Toolkit · Free
Pricing & Optimization Toolkit
What should we charge, and what should we sell. These get argued in separate meetings and they are the same question, because a price you cannot supply is not a price and a product you cannot sell at a profit is not capacity well spent. Hand it a portfolio, get back a ranked set of decisions where every recommendation carries what it is worth and what has to be true.
What It Does
It decides, and it shows its work
A diagnostic explains what already happened. This one chooses what to do next, which means every recommendation has to survive being checked. Four layers stack, and each needs a little more from your data than the last. You are told which ones are live rather than being shown a confident answer built on a column you never supplied.
1
True contributionNeeds price, variable cost and volume
What each line actually earns, and for customers what survives the cost of serving them: order handling, delivery, returns, support and the financing of payment terms.
2
PositionNeeds a customer value or a competitor price
Where each price sits between the floor your cost sets and the ceiling your customer sets. Cost-plus only ever consults the first, which is how money gets left on the table at one end and priced out of the market at the other.
3
The move gateNeeds a move you are considering
How much volume a price rise can afford to lose, or a cut has to win, before it stops paying. This is arithmetic, so it settles the argument that opinion cannot.
4
The mixNeeds a binding constraint and its consumption per unit
When something runs out, the best product is the one that earns most per unit of the scarce thing, not per unit sold. It ranks on that, fills capacity against real demand ceilings, and prices the constraint itself.
Then it turns all of it into a ranked decision set, and the impacts reconcile: base contribution plus everything you accept equals the contribution you end up with, to the dollar. If it does not tie, the decision set is wrong, and the toolkit is built to say so rather than to plug the gap.
The Package
Two files, no email wall
Install the skill once and it is available in every Claude conversation. The template is optional, and worth opening first: its sample rows are the worked examples from the articles behind this toolkit, so a run against the untouched file reproduces numbers that already exist in public.
Free download: the Pricing & Optimization Toolkit
Install once, then reuse it on every price review, capacity argument and customer rationalisation.
The Skill
pricing-optimization-toolkit-skill.zip: the pricing and optimization engines, the dashboard specification, and a validator that checks the decisions reconcile before the file reaches you.
Download Skill ↓
Input Template
Pricing_Optimization_Input_Template.xlsx: products, customers and the constraint, with a column guide and a sample portfolio whose expected answers are printed alongside it.
Download Template ↓
Requires a Claude Pro, Max or Team account. For the strongest reasoning about which decisions to put first, run it on Claude Opus. The arithmetic is deterministic either way, so the model choice changes the judgment, not the break-evens. Something not working, or a question about your own portfolio? Get in touch.
Setup
Running it the first time
About 2 minutes to install. Then as long as it takes you to find a price, a cost and a volume for each line.
1
Install the skill
Download the zip and add it to Claude as a skill. It stays available in every conversation afterwards, so this happens once.
2
List your lines
One row per thing you would actually reprice or stop selling. Price, variable cost and volume are the only columns you need. Make sure the cost is genuinely variable: if it has overhead absorbed into it, every conclusion will be wrong in the expensive direction, and it is better to say so than to find out later.
3
Add whatever else you have
A value ceiling, a competitor price, a constraint and its consumption, demand ceilings, a move you are considering. Each one turns on another layer. A column you leave out produces an honest not asked, never a silent zero.
4
Upload it and ask
Say what you want in your own words. Can we raise prices? or we are at capacity, what do we drop? is enough. You get back a single HTML file that opens by double-click and works offline.
The Output
What comes back
One interactive file, built to be argued with in a meeting rather than admired afterwards.
The decision setRanked actions, each with its dollar impact, its break-even stated in words, and a confidence label. Accept and reject them and the totals move with you.
A bridge that tiesA waterfall from today's contribution to the decided one, a step per accepted action. It is the panel that proves the plan is arithmetic and not a wish list.
The contribution mapEvery line plotted by what it earns per unit against what it earns per unit of the constraint. The lines far off the diagonal are the finding.
Price anchorsEach price on a track from its cost floor to its value ceiling, with the competitor and the cost-plus price marked, sorted by the size of the gap.
The move gateBreak-even response against expected response, coloured by headroom rather than by size, so a large but fragile move looks fragile.
Mix allocationCurrent against recommended capacity fill, unused capacity shown explicitly, and the shadow price of one more unit of the constraint.
Cost to serveWhere customer data allows it: accounts ranked by what they contribute after service costs, with the price lever and the service lever priced side by side.
A stress testHow many decisions and how many dollars survive as volume responses get worse. This is what separates a robust plan from a hopeful one.
Honesty
What it will not do
The limits are part of the deliverable, and the toolkit states them on the dashboard rather than in a footnote.
Where the arithmetic stops
- It will not tell you what the market will pay. A value ceiling is an input, not an output. Without one it says the ceiling is unknown, rather than dressing up cost-plus as a target.
- It will not forecast a price move. The break-even is arithmetic and the volume response is your assumption. Both appear, and they are never allowed to look like the same kind of number.
- It will not solve a multi-constraint problem with a ranking. The ranking is optimal for one binding constraint. When several bind at once it says so and hands off, which is what the Solver model is for.
- It will not book a fixed-cost saving nobody agreed to. Dropping a line removes its contribution immediately and its cost only if the capacity behind it genuinely goes. It models the first and flags the second.
- It will not price strategy. A line that anchors the brand or keeps a plant economic can be worth its capacity while losing on the numbers. It shows what that choice costs, then leaves the choice to a person.
The Thinking Behind It
Why it is built this way
Four articles set out the frameworks, and four calculators let you check any single piece of it in your browser before you commit a portfolio to the toolkit.
Cost-Plus Pricing Is a Confession
Article
The three anchors, and why a price set from cost alone never asked the question that decides it.
Elasticity Without Econometrics
Article
The break-even gate: how much volume a move can afford to lose, without needing a demand model.
Why More Volume Doesn't Always Mean More Profit
Article
Contribution per constraint unit, and the factory that made more revenue and less money.
The Revenue You Should Want to Lose
Article
Cost to serve, and the account at 3.5 times the revenue returning an eighth of the contribution.
Check any single piece in your browser first, no download: the value-based price calculator, the break-even calculator for either direction of a move, the constraint mix calculator, and the cost-to-serve calculator. The toolkit is what runs all 4 across a whole portfolio at once, and then decides.