create-prd

what it does

Create a lean, decision-ready Product Requirements Document with an 8-section template: summary, contacts, background, objective, market segments, value propositions (6-part JTBD template), solution, and release. Lean by default (about one to two pages), with a fuller spec on request. Use when writing a PRD, documenting product requirements, preparing a feature spec, or reviewing an existing PRD.

by Pawel Huryn · Pawel Huryn, MIT · version 1 · license MIT

Verified Before and after: 2 models 1 try per case View source Download .zip

Compare the answers

Answers from Claude Sonnet 5.5 and GPT-6.1 Sol · try 1 of 1

Same request, same model, once with the skill and once without. Read both and judge for yourself.

“Turn these kickoff notes into a PRD.”

Source material given with the request

Kickoff, Oct 14. Dana (PM), Luis (eng lead), Mei (design). - Groomers lose money on no-shows. No-show rate across our 340 grooming salons averaged 18% in Q3. - Idea: SMS and push reminders 48 hours and 3 hours before the appointment, with one-tap confirm or cancel. Cancelled slots get offered to a waitlist. - Luis: deposits would need a payment integration, about 2 extra months. Decided: no deposits in v1. - Decided: iOS and web first. Android later (only 9% of salon clients use Android). - Goal: bring no-shows from 18% to 12% within 3 months of launch. - Mei worried clients will get too many texts. Agreed: max 2 reminders per booking. - Unknown: whether salons will pay extra for SMS (costs us about $0.01 per text).

with the skill

I couldn't save a file in this chat, so the PRD is below. Save it as `PRD-grooming-reminders.md`. --- # PRD: Appointment Reminders for Grooming Salons ## 1. Summary We will send clients SMS and push reminders before their grooming appointment. They can confirm or cancel with one tap, and cancelled slots go to a waitlist. This is for our 340 grooming salons, which lose money when clients don't show up. The Q3 no-show rate averaged 18%. ## 2. Contacts - Dana: Product Manager - Luis: Engineering Lead - Mei: Design ## 3. Background Across our 340 grooming salons, 18% of booked appointments were no-shows in Q3 (average). An empty slot is lost income that the groomer can't get back. Reminders are a simple first step. A client who can cancel in one tap frees the slot early, and the waitlist can fill it. ## 4. Objective

Show the rest (84 more lines)

Help salons keep their calendars full by cutting no-shows. Clients get an easy way to confirm or cancel. **Key Result:** Bring the no-show rate from 18% to 12% within 3 months of launch. That is about one-third fewer no-shows. **Guardrail:** No client gets more than 2 reminders per booking. ## 5. Market Segments 1. **Salon owners and groomers.** They need to fill every slot, and every no-show costs them money. 2. **Salon clients (pet owners).** They want to keep or change an appointment without calling the salon. **Constraints:** v1 covers iOS and web. Android clients are only 9% of salon clients, so Android comes later. ## 6. Value Propositions ### Segment 1: Salon owners and groomers - **Who:** Owners and groomers at our 340 salons. - **Why:** Get a full day of paid appointments. No-shows get in the way. - **What before:** About 18% of appointments are no-shows. The slot sits empty and the income is lost. - **How:** Automatic reminders with one-tap confirm or cancel. Cancelled slots are offered to a waitlist. - **What after:** Fewer no-shows, and cancellations arrive early enough to refill the slot. - **Alternatives:** Calling clients by hand, or taking no action and absorbing the loss. We are automatic and built into the booking. ### Segment 2: Salon clients - **Who:** Pet owners who book grooming appointments. - **Why:** Keep their appointment, or cancel it without hassle, when life gets busy. - **What before:** They have to remember the date. Cancelling may mean calling the salon, so some people just don't show up. - **How:** A reminder 48 hours and 3 hours before, with one tap to confirm or cancel. - **What after:** They don't forget, and they can cancel in seconds with no awkward call. - **Alternatives:** Their own calendar or memory, or a phone call to the salon. **Value proposition statement:** Salons get fewer empty slots, and clients get a simple reminder and a one-tap way to confirm or cancel. ## 7. Solution ### 7.1 Key flow 1. A client books an appointment. 2. They get a reminder 48 hours before, and another 3 hours before. 3. Each reminder has a one-tap Confirm or Cancel. 4. If they cancel, the slot is offered to clients on the waitlist. Mei will design the reminder and the confirm/cancel screens. ### 7.2 Key Features (v1) - **SMS and push reminders** at 48 hours and 3 hours before the appointment. - **One-tap confirm or cancel** from the reminder. - **Waitlist:** a cancelled slot is offered to waitlisted clients. - **Reminder cap:** a hard limit of 2 reminders per booking. Mei raised the concern that clients will get too many texts. - **Platforms:** iOS and web. ### 7.4 Assumptions and open questions **Assumptions (not yet proven):** - Reminders with easy cancelling will cut no-shows from 18% to 12%. This is the main bet. - Early cancellations will be refilled from the waitlist often enough to matter. **Open questions:** - **Will salons pay extra for SMS?** Each text costs us about $0.01. With 2 reminders per booking, that is about $0.02 per booking, before any other texts. We need to decide whether we absorb this cost or charge for it. - **Does the 2-reminder cap cover the waitlist offer message?** The notes don't say. - **Do the 2 reminders mean 2 total, or 2 per channel (SMS and push)?** The notes don't say. We should settle this before building. - **What do Android clients get in v1?** We assume they still get SMS, since SMS doesn't depend on the app. This needs confirming. - **How long will v1 take?** The kickoff didn't estimate it. ## 8. Release **In v1:** - SMS and push reminders at 48h and 3h - One-tap confirm or cancel - Waitlist for cancelled slots - Max 2 reminders per booking - iOS and web **Later:** - Android app support - Deposits (needs a payment integration, about 2 extra months) **Not doing in v1:** - Deposits - Android-specific features - More than 2 reminders per booking **Timeline:** Not yet estimated. We will measure the no-show rate for 3 months after launch. --- **Left out on purpose:** competitive analysis, detailed UX flows, pricing options for SMS, and a rollout plan. I can expand any of these. I also didn't add Key Results beyond the one you set, because the notes didn't give targets for waitlist fill rate or client response rate.

without

# PRD: Appointment Reminders & Waitlist Backfill (v1) **Source:** Kickoff, Oct 14 **Participants:** Dana (PM), Luis (Eng Lead), Mei (Design) **Status:** Draft --- ## 1. Problem Groomers lose revenue when clients don't show up. The no-show rate across our 340 grooming salons averaged **18% in Q3**. Empty slots are rarely refilled because salons have no easy way to learn of a cancellation early or to offer the slot to someone else. ## 2. Goal and Success Metric | | | |---|---| | **Primary goal** | Reduce no-show rate from **18% to 12%** within **3 months of launch** | | **Baseline** | 18% (Q3 average, all 340 salons) | Suggested supporting metrics (not yet agreed): confirmation rate, cancellation rate, % of cancelled slots refilled from the waitlist, and opt-out rate for reminders.

Show the rest (67 more lines)

## 3. Proposed Solution ### 3.1 Reminders - Send reminders **48 hours** and **3 hours** before each appointment. - Channels: **SMS and push notifications**. - Each reminder includes **one-tap confirm or cancel**. - **Cap of 2 reminders per booking**, to avoid over-messaging (Mei's concern). ### 3.2 Waitlist backfill - When a client cancels, the freed slot is **offered to a waitlist**. ## 4. Scope ### In scope for v1 - 48h and 3h reminders via SMS and push - One-tap confirm/cancel - Waitlist offer for cancelled slots - **iOS and web** clients ### Out of scope for v1 - **Deposits / prepayment.** This needs a payment integration, which Luis estimates at about 2 extra months. Revisit after v1. - **Android.** Deferred because only 9% of salon clients use Android. Android clients can still receive SMS, but this needs confirming (see open questions). ## 5. Requirements | # | Requirement | Notes | |---|---|---| | R1 | Send a reminder 48h before the appointment | | | R2 | Send a reminder 3h before the appointment | | | R3 | Never send more than 2 reminders per booking | Hard cap | | R4 | Reminder includes one-tap confirm and cancel actions | | | R5 | Cancelling frees the slot and triggers a waitlist offer | Rules TBD | | R6 | Supported on iOS and web | Android is post-v1 | ## 6. Risks and Constraints - **Message fatigue:** Mitigated by the 2-reminder cap. - **SMS cost:** About $0.01 per text. At the cap, SMS costs up to roughly **$0.02 per booking**. Whether salons will pay extra for SMS is **unknown**, and this affects pricing and packaging. - **Scope trade-off:** Without deposits, there is no financial commitment from clients. The 12% target relies on reminders and backfill alone. ## 7. Open Questions 1. **Pricing:** Will salons pay extra for SMS? Should it be included, an add-on, or usage-based? *(Owner: Dana)* 2. **Cap definition:** Does "2 reminders per booking" mean 2 reminder moments (48h and 3h), each sent by SMS or push, or 2 total messages across channels? Should a client get both SMS and push for the same reminder? 3. **Channel logic:** Is SMS sent to everyone and push only to app users? Is there a fallback if push is disabled? 4. **Waitlist mechanics:** How do clients join? How are they ranked (first come, first served)? How long do they have to accept an offer? Is the offer sent by SMS or push, and does it count toward the 2-message cap? 5. **Late cancellations:** Is there any treatment of a cancel within the 3h window, such as a flag for the salon? 6. **Unconfirmed clients:** Does no response to a reminder trigger anything, such as a salon alert or an automatic release of the slot? 7. **Android clients:** Do they receive SMS reminders in v1 even without an app? 8. **Opt-out and compliance:** SMS consent and opt-out handling requirements. 9. **Timeline and launch plan:** No dates were set at kickoff. Is a pilot with a subset of salons planned? 10. **Measurement:** How will we attribute no-show reduction to this feature, for example with a holdout group, given seasonality? ## 8. Decisions Log | Decision | Rationale | Source | |---|---|---| | No deposits in v1 | Needs a payment integration, about 2 extra months | Luis | | iOS and web first, Android later | Android is only 9% of salon clients | Team | | Max 2 reminders per booking | Avoid over-texting clients | Mei / Team | | Target 18% → 12% in 3 months | Goal-setting | Team | ## 9. Owners - **Product:** Dana - **Engineering:** Luis - **Design:** Mei

Answers are shown exactly as the model wrote them, from the first try, and nothing is picked by hand.

The models

2 models

Claude Sonnet 5.5

made by Anthropic

claude-sonnet-5-5 · revision not reported by the provider

1 try per case · tested 3 Nov 2026

GPT-6.1 Sol

made by OpenAI

gpt-6.1-sol · revision not reported by the provider

1 try per case · tested 3 Nov 2026

The tasks

written by Carte Blanche

bus-driver-assignment

“Write the PRD for a driver-assignment feature in our school bus software.”

dental-intake-lean

“Write a PRD for this. Keep it to a page or two so the clinic managers will actually read it.”

kickoff-notes-prd

“Turn these kickoff notes into a PRD.”

review-checkout-prd

“Can you review this PRD before I send it to engineering?”

vague-ai-prd

“Write a PRD for adding AI to our invoicing app.”

Install

Download the .zip and unzip it. Put its files, in a folder named create-prd, here:

Claude Code~/.claude/skills/Every project. For one project only, use .claude/skills/ inside it.

Codex~/.agents/skills/Every project. For one repository only, use .agents/skills/ inside it.

Cursor~/.cursor/skills/Cursor also reads the Claude Code and Codex folders above.

Then start a new session and ask for what the skill does. Other agents that read SKILL.md have their own skills folder; check their docs.

Coming after launch: get skills from inside your agent, no download needed.