Tracking time on commissions: a simple method for better quotes
Updated · 7 min read
Most makers know roughly how long a piece takes. Roughly is the problem. A quote built on a guess that is a fifth too low gives away a fifth of your labour on every commission, and you will not see it until the year's accounts.
This guide is for workshops of one to five people working to order. It covers why tracking time pays off, simple methods that survive a busy week, how to track by stage, how to compare your estimate with what actually happened, and what research says about why our estimates run short.
The examples come from furniture, jewellery, ceramics, leather and instrument making. The worked example is illustrative, with invented figures.
Key takeaways
- Track hands-on time per stage for every commission, and record waiting time (drying, curing, firing) separately.
- A paper sheet, a spreadsheet or a timer app all work; consistency matters more than the tool.
- Compare estimate and actual per stage when each piece is finished. The stage that overruns is where your quotes are weak.
- Buehler, Griffin and Ross found that students predicted 33.9 days for their thesis and took 55.5; fewer than a third finished by their own best estimate.
- In the same research, the optimistic bias disappeared when people were asked to connect relevant past experiences to their prediction. Your time records are those past experiences, written down.
Why track time at all
Pricing accuracy. In made-to-order work, your hours are a large part of what the client pays for. If you charge by the hour or build a fixed price from hours, every hour you do not count is an hour you work for free. See how to price custom furniture for how hours feed into a price.
Better quotes. When a client asks for a walnut sideboard or a pair of gold earrings with a particular setting, you can look up what similar pieces actually took, not what you remember.
Honest lead times. Hours per stage, plus waiting time, tell you when a piece will realistically be ready. That makes delivery dates easier to keep.
Spotting problems. If finishing always runs over, you may need a better spray set-up, a different finish or a higher price for that stage. You cannot see that without data.
What research says about estimating
The term planning fallacy comes from Daniel Kahneman and Amos Tversky (1979). It describes holding a confident belief that your own project will go to plan, while knowing that most similar projects run late.
Roger Buehler, Dale Griffin and Michael Ross tested this in a 1994 paper in the Journal of Personality and Social Psychology. In their first study, honours thesis students at the University of Waterloo were asked to predict when they would submit. On average they predicted 33.9 days and took 55.5 days. Only 29.7% finished within their own best estimate. Even their worst-case predictions, made on the assumption that everything went as poorly as it possibly could, averaged 48.6 days, still a week short.
The paper's think-aloud studies found that people focused mainly on future scenarios, how they planned to do the task, rather than on how long similar tasks had taken before. They also tended to explain away past overruns as one-off events. In Study 4, the optimistic bias was eliminated for participants instructed to connect relevant past experiences with their predictions.
Students writing theses are not makers building cabinets, so take the exact figures with care. The pattern is the useful part: estimates built from a mental plan run short, and estimates anchored in records of past jobs do better.
Simple ways to track time
Pick the method you will still use in a busy week.
| Method | Good for | Watch out for |
|---|---|---|
| Paper sheet on the wall | Workshops with dirty hands and no phone at the bench | Transcribe weekly or it gets lost |
| Spreadsheet | Makers who already keep orders in a sheet | Typing times in from memory at the end of the day |
| Timer app | People who forget to write things down | Free plans limit features; check before relying on reports |
| Phone photos | A rough check of when stages started and ended | Gives dates, not hands-on hours |
Timer apps such as Toggl Track and Clockify have free plans. Check the limits on their pricing pages: Toggl Track's free plan has personal reports only and billable rates start on a paid plan, and Clockify lists budgets and estimates as unavailable on its free plan.
Whatever you use, record four things per session: commission ID, stage, start time, end time. Add a note when something unusual happens, such as a recut or a failed firing.
Track by stage, not just by piece
A total of 62 hours tells you the price was wrong. Hours per stage tell you where. Use the same stage list for every piece of a given type so the numbers compare.
| Craft | Example stages |
|---|---|
| Furniture | Design and client time, timber selection and milling, joinery, dry fit and glue-up, finishing, delivery |
| Jewellery | Design and client time, model or wax, casting or fabrication, setting, polishing, hallmarking and dispatch |
| Ceramics | Design and client time, throwing or building, trimming, bisque, glazing, glaze firing, checking and packing |
| Leather | Design and client time, pattern, cutting, skiving and preparation, stitching, edges and finishing |
| Instruments | Design and client time, wood preparation, body or plates, neck, assembly, finishing, set-up |
Two separate lines are worth adding to every list:
- Design and client time. Sketches, emails, calls, revisions. This is real work and it varies hugely between clients.
- Waiting time. Oil curing between coats, glaze firings, adhesive cure times. Record it in days, not in hours, and keep it out of labour cost.
Compare estimate and actual
When you quote, write your estimate per stage. When the piece is finished, put the actual hours next to it.
Worked example (illustrative)
The figures below are invented to show the method. A maker quoted an oak bedside cabinet with one drawer.
| Stage | Estimate (h) | Actual (h) | Difference |
|---|---|---|---|
| Design and client time | 2 | 5 | +3 |
| Timber selection and milling | 4 | 4.5 | +0.5 |
| Joinery | 8 | 9 | +1 |
| Drawer | 4 | 6.5 | +2.5 |
| Dry fit and glue-up | 3 | 3 | 0 |
| Finishing | 3 | 5 | +2 |
| Delivery | 1 | 1 | 0 |
| Total | 25 | 34 | +9 |
The total overran by 36%. Three stages explain nearly all of it: client time, the drawer and finishing. Each points to a different fix:
- Client time. The client changed the handle twice. The next quote could include one round of revisions and price further changes.
- Drawer. If drawers overrun on every piece, the estimate for a drawer should be closer to the actual figure, not the hoped-for one.
- Finishing. If this repeats, look at the process or the price of that stage.
Turn records into a quote
After five or six similar pieces, take the actual hours per stage, not the estimates. Use the middle value (the median) for each stage, so one disaster or one lucky job does not distort it. Add the stages up. That is your starting estimate for the next similar commission. Then adjust for what is genuinely different: a larger size, an unfamiliar timber, a stone you have not set before.
This is the "outside view" in practice. Instead of asking how you plan to make this piece, you ask how long pieces like it have taken you.
Common pitfalls
- Recording at the end of the week. Memory shrinks hours. Write times down the same day.
- Leaving out small sessions. Ten minutes sharpening, sanding a test piece, packing. They add up and belong to a stage.
- Counting waiting time as labour. It inflates labour cost and hides the real cause of long lead times.
- Changing stage names. Keep the list stable for a type of piece, or comparisons stop working.
- Tracking but never reviewing. Set a time, for example the day you send the final invoice, to compare estimate and actual.
A one-page routine
- Before you quote, look up actual hours per stage for similar past pieces.
- Quote with an estimate per stage and keep it with the commission.
- Record every working session: ID, stage, start, end, short note.
- Record waiting time separately, in days.
- When the piece is finished, fill in actual hours next to the estimate.
- Once a quarter, update your typical hours per stage for each type of piece.
If you keep orders in a spreadsheet, add the time log as a tab linked by commission ID. See spreadsheets vs workshop software for how to keep it reliable.
Sources
- Exploring the "Planning Fallacy": Why People Underestimate Their Task Completion Times, R. Buehler, D. Griffin and M. Ross, Journal of Personality and Social Psychology 67(3), 366–381, 1994: Study 1 figures (33.9 and 55.5 days, 29.7%, worst case 48.6 days), think-aloud findings, Study 4 result, attribution of the term to Kahneman and Tversky (1979) (accessed 3 November 2026)
- Toggl Track pricing, Toggl: free plan with personal reports, billable rates from Starter (accessed 3 November 2026)
- Clockify pricing, Clockify: budgets and estimates not on the free plan (accessed 3 November 2026)
Questions this guide gets asked
Do I need an app to track time?
No. A paper sheet on the wall with the commission ID, the stage and start and stop times works. Apps help if you forget to write things down or want reports, and several have free plans with limits. The method matters more than the tool: record every session, by stage, the same day.
Should I count drying, curing and firing time?
Record it separately. Hands-on hours drive your labour cost; waiting time drives your lead time. A glaze firing or an oil finish that needs days between coats costs little labour but a lot of calendar, and both matter when you quote a delivery date.
What about time spent on emails, sketches and client meetings?
Track it as its own stage, such as design and client time. It is often larger than makers expect, especially for commissions with many changes. If you leave it out, your quotes will be too low on exactly the most demanding clients.
How many commissions do I need before the data is useful?
You will learn something from the first one. For quoting, aim for five or six pieces of a similar type, so a single unusual job does not distort the picture. Compare like with like: a dining table with other tables, a ring with other rings of similar complexity.
Why are my estimates always too low?
This is a well-documented bias called the planning fallacy. In a 1994 study, students predicted they would finish their thesis in 33.9 days on average and took 55.5. People focus on the plan for this job and give little weight to how long similar jobs took. Your own records are the fix.