Cost a single terracotta planter from clay to sale price
Small pottery units often price by habit, so they cannot tell whether a planter sold to a nursery or a hotel actually covers clay, firing fuel, labour and breakage.
A spreadsheet that costs one terracotta planter step by step with every assumption in a labelled cell, shows the cost at two batch sizes, and a short note on which input moves the cost most.
This is a financial model project. Build a small model of costs, revenue or cash and test what changes the answer.
The approach: Write every assumption in its own labelled cell → Build the calculation from the assumptions → Change one assumption at a time → Say which assumption the answer depends on most.
Think first
You are working out what one planter really costs to make, so the potter can see whether each sale earns anything. The model must make every assumption visible so the potter can change it and trust the answer.
- Which product exactly: what size and finish of planter, made how many at a time?
- What goes into it: clay, water, colour or slip, firing fuel, packing, transport to the buyer?
- How much time does each step take the potter or helpers, and what is that time worth?
- How many pieces crack or break in drying and firing out of a typical batch?
A sheet with an assumptions block, a cost build-up per batch and per good piece, two batch sizes side by side, and a two-line note naming the input that changes the cost most.
Leaving out the potter's own time and breakage. A planter that looks profitable on materials alone may lose money once those are counted.
Gather
- Notes from a consented interview with a potter on steps, inputs and time, with no name recorded
- What a batch looks like: how many planters fit a firing, and how often they fire
- Typical breakage in drying and firing, in the potter's own words
- If no interview is possible, a set of made-up practice numbers, labelled as made up in the sheet
AI must not supply clay or fuel prices as if they were real. Every number comes from the potter or is visibly marked as made-up practice data.
Build with AI
- 3.1 Challenge the cost list before buildingCLAUDE CHAT
Why this tool: Claude can spot cost items a first model usually forgets.
Describe the process and your list of inputs and ask what is missing. Add only items that apply to this potter.
- 3.2 Build the batch cost modelCLAUDE FOR EXCEL
Why this tool: Claude in Excel writes formulas that point at labelled cells and explains them.
Create an assumptions block (batch size, clay per piece, fuel per firing, hours per step, hourly value of time, breakage) and a calculation block that builds cost per batch and per good piece. Ask Claude to explain each formula.
- 3.3 Run the model through a Skill for two batch sizesCLAUDE SKILLS
Why this tool: A saved Skill re-runs the same cost check whenever the potter changes an input.
Save a Skill that reads the sheet, recalculates cost per good piece at the current and a larger batch size, and lists which single assumption moves the result most when nudged up and down.
Before you let it act: Before running it, confirm the Skill only reads and reports and does not overwrite any assumption cell; compare its output for the current batch with your own hand calculation.
- 3.4 Write the potter's noteCLAUDE CHAT
Why this tool: Claude can help say the finding in two plain sentences.
Give Claude the results and ask for a short note for the potter. Rewrite it in your own words.
Check the AI's work
- No number is typed inside a formula; every input sits in a labelled assumption cell
- Recompute cost per good piece by hand for one batch size and match the sheet
- Breakage reduces the number of good pieces rather than being ignored
- Every made-up number is marked as practice data
- No number typed inside a formula: every input is a labelled cell
- Recompute the headline result by hand
AI often spreads the firing fuel over every planter loaded, including ones that break, which understates the cost of each good piece.
Finish and show
- Assumptions block with sources or 'made up' labels
- Cost per batch and per good planter
- Two batch sizes compared
- Two-line note on the input that matters most
One screen of the sheet with the assumptions on the left, results on the right, and the note at the top.
Show the cost per good piece and explain how counting breakage and time changed the picture.
Which tool fits
- Claude chat: Claude can spot cost items a first model usually forgets.Free plan works; Pro gives longer conversations.
- Claude for Excel: Claude in Excel writes formulas that point at labelled cells and explains them.Paid plans; needs Excel.
- Claude Skills: A saved Skill re-runs the same cost check whenever the potter changes an input.Pro or above, with code execution switched on.
- Google Sheets (free alternative): a free alternative for the same modelFree with a Google account.
Add a second product, such as a water pot, that shares the same firing, and decide how to split the fuel cost between the two.
AI helps you think, check and polish. The work, and the judgement, stay yours.
This is a free sample. Full Access opens every project in full and lets you submit your finished work.
See Full Access →