the job's tasks correctly on 3 jobs without correction |
| 3. Lead with a net | New tech, whole job | New tech | Stay silent unless safety or a wrong diagnosis is at stake; debrief in the truck | 3 clean jobs of this type: right diagnosis, readings recorded, options presented |
| 4. Solo with review | New tech, alone | New tech | Review photos, readings and the invoice the same day; take calls | 5 jobs of this type with no callback within 30 days and complete documentation |
| 5. Solo | New tech | New tech | Spot-check monthly | Ongoing: callback rate and documentation in line with the team |
Two rules make the ladder work.
Rungs are per job type. A new tech can be at rung 5 on maintenance tune-ups, rung 3 on no-cool diagnostics and rung 1 on heat pump defrost problems, all in the same week. That's normal, and it's how you get someone producing in week three instead of week twelve.
Rung 3 is the one that matters. The mentor keeps quiet while the new tech leads the customer conversation. It's uncomfortable, and it's the only rung that builds the confidence the research points to. Most informal programs skip straight from 2 to 4.
Sign off by job type: the competency matrix
The ladder needs a short list of job types. Use the ones in your price book or dispatch system, not a training manual's version. For a residential HVAC shop, eight to twelve is plenty.
| Job type | Typical first rung-5 week (example) | Exit test detail |
|---|---|---|
| Cooling or heating tune-up | Weeks 2 to 4 | Full checklist, readings logged, photos of the data plate and coil |
| Filter, thermostat and simple swaps | Weeks 2 to 4 | Correct part, tested operation, customer shown the change |
| No-cool or no-heat diagnosis | Weeks 6 to 10 | Correct root cause on 3 observed calls; options presented on every one |
| Capacitor, contactor and motor repairs | Weeks 5 to 8 | Correct diagnosis before parts; amps recorded before and after |
| Refrigerant leak search and repair | After 608 certification | Leak located and documented; charge verified by method, not by guess |
| Heat pump defrost and reversing valve | Weeks 10 to 16 | Sequence of operation explained; correct component on 3 observed calls |
| Water heater replacement (plumbing) | Weeks 6 to 12 | Code items checked against your local requirements; permit handled |
The week ranges are an example for planning, not a standard. Nobody publishes a reliable time-to-proficiency figure for these jobs, and your own will depend on who you hire. A helper with two years at another shop will move very differently from someone just out of a six-month program.
What makes a matrix useful is that dispatch can read it. If a new tech is at rung 5 on tune-ups and rung 3 on diagnostics, the board should send them solo to tune-ups and with a mentor to no-cool calls. When that lives in the dispatcher's memory, it breaks the first busy Monday.
A 30/60/90 day plan built on gates
Bauer's guide recommends milestones at 30, 60, 90 and 120 days and up to a year4. Here's how they map to the ladder. The dates are checkpoints, not promotions; promotions happen at the gates.
Day 1
Compliance and the truck
Paperwork, safety training for the hazards of your work (OSHA requires instruction in recognizing and avoiding unsafe conditions), ladder and fall rules, vehicle policy, phone and app setup. Meet the dispatcher and the warehouse lead.
Week 1
Rung 1 on everything
Ride with one mentor. Every evening, the new tech writes down three things they learned and one question. The mentor answers the question the next morning.
Day 30
First review
Rung 5 on two or three simple job types. Rung 2 or 3 on diagnostics. Review: what's clear, what isn't, who they go to for help. Ask about the crew, not just the work.
Day 60
Producing on simple work
Running solo maintenance days, and rung 3 or 4 on common repairs. First look at their own numbers: documentation, callbacks, options presented.
Day 90
Rung 4 or 5 on the core
Most common repair types at rung 4 or 5. Set the next three job types to learn, and a date to revisit pay.
Day 120 to 1 year
The long tail
Advanced diagnostics, commercial or specialty work, a license or certification if your trade requires one. This is where the long-term training BLS describes actually happens.
OSHA's construction rules require employers to instruct each employee in recognizing and avoiding unsafe conditions and the regulations that apply to their work7, and there are separate training rules for fall protection and ladders. Service work can fall under OSHA's general industry rules instead of construction, so check which apply to your jobs. Either way, the day-one safety block isn't optional.
Pay the trainee, and pay the mentor
The trainee's time is work time. Under federal rules, training time can be left unpaid only if it is outside normal hours, truly voluntary, not directly related to the job, and involves no productive work8. A ride-along fails every one of those tests. Pay new hires for all of it, including the evening question log if you require it. If you use a paid trial day for candidates, pay it at the full rate too. The commission and overtime math for new techs is in how to pay technicians.
The mentor needs a reason. A senior tech on performance pay gives up production to train. Three fixes, in order of how much they help:
- A sign-off bonus per job type. A fixed amount each time a new tech passes a rung-4 exit test on a job type the mentor trained. It pays for the outcome, not for the hours of riding along.
- Protected commission. On jobs where the trainee leads (rung 3), credit the sale to both, or to the mentor. Otherwise you're asking your best seller to hand their customers to a beginner for free.
- A lighter board. Schedule mentor and trainee for fewer jobs on rung 1 to 3 days. Pretending they'll run a full board is how mentors learn to skip the teaching.
Pick one mentor per new hire for the first 30 days. The research on social acceptance is a good reason: one person who answers the questions and introduces the new tech to the crew does more than five who each ride along once.
What a faster ramp is worth: a worked example
All numbers here are hypothetical. Replace them with yours.
Say a new tech earns $24 an hour. BLS's June 2026 compensation survey puts total cost for installation, maintenance and repair workers at $49.01 per hour against $33.61 in wages, about 1.46 times9. That makes the new tech cost about $35 an hour loaded, or $1,400 a week.
Hypothetical weekly contribution of a new tech
Assume a fully producing tech generates $4,000 a week of gross profit from their own work, and that a new tech produces 0%, 20%, 50% and 80% of that on rungs 1 to 4. Mentor drag (the mentor's lost output) is $600 a week while the new tech is on rungs 1 to 3 and $200 on rung 4.
Compare two ramps for the same person. The calendar ramp moves everyone one rung every four weeks. The gated ramp moves job types as soon as they pass, which in this example gets the new tech's average rung up about a week or two sooner at each stage.
| Cumulative contribution (hypothetical) | Calendar ramp | Gated ramp |
|---|---|---|
| End of week 4 | −$8,000 | −$6,400 |
| End of week 8 | −$12,800 | −$7,600 |
| End of week 12 | −$12,800 | −$200 |
| End of week 16 | −$6,400 | +$10,200 |
| End of week 26 | +$19,600 | +$36,200 |
Invented inputs: $1,400 a week loaded cost, $4,000 a week full gross profit, output of 0/20/50/80/100% by rung, mentor drag $600 on rungs 1 to 3 and $200 on rung 4. Calendar ramp: one rung per four weeks. Gated ramp: rungs at weeks 3, 6, 9 and 12.
| Calendar ramp | 19 wks |
|---|---|
| Gated ramp | 13 wks |
In this example the calendar ramp breaks even in week 19 and the gated ramp in week 13. By six months the difference is about $16,600 for one hire. The exact figure doesn't matter. The shape does: the cost of a new tech is front-loaded, so every week you shave off the early rungs pays back twice, once in less cost and once in earlier production.
The same math shows why losing a new tech in month two is so expensive. You've paid for the bottom of the curve and left before the top. That's the subject of why technicians quit.
The weekly check-in: five questions, 15 minutes
A ladder needs a short, fixed weekly conversation with the new tech, the mentor and the service manager. Keep it to five questions so it actually happens:
- 1
What did you sign off this week?
Name the job types and rungs. Update the matrix on the spot.
- 2
What's next, and when?
One or two job types to move up next week, and which days the board will give them those calls.
- 3
What did the numbers say?
Documentation complete, readings recorded, options presented, any callbacks. Look at the actual jobs, not a summary.
- 4
What are you unsure about?
Ask the new tech directly, then ask the mentor separately. The answers are often different.
- 5
Who have you worked with?
Have they met the dispatcher, the warehouse lead, another senior tech? Connection is the C most programs skip.
A rough number from Bauer's 2010 guide gets quoted a lot: half of hourly workers leave new jobs within the first 120 days4. The guide cites it from a 2010 conference presentation, so treat it as a warning, not a current measurement. BLS's tenure data is a different measure but points the same way: in January 2024, about 22% of wage and salary workers had been with their current employer a year or less10. The first four months are when a new tech decides whether you're worth staying for.
How TradesWork helps run the ladder
You can run all of this on a whiteboard and a spreadsheet. Two pieces of TradesWork fit it directly. Technicians carry a skills and certifications list (with expiry dates), and job types can require skills, so the dispatch board flags a missing skill when someone tries to send a new tech solo to work they haven't been signed off on. And forms and checklists can be attached on demand to a job, so a mentor can fill a rung sign-off form on the job where it happened, with photos. There's no dedicated training-plan screen; the matrix itself lives in your own document.
What the ladder buys you
- Clear exit tests make promotion decisions fair and explainable
- New techs produce on simple work weeks earlier
- Mentors are paid for outcomes, so the best people volunteer
- Dispatch can follow the matrix instead of guessing
What it costs
- Takes a few hours to write the job-type list and exit tests
- Rung 3 days run slower and need lighter boards
- Needs a service manager who holds the weekly check-in every week
- Mentor bonuses add cost before the new tech produces
Ride-along program checklist
- Job types for the matrix taken from your price book or dispatch list
- Written exit test for each rung on each job type
- One named mentor per new hire for the first 30 days
- Day-one safety training documented
- 608 status and license supervision rules checked before refrigerant or licensed work
- Trainee paid for all ride-along and training time
- Mentor sign-off bonus and commission credit written into the pay plan
- Lighter boards scheduled for rung 1 to 3 days
- Weekly five-question check-in on the calendar
- 30, 60 and 90 day reviews booked on the day they start
If you're building a longer pipeline than one hire at a time, the ladder plugs straight into a registered apprenticeship, which adds classroom instruction and a wage schedule around the same on-the-job learning.
Footnotes
-
2026 Report on Employer Firms: Findings from the 2025 Small Business Credit Survey, Federal Reserve Banks, March 2026. ↩
-
Heating, Air Conditioning, and Refrigeration Mechanics and Installers: How to Become One, Occupational Outlook Handbook, U.S. Bureau of Labor Statistics; training category definitions at Measures of education and training. ↩ ↩2
-
Newcomer adjustment during organizational socialization: A meta-analytic review of antecedents, outcomes, and methods, Talya N. Bauer, Todd Bodner, Berrin Erdogan, Donald M. Truxillo and Jennifer S. Tucker, Journal of Applied Psychology 92(3), 2007. ↩
-
Onboarding New Employees: Maximizing Success, Talya N. Bauer, SHRM Foundation Effective Practice Guidelines Series, 2010 (copy hosted by the University of Georgia). ↩ ↩2 ↩3
-
Cognitive Apprenticeship: Making Thinking Visible, Allan Collins, John Seely Brown and Ann Holum, American Educator, Winter 1991. The original framework is Collins, Brown and Newman, "Cognitive apprenticeship," in Knowing, Learning, and Instruction (Erlbaum, 1989). ↩
-
Section 608 Technician Certification, U.S. Environmental Protection Agency; rule text at 40 CFR 82.161. ↩
-
29 CFR 1926.21, Safety training and education, Occupational Safety and Health Administration; see also 1926.503 (fall protection training) and 1926.1060 (ladder training). ↩
-
29 CFR 785.27, General (training programs), Code of Federal Regulations, U.S. Department of Labor rules on hours worked. ↩
-
Employer Costs for Employee Compensation, Table 4, private industry by occupational group, U.S. Bureau of Labor Statistics, June 2026 data, released September 2026. ↩
-
Employee Tenure in 2024, U.S. Bureau of Labor Statistics, January 2024 data, released September 2024. ↩