agentleFS
MadalaVijayMadalaVijay/marketing-skills-deep/skills/pricing-decisions/SKILL.md

pricing-decisions

Skill0 starsChanged 2 months agoMITView on GitHub

Reason about price level, tier structure and discounting without guessing. Covers what a price test can and cannot tell you, willingness-to-pay research, and why discounting usually costs more than it earns. Use when setting or changing price, designing tiers, or evaluating a discount.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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
  1. Distinctive:

    Prefer, in order: longer commitment, bundled value, then price cuts last.

    line 61

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Pricing decisions

Price is the fastest lever on profit and the one most often set by feel. It is also the hardest thing to A/B test honestly.

Why price tests are different

You usually cannot run a clean price test:

  • Charging different customers different prices for the same thing creates fairness and, in some markets, legal exposure
  • The effect is slow. Price changes churn and expansion for months. A two-week test measures signup rate, which is not the thing you care about
  • Signup rate is the wrong metric anyway. A higher price with fewer signups can be strictly better. Judge on contribution margin per visitor, not conversion

Where a test is possible, run it on new cohorts only, hold existing customers harmless, and commit up front to measuring for a full retention cycle.

When you cannot test, do this instead

  • Van Westendorp four questions (too cheap / cheap / expensive / too expensive) gives an acceptable range from a modest sample. Directional, not precise
  • Compare against the category anchor, not against cost. Buyers price you against the nearest familiar alternative. If the category trains people to expect a low entry price, being several times it needs a visible reason
  • Watch what people already pay for the adjacent thing they use today
  • Sell the duration, not just the tier, where the purchase is recurring. Longer commitments often price better than deeper discounts

Tier design

  • Three tiers works because it gives a reference point, not because three is magic. The middle tier is chosen most often when the top exists
  • Differentiate on a dimension the buyer can predict for themselves — seats, volume, frequency. Feature-gating on things buyers cannot forecast produces wrong-tier purchases and churn
  • Do not let the cheapest tier be good enough for the core use case, or the ladder never lifts
  • Check the payment ceiling. Many markets have mandate or auto-debit limits above which recurring payment silently fails or requires re-authentication. A price just over that ceiling can look fine and quietly break renewals

Discounting

Model it before you approve it. A 20% discount on a 40% contribution margin removes half the margin, so it needs to double volume just to break even. State that arithmetic explicitly whenever a discount is proposed.

Worse, discounts train the buyer. A brand that discounts predictably teaches people to wait, which shifts full-price demand into the discount window and looks like the discount worked.

Prefer, in order: longer commitment, bundled value, then price cuts last.

When to refuse

  • Contribution margin unknown. Ask first
  • Asked to project revenue from a price change with no retention data
  • Asked to justify a price already decided. Say what the evidence supports

Output

The recommendation, the evidence class behind it (test, research, or judgement), what it would take to be wrong, and the one metric to watch after the change. For a discount: the break-even volume lift, stated as arithmetic.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.