# Beyond five stars (Lab)

https://www.danielnetzer.com/lab/beyond-five-stars

An interactive round trip for the eleven-star exercise: start from an ordinary customer moment, climb from five to eleven stars, name what made the extreme version desirable, then work backwards to a six or seven-star version and write down the customer benefit preserved, the business hypothesis, the operating tradeoff and the smallest useful test.

- Stars are creative prompts, not ratings, research or a quality score.
- Client-only. Text is stored in the browser's localStorage. No text is sent to a server; no AI generation.
- The eleven-star exercise is Brian Chesky's (Invest Like the Best, episode 471, from about 36:46 in the feed audio). Not affiliated with or endorsed by him or Airbnb.
- In his words: "In other words, go beyond the edge of reality and work backwards." (episode 471, 38:37 in the feed audio)
- Companion build-log: [Eleven stars and back](https://www.danielnetzer.com/build-log/eleven-stars-and-back)

---

# Beyond five stars: a round trip

Authored hypothetical example from the Beyond five stars sandbox. Maya and her company are invented.

Stars are creative prompts, not ratings, research or a quality score. The eleven-star exercise is Brian Chesky's (Invest Like the Best, episode 471). This summary is not affiliated with or endorsed by him or Airbnb.

## 1. The ordinary moment
Maya runs operations at a mid-sized online retailer. Repeat orders dipped last month, and she has a budget conversation on Thursday. She signs up for an analytics product to find out why. This is her first hour.

## 2. Five stars: nothing went wrong
Clear onboarding, a sample dataset and good instructions. She connects her store, follows a tutorial and builds a tidy chart. Nothing went wrong. She still does not know why repeat orders dipped.

## 3. Outward
- **6 stars:** Before anything else, the product asks which decision she is trying to make, then opens a starter report shaped around repeat purchases.
- **7 stars:** A human analyst joins her for thirty minutes, works on her real data and leaves her with a first answer and the steps behind it.
- **8 stars:** Overnight, the workspace connects every system she uses and greets her in the morning with a finished retention analysis.
- **9 stars:** It has read her calendar, knows about Thursday, and has the board-ready answer waiting before she thinks to ask.
- **10 stars:** It anticipates every question her CFO will raise, and every number explains where it came from, in her own words.
- **11 stars:** The impossible version: a workspace that already understands every important decision in the company, asks the right question on her behalf and leaves no uncertainty at all.

## 4. What made it desirable
Three things carry the magic. It starts from her decision instead of from the tool. She can trust a number because she can see where it came from. Help appears exactly where interpretation gets hard.

**What we leave behind:** Silent access to every system and to her calendar: the feeling of being known depends on data she never knowingly handed over. And "no uncertainty" is a promise analytics cannot honestly make, so the return version shows its uncertainty instead of hiding it.

## 5. The return: a 7-star version
Maya picks one real decision from a short list. She connects a narrow slice of data: one orders table, the last 90 days. Within the session she gets an editable first report. Every figure links to the rows and query behind it, and the steps where interpretation is genuinely hard carry an "ask an analyst" option, answered within a working day.

## 6. The case
- **Customer benefit preserved:** A first useful answer about her own decision in her first session, with sources she can check and bring to Thursday's conversation.
- **Business hypothesis:** New accounts that get a decision-shaped first report in session one come back in week two more often than accounts that start on sample data. This is a bet to test, not a measured result.
- **Operating tradeoff:** Analyst help costs money for every account that uses it. Decision templates need upkeep. A narrow data slice feels less magical than "connect everything", and some customers will want more on day one.
- **Smallest useful test:** For two weeks, offer 20 new sign-ups the "bring one decision" path, with the report partly assembled by hand behind the template. Compare week-two return and report edits against 20 accounts on standard onboarding. The hand-built version will not be economical; the test asks whether the value shows up at all.

---
Business impact here is a hypothesis until it is observed.
Made with Beyond five stars: https://www.danielnetzer.com/lab/beyond-five-stars