Ask about pricing in any trade forum and you will find a holy war. Flat-rate shops call time-and-materials shops amateurs; T&M shops call flat-rate shops thieves. Both are arguing about identity when the real question is fit.
Full disclosure: we build price-book software, so we have an interest here. But a price book supports either model, so this piece can afford to be honest about when each one is the right choice.
What flat rate is actually good at
Flat rate shines on repeatable work: the water heater swap, the capacitor, the wax ring, the standard panel change. The task is known, the materials are known, and your own history says how long it takes. Pricing it once, carefully, beats pricing it forty times in forty driveways.
The customer gets certainty: a number before work starts, and no meter running while the tech waits on a part. The shop gets consistency: the price is the book's, not the tech's mood, so quoting stops being a personality trait. A good flat price also survives the slow tech and the fast tech alike, because it was averaged from real history instead of estimated in a driveway under pressure.
When T&M is the honest choice
Time and materials is honest exactly where flat rate is a guess: diagnostic-heavy and open-ended work. Chasing an intermittent electrical fault, opening a wall you cannot see behind, clearing a line that might be roots and might be a toy. Nobody knows the scope yet, so a fixed price is either padded against the worst case or a coin flip against your margin.
The padded price makes the easy version of the job feel like a rip-off, and the coin-flip price makes the hard version a loss. T&M with a posted hourly rate, a diagnostic fee, and a not-to-exceed number tells the customer the truth: we will know the size of this when we are inside it.
Flat rate prices the task; T&M prices the uncertainty, and a shop that admits which one it is holding quotes better with both.
Running the hybrid
Most shops that make peace with this run both, governed by a rule instead of a debate. Write the rule down and keep it where estimates get built, because a rule that lives only in the owner's head is just the holy war with extra steps. A workable version:
- If the task shows up in your job history ten or more times, it goes in the book at a flat price.
- If the job starts with the word 'diagnose', it starts as T&M with a posted diagnostic fee, and converts to a flat book entry the moment the diagnosis names a known task.
- If the scope is open-ended (remodel work, anything behind a closed wall), it runs T&M or as a proposal with allowances, and the estimate says so in writing.
- Either way, the customer sees the pricing basis before work starts. The model is allowed to vary; the disclosure is not.
Build the first fifty entries from your own jobs
Do not build your opening book from guesswork or from someone else's market, and do not stall for a year because a full book feels like a winter project. Fifty entries built from work you have already done and been paid for will carry you. The procedure:
- Pull the last twelve months of invoices. Ninety days will do if that is what you have.
- Tally which tasks repeat, sort by frequency, and take the top fifty. Those are your book, whatever a catalog vendor says your book should be.
- For each task, compute your real numbers from those invoices: average tech hours, actual materials cost, and how often it came back as a callback. Your numbers, not a rule of thumb.
- Price each entry as materials plus loaded labor plus overhead recovery plus profit. If you do not know your loaded labor cost (wage, taxes, insurance, the truck, the unbillable hours), compute that first, because every price downstream of a wrong labor number is wrong.
- Write each entry so a tech can read it aloud at a kitchen table: what is included, what is not, roughly how long it takes.
- Put a review date on the book. Materials prices move, and a book nobody reprices is a slow leak.
Fifty entries will cover most of what you quote, and growth after that is one entry at a time, whenever a job repeats. When someone insists their model is the only professional one, ask what they do with a mystery leak. The answer is usually a hybrid with better branding.