Pricing
by
Price decisions for ecommerce teams: read how your own demand responds to price on a product, then recommend raising it, lowering it, or leaving it alone.
A price change gets debated on instinct because the evidence needed to settle it is scattered across order history, promotion history and cost records that nobody joins. So the conversation runs on competitor screenshots and a feeling about what customers will tolerate, and it usually ends with no change, which is itself a decision nobody priced. Meanwhile the cost of goods has moved twice since launch. This skill reads how your own demand has responded to price on a product and recommends raising it, lowering it, or holding.
What It Does
- Estimates demand response from your own price and promotion history.
- Separates a price effect from a seasonal or campaign effect underneath it.
- Compares the current price against cost movements since it was set.
- Projects the contribution at each candidate price, not just the margin rate.
- States the confidence, so a thin history is visible rather than hidden.
- Recommends raise, lower or hold, with the contribution at stake.
Use Cases
- Settle a price debate with your own demand curve rather than a competitor's price.
- Catch the product still priced at launch after two cost increases.
- Find the product where lowering the price actually raises contribution.
- See what holding a price is costing before defaulting to it.
Example Session
Should we raise the price on the navy jacket?
Yes, to $198. Demand on this product has never responded to discounting.
- Price | Projected units, quarter | Margin per unit | Contribution | Change
- $162, two years ago | 1,180 | $58 | $68,440 | Historical
- $180, current | 1,090 | $71 | $77,390 | Baseline
- $198 | 1,020 | $89 | $90,780 | +$13,390
- $218 | 840 | $109 | $91,560 | +$780 vs $198
Going from $180 to $198 costs about 70 units and adds $13,390 of contribution. Going further to $218 adds almost nothing, because the volume loss finally catches up with the margin gain. That flat top is the signal to stop: $198 captures nearly all of the available upside and leaves the customer base intact.
Why This Beats Prompting It Yourself
A Competitor's Price vs Your Own Demand Response: The Hidden Risk
A generic prompt looks at what similar products cost elsewhere, which is the input it can find. That number reflects someone else's cost base and margin target. A product whose demand has never moved with price is underpriced regardless of what a competitor charges, and matching them gives away margin for a comparison your customers are not making.
When You Still Need The Point Where It Flattens
Most price analyses find the direction and stop. The useful part is where the curve turns over: $218 earns $780 more than $198 and costs you 180 customers a quarter. That is a bad trade even though the arithmetic is positive, and it is invisible unless you project the whole curve rather than one candidate price.
Known Limitations
Demand response is estimated from your own price and promotion history, so a product that has never varied in price returns a wide band the output states rather than a point estimate. It does not model competitor reaction, which matters most on directly substitutable products. It recommends the price and does not change it in your storefront.
Install It
- Download the ZIP.
It is free and there is no account to create. - Unzip it into your agent's skills folder.
Claude Code reads ~/.claude/skills/, which is hidden by default: the command in the folder block opens it. Other agents scan their own directory, so drop the same folder there instead. - Ask your agent to use it.
Restart the agent if it was already running, then it picks the skill up with no config.
skills folder
Copy
~/.claude/skills/pricing/
SKILL.md
references/demand-curve.md
decisions/
# one file per product, with the projection
# macOS: create the folder and open it in Finder
mkdir -p ~/.claude/skills && open ~/.claude/skills
# Windows: paste in the Explorer address bar
%USERPROFILE%\.claude\skills\Frequently asked questions
How is this different from a competitor price check?
A competitor check tells you where you sit against someone else's economics. This estimates how your own customers respond to price, projects contribution across a range, and says where the curve flattens. It frequently recommends a price no competitor is charging.
Which agents does the skill run in?
Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision.
What does it need connected to work?
Order history with prices paid, promotion history, and product cost over time. The more price variation in your history, the tighter the estimate, and the output always states which it is.
Is a 70 unit loss acceptable?
At $18 more margin per unit on a thousand units, yes: the arithmetic is $13,390 in your favor. The skill shows the units lost explicitly so the trade is visible rather than buried in a margin percentage.
Why not go to the highest contribution price?
Because $218 earns $780 more and costs 180 customers, many of whom would have bought again. The skill flags flat tops for exactly this reason: a rounding-error gain in contribution is rarely worth a real reduction in your customer base.
What is included with the skill?
The skill itself, the demand curve reference, the decision format, and the confidence thresholds. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your projections start shifting.
A price move needs a margin floor you trust.
Polar defines contribution margin once, across 45+ sources, so raise, lower or hold is decided on the same number your finance team closes on.
<script type="application/ld+json">{"@context":"https://schema.org","@type":"FAQPage","mainEntity":[{"@type":"Question","name":"How is this different from a competitor price check?","acceptedAnswer":{"@type":"Answer","text":"A competitor check tells you where you sit against someone else's economics. This estimates how your own customers respond to price, projects contribution across a range, and says where the curve flattens. It frequently recommends a price no competitor is charging."}},{"@type":"Question","name":"Which agents does the skill run in?","acceptedAnswer":{"@type":"Answer","text":"Any agent that supports the open SKILL.md format: Claude Code, Cursor, Codex CLI, GitHub Copilot, Gemini CLI, Manus, Grok and others load it unmodified. The format is portable, the location is not. Each agent scans its own skills directory, so you drop the same folder into whichever one yours uses. There is no config file to edit and no API key to provision."}},{"@type":"Question","name":"What does it need connected to work?","acceptedAnswer":{"@type":"Answer","text":"Order history with prices paid, promotion history, and product cost over time. The more price variation in your history, the tighter the estimate, and the output always states which it is."}},{"@type":"Question","name":"Is a 70 unit loss acceptable?","acceptedAnswer":{"@type":"Answer","text":"At $18 more margin per unit on a thousand units, yes: the arithmetic is $13,390 in your favor. The skill shows the units lost explicitly so the trade is visible rather than buried in a margin percentage."}},{"@type":"Question","name":"Why not go to the highest contribution price?","acceptedAnswer":{"@type":"Answer","text":"Because $218 earns $780 more and costs 180 customers, many of whom would have bought again. The skill flags flat tops for exactly this reason: a rounding-error gain in contribution is rarely worth a real reduction in your customer base."}},{"@type":"Question","name":"What is included with the skill?","acceptedAnswer":{"@type":"Answer","text":"The skill itself, the demand curve reference, the decision format, and the confidence thresholds. It is free and security scanned. We re-publish the ZIP when the model changes, so download it again if your projections start shifting."}}]}</script>