The usual assumption in a small operation is that the busy work is the profitable work, and that a year with more jobs in it is a better year than one with fewer. Both are frequently wrong, and the only way to find out is to attach costs to individual jobs rather than to months. Almost nobody does this, not because it is difficult but because it looks like accounting and everybody has enough of that already. It is not accounting. It is four numbers per job, recorded as the job happens, and it usually changes what gets quoted within a quarter.
What to Record on Each Job
Four fields cover it: what was quoted, hours actually worked by everybody who touched it, materials actually consumed, and anything paid to somebody else. A note field for what went unexpectedly wrong is worth adding and is the one people find most useful later. That is the entire sheet, one row per job, and it works equally well in a spreadsheet, a notebook, or the back of the job folder. The format matters far less than whether it gets filled in on the day the work happened.
The discipline that matters is recording hours on the day rather than at the end. Reconstructed hours are systematically low, because nobody remembers the second trip to the supply house or the forty minutes spent on the phone with the customer about a change, and those omissions land almost entirely on the jobs that went badly. A costing sheet built from memory will make a difficult job look like a fine one, which is precisely backwards. The jobs most in need of examination are the ones least likely to be remembered accurately.
The Costs People Forget to Attach
Three categories go missing from nearly every attempt. Drive time and fuel is the first, and it is the one that quietly decides whether small jobs across a wide area are worth taking. Callbacks and warranty visits are the second, and they belong on the original job rather than being treated as a separate small job that shows up as a loss on its own. The third is the owner's own hours, which get left out because they do not appear on a payroll.
Leaving out the owner's time makes every job look profitable and makes the business look like a hobby that generates cash. Assigning a rate to those hours, even an arbitrary one held constant across all jobs, is what makes the comparison meaningful, since the question being asked is which work is worth doing rather than whether the business survives. Overhead can be left out entirely at this stage, because it is roughly constant and does not change the ranking.
What to Do With Three Months of It
After a quarter there are enough rows to sort, and sorting is the whole analysis. Put them in order of gross margin as a percentage rather than as a dollar amount, since a large job with a thin margin and a small job with a fat one are easy to confuse when looking at totals. Then look at the top five and the bottom five and describe what each group has in common, which usually takes about ten minutes and is where the entire value of the exercise sits.
The comparison people find most useful is quoted hours against actual hours, expressed as a ratio. A trade that consistently runs twenty percent over on one type of work and exactly to estimate on another has learned something specific and actionable, and it is invisible in any monthly total. That ratio also tends to be stable, which means it can be applied to the next quote immediately rather than waiting for another year of data.
The Two Patterns It Usually Reveals
The first is that a category of work everybody assumed was fine is being subsidized by another category. Emergency call outs, small repairs, and anything involving a great deal of travel are the usual suspects, and they frequently show a negative margin once drive time and the owner's hours are attached. The second is a customer rather than a job type: one account whose work always takes longer than quoted, always involves a change, and always pays slowly.
Neither finding necessarily means dropping anything. A loss making category may be worth keeping as a route to larger work, and a difficult customer may be worth keeping if the volume matters. What changes is that the decision becomes deliberate. An owner who knows a job type runs at a thin margin can price it accordingly, schedule it into the gaps, or take it knowingly, and all three are better than discovering at the end of a busy year that the busiest quarter was the least profitable one.
What It Changes About Estimating
The immediate effect is on the next quote, which stops being a guess anchored to the last similar job and becomes an arithmetic problem with a known correction factor. The longer effect is on what gets pursued, because an operation that knows which work pays will start declining the kind that does not, and declining work is the single hardest thing for a small business to learn to do. An evening with a spreadsheet is a strange way to acquire that particular nerve, and it is the most reliable one available.