pricing-and-packaging
Skill2 starsChanged 11 days agoMITView on GitHub
Use whenever what to charge, or how to package it, is being decided or checked, in any language and at any length - "how should I price this", "we are undercharging", "set up our plans", "should we have a free tier", "what should the three tiers be", "when and how to raise prices", "are these prices right", price this service or subscription or software product. Covers what the price is tied to (the charging unit), what different buyers will pay and how to find out, tier and bundle design from stated customer segments and costs, anchoring and decoys used honestly, discounts and free offers, changing prices for existing customers, and checking a price page or plan table against these choices. Not for the wording of offers, guarantees and tier names on a page (offers-and-value-propositions), persuasion ethics and defaults (fair-persuasion), the layout of a comparison table (data-tables) or financial modelling and accounting.
In this repository25 other files it gives its agents.
- agent-interface-designSkill
- brand-identitySkill
- chart-designSkill
- color-paletteSkill
- data-tablesSkill
- design-critiqueSkill
- design-system-builderSkill
- form-designSkill
- generated-ui-restyleSkill
- interface-behaviorSkill
- journey-and-service-mappingSkill
- layout-structureSkill
- practice-and-assessmentSkill
- presentation-designSkill
- responsive-layoutSkill
- sales-conversationsSkill
- showing-workSkill
- spacing-and-groupingSkill
- teaching-interfacesSkill
- typeface-selectionSkill
- typesettingSkill
- ui-motionSkill
- ui-polish-passSkill
- user-research-basicsSkill
- visual-hierarchySkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 36
If the prices and plans already fit the buyers and the arithmetic holds, say so and change at most one thing.
- Distinctive: line 62
Every price traces to buyer value, the alternative or shown arithmetic, not to cost-plus or a rival alone.
- Distinctive: line 63
Every figure the user did not give is marked assumed.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Pricing and packaging
A price is a claim about what the buyer gets compared with their best alternative, including doing nothing. Set it from that value, treat cost as a floor, and split plans by what different buyers value, not by what was easy to build. The usual failure is a confident number with no basis: a margin on cost, "a bit under the leader", or three plans with rank names and no buyer behind them. Deliver a decision with its arithmetic, say what you assumed, and say what would change it.
Do not produce
- A price, conversion rate, willingness-to-pay figure, competitor price or customer count the user did not give, stated as fact. Mark it "assumed" or leave a placeholder.
- A price justified by cost plus margin, a percentage under a rival, or what the thing cost to build.
- One bare number: no range, no arithmetic, no line on what would change it.
- Tiers that differ only in quantity, carry rank names with no segment, or include one nobody is meant to buy.
- A cut, sale or discount with no diagnosed cause, no end and nothing asked in return.
- A price rise that does not say what existing customers pay and from when.
- A free tier with no cost per free user and no limit.
- Sections the user did not ask for, such as a models overview or a next-steps list.
When to use
- Setting a first price, choosing plans or tiers, picking what the price is tied to, deciding on a free tier, discount or annual plan.
- Raising prices, or reacting to weak sales, or to a request to halve the price.
- Checking a price page, plan table or price list against the choices below.
When not to use
- Words on the page (tier blurbs, guarantees, renewal and price-change notices): offers-and-value-propositions. Who it is for and the category: positioning-and-messaging. A whole pricing page: landing-page-copy. Table layout: data-tables. Interview craft: user-research-basics. All if installed; this skill supplies the decision and hands the words on.
- Whether a device is a pressure or a rigged choice: fair-persuasion. Whether a number or comparison is true: honest-claims. Taxes, legal terms, forecasts and accounting: say they need a specialist.
- A request for only the copy, with the structure supplied, needs no decisions from here.
- An explicit instruction, an existing price list, a contract, a design system or platform conventions (app-store tiers, payment-provider limits) win over every default below.
If the pricing already works
If the prices and plans already fit the buyers and the arithmetic holds, say so and change at most one thing. Do not add a tier, a free plan or a discount for taste.
Facts first
Look for: who pays and what they do now, what one more customer costs to serve (support, payment fees, any channel share, free users), current prices and how many buyers sit on each, the goal (profit, share, cash), and any evidence of willingness to pay (calls, lost deals, paid pilots). Whatever is missing, write one line "Assumed: ..." and decide anyway; name the one or two facts that would move the answer most. Ask a question only if the chat is interactive and one answer would flip the decision. Never fill a gap with an invented figure.
Decisions, in order
A small task starts at the decision it touches.
- Buyer and alternative. Name the buyer and what they do or pay today. Default: price against that alternative and the value gap, stated as a range in the buyer's terms (hours saved, money earned, risk avoided). Change to the rival's list only when buyers truly compare the two side by side. Cost sets the floor, never the price; leave out what it cost to build. See finding what buyers will pay.
- Evidence. Default: conversations about the alternative and the value, then range questions with a "why" after each, then real money (a paid pilot, a deposit, a real quote). Read the spread of answers, not the average; two clusters mean two offers. Change to quick range questions only when the stakes are small. A stated "yes" is weak; payment is strong.
- Level. Default: one goal, a profit table at three to five prices using supplied or assumed unit cost, and a range around the peak. Revenue and profit peak at different prices. Change to a deliberately low entry only with a written reason and a plan to raise. For a proposed cut, match or cost-driven rise, show the arithmetic in price moves. See level, discounts and changes.
- What the price is tied to. Default: the unit that rises with the value the buyer feels, that they can predict, that you can measure, and that does not shrink as customers or hardware grow. Check seats for tools where one person does the work and many only look. Change to a flat price when usage is similar across buyers. Tests and examples: tiers, units and free offers.
- Segments and tiers. Default: three or four buyer groups at most, each tier built for one group, each higher tier containing everything below plus something that group pays for; label each tier with its group. Three tiers when buyers compare side by side; two when only two groups differ in value; four when a group buys through a different route. Change to one price when no group differs in value or cost to serve. A cheap tier must truly lack what the next adds.
- Free and trial. Default: no free tier unless a free user costs close to nothing and the limit stops the free plan covering what paid buyers value. Change to a time-limited trial, a free narrow job, or a paid pilot when the cost per free user is real. Trials fail where one use solves the whole problem.
- Discounts. Default: no blanket discount. Trade each one for something (a longer term, volume, prepayment, a case study), with a floor and a named approver set in advance. Compute an annual-plan discount from how long monthly buyers actually stay, and from who will choose annual. Change to a dated launch price only if the later price is already decided. For a negotiated account, a volume ask, an exception, approval limits, a tender or an audit of what accounts pay, see discounts and negotiated prices.
- Anchors and thresholds. Default: every tier is something a real buyer would sensibly choose; a top tier may anchor only if it is a real product with buyers. Find, do not guess, the amounts where a buyer needs sign-off or a round number is a barrier, and price below one only if the product still earns. Route the audit to fair-persuasion.
- Changing prices. Default: raise in steps small enough that the reason to stay still plainly outweighs them, with a visible improvement; keep current customers at their price for a stated period; define what you will watch and what would reverse it. Prefer a new tier when only some buyers value the new thing. For weak sales, find the cause before any cut. See level, discounts and changes.
Output
Give the recommendation first: price or plans, with the range and the arithmetic in a few lines or a small table. Then one line "Assumed: ..." and one line "Would change if: ...". Stop there unless more was asked. For a review, quote each weak price or plan, say why in a clause, and give the fix. Honor a stated count or length. A request for plans gets a tier-by-group line, then hands blurbs to offers-and-value-propositions.
Quick checks
- Every price traces to buyer value, the alternative or shown arithmetic, not to cost-plus or a rival alone.
- Every figure the user did not give is marked assumed.
- The answer holds a range or a profit calculation, and says what would change it.
- Each tier names its group and one feature that a lower tier lacks.
- Cost used excludes build cost and includes support, fees, channel share and free-user cost.
- A discount or cut gives its cause, what comes back and when it ends.
- A rise says what existing customers pay and from when.
- Nothing appears that the user did not ask for.
References
- finding-willingness-to-pay.md: load when the user has no evidence of what buyers pay, wants research, or has survey answers to read.
- tiers-and-units.md: load for plan design, feature placement, bundles, the charging unit, a free tier, or a plan mix that looks wrong.
- level-and-changes.md: load for a profit table, back-solving volume, discounts and annual plans, weak sales, raising prices, or reviewing a price page.
- price-moves.md: load when a cut, discount, promotion, price match or cost-driven rise is proposed or questioned: the extra sales a cut needs, the sales a rise can lose, which costs count, a rival's cut.
- discounts-and-negotiated-prices.md: load when a customer asks for a better price, a volume deal, a discount on a rise or compensation; when discount rules, approval limits or exceptions are set; for tenders; or to audit what accounts actually pay.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

