🛠️ Scorecard component sketches

💡 Use Case

• Early components for 📂 Carrier Scorecards discovery, started on Friday 24 October as rough boxes of my own and taken over, mercifully, by Tomasz Wilk the following week.
• The range bar is the one that matters: a carrier shown as a stretch and not a point. A carrier with thousands of shipments gets a short bar; a small one gets a long bar, because we know less about it. It replaced the rank, the column of 1 to 20 that everybody asks for first: 🔍 Show carrier scores as ranges, not ranks.
• The others so far: the carrier row, a how many shipments is this based on line, a reason breakdown, and a not enough data state that says so plainly.
• Still sketches. The build decision is in January, and nothing here is promised.

⚖️ Pros & Cons

Pros

• Two bars that overlap cannot be ranked, and the picture says so without a paragraph about uncertainty.
• It answers the objection Gideon Asante raised on 22 October, that a small carrier is all noise. The long bar is his noise, drawn.
• In the November prototype sessions nobody asked what the bar meant. They asked why one was so long, which is the right question.

Cons

• People still want a number one. A range feels like a refusal to answer.
• Sorting a list of ranges needs a rule, and every rule quietly brings the rank back.
• My first version used the interval from the course as it stood. Bao Nguyen has views on which interval, and they are better than mine.

🔄 Alternatives

• A league table with a rank. Simple, wanted and unfair.
• Three bands, such as good, mixed and poor. Kinder than a rank, and it hides the same thing.