Gross Margin by Job: The Quarterly Pricing Review That Catches Drift Before It Compounds

One table of jobs with revenue, direct cost and margin percent, a floor that decides which jobs earn a question, the two drifts — price drift and cost drift — each with its own lever, the 45-minute quarterly ritual, and the five traps that turn margin into a number somebody reads and nobody moves.

Ask a small-business owner what their gross margin is and you will usually get one confident number — “about 40%” — which is the average. Averages are where pricing problems hide: the 60% jobs subsidize the 15% jobs, the blend looks healthy, and the losing jobs keep getting quoted because nobody ever put them side by side. Margin is not one number; it is a distribution over the jobs and products you actually sell, and once a quarter you owe that distribution forty-five minutes. This page is that forty-five minutes: it sits next to the monthly budget-vs-actuals hour (which reads the whole business) and the month-end close (which produces the numbers), and it answers the question neither of them can: which specific work is priced wrong, and which lever — price or cost — fixes it. It is also the evidence base for the price increase letter: an increase announced from a margin table is a decision, an increase announced from a feeling is a coin flip.

1. The one table — and the floor that decides what earns a question

The whole review is one table, one row per job type or product line (not per invoice — per kind of work):

• Revenue — what customers actually paid, discounts included. The quoted price is a hope; the invoice is the fact.
• Direct cost — materials, subcontractors, the labor hours the job consumed (from timesheets, not from the quote’s estimate), freight, and the platform or card fees on the sale. Overhead stays out — rent does not belong in a job’s margin.
• Gross margin — dollars and percent, both. Dollars tell you what the job contributes; percent tells you how hard it worked for that contribution.

Then one rule keeps the 45 minutes to 45 minutes:

• Set the floor once: a target margin per row type — for example 50% on service labor, 30% on product resale, 40% blended — agreed with the annual budget, not negotiated per quarter with your own anxiety.
• The threshold: any row more than 5 margin points below its floor gets a written one-line cause and a one-line action. Everything else is noise this quarter.
• Every below-floor row gets a name, not a mood: “emergency installs — 18% vs 50% floor — after-hours labor not quoted” is a finding. “Installs feel thin” is a complaint.

The floor is the whole trick. Without it, the review drowns in a spreadsheet and dies by winter; with it, two or three rows a quarter get read properly, and the rest are allowed to be fine.

2. The two drifts — and which lever each one pulls

Every margin miss is one of two drifts, and naming it correctly picks the lever for you:

• Price drift — revenue is the problem: discounts granted to win the job, scope that grew past the quote, a rate set two years ago and never revisited. The lever is the price increase letter, a tighter quoting rule, or a scope-change clause — never silent absorption.
• Cost drift — direct cost is the problem: a supplier’s price crept up, overtime ate the install, a subcontractor’s rate reset, freight went from free to $40 a carton. The lever is renegotiation on the supplier terms you already negotiated, a second quote on the offending line, or a materials surcharge line on the next quote.

Test before you pull: take the row’s price per unit and cost per unit and compare each against last quarter. The one that moved is the drift. Pulling the price lever for a cost problem poisons the customer relationship for nothing; pulling the supplier lever for a price problem burns goodwill internally and changes nothing.

3. The quarterly ritual — forty-five minutes, the week after quarter close

1. Pull closed revenue and closed costs for the quarter from the books — not the pipeline, the closed jobs. Comparing against half-finished work is how phantom margins get born.
2. Build the table — one row per job type, the three columns, margin percent beside floor. Fifteen minutes with a spreadsheet you keep; this page’s table never changes shape, so the build gets faster every quarter.
3. Mark every row past the threshold — in a healthy quarter that’s two to four rows. More than six means the floors are fantasy or the coding is drifting — fix the instrument before trusting the reading.
4. Write the one-liner per row: which drift, then which action. “Maintenance contracts — 31% vs 50% — price drift; 4% increase letter at renewal, March.” If the action takes more than one line, it goes on the task list, not the table.
5. Close the loop: price actions go into the renewal calendar with a date; cost actions go to the named supplier conversation; anything structural (a product line that cannot hold its floor) goes to the decision log — keep, reprice, or retire — so the same conversation does not restart every quarter.

Forty-five minutes, four times a year, calendar-blocked the week after the close. The quarter it feels skippable is always the quarter a job was quietly working for free.

4. The three honesty rules

• Labor at true cost, not at quote. The quote assumed eight hours; the timesheet says fourteen. Job margin computed from the quote is a forecast wearing a result’s clothes. Use the timesheet.
• Never average your way past a loser. A blended 41% can contain a 12%. The blend is for the bank; the distribution is for you. Read the rows, not the total.
• Every review ends in a decision or a date. A row that stays below floor for three quarters with no action is not a finding anymore — it is a subsidy you operate on purpose, and it should be written down as one (the decision log entry takes one line and ends the drift).

5. The five traps

• The blend that hides everything. One margin number for the whole business. The mix of jobs is the margin; the average is a costume the mix wears.
• Revenue masquerading as profit. The biggest job by revenue is often the thinnest by percent — scale is not margin, and the row that pays the rent is not automatically the row that pays you.
• The hero job at zero margin. The work you do “as a favor” or “because they’ve been with us forever.” Name it, floor it, or retire it — sentiment is a supplier you pay in hours.
• Repricing the losers, never the winners. The rows holding 65% get no attention for years, then quietly drift to 45% as costs rise. The floor works both ways — investigate the top too, because that is where the pricing power is.
• Review without a lever. Reading the table, sighing at the install line, closing the laptop. A margin review that produces no letter, no negotiation, and no decision log line was forty-five minutes of theatre.

6. Worked example — the signage shop’s two free jobs

A twelve-person signage company ran a blended gross margin of 41% and considered itself healthy. The quarterly table told a different story: fourteen rows, and two of them — after-hours emergency installs at 12% (floor 50%) and large-format printing at 18% (floor 35%) — were dragging the rest down. The cause read: emergency installs quoted day rate but consumed overtime at 1.5× (price drift); large-format media costs had risen twice in eight months while the price list had not moved since 2023 (cost drift). The actions: a scope-change clause and a 1.5× after-hours rate on the next quoting template, and a 6% price move on large-format at the next customer notice, with the media line re-quoted to a second supplier. Result at the next quarter’s table: installs 44%, large-format 33%, blended margin 41% → 46% — roughly $2,100 a month recovered on the same revenue, for one hour of reading and two letters. The rows that were already strong were left alone, because the review that attacks everything changes nothing.

Kits

Every page ships with a kit block — the paid tools behind the free advice:

• The First 30 Minutes — free incident quick-start checklist
• Ops Starter Kit — incident response for small teams — $14
• Ops Starter Kit Vol. 2 — advanced incident response & communications — $27
• Ops Mega Bundle — all 5 kits in one download — $49

Related: the weekly ops review is where the operational drift behind many a margin row gets caught weekly; the budget vs actuals monthly review is where a margin problem first shows up as a revenue or cost variance — this page is where it gets traced to the specific job and lever; the price increase announcement template is the lever for price drift — five lines, sent with notice, grounded in this table; the supplier payment terms checklist is the lever for cost drift — renegotiate the line, don’t absorb it; the debtor days review is the cash-side twin — margin earned late is margin financed by you; and the 13-week cash flow forecast is where the recovered dollars become visible weeks earlier than the P&L can show them.

The annual pricing decision runs on year-end margin — the year-end close checklist is what turns January’s numbers into December’s real ones.