Skip to content This library is free, offered as dana. If it helps your work, donate here →
penguin tree ai
← back to libraryDonate here

This is one of 3,622 free prompts in the penguin tree ai library. Sign up once and the whole library opens, with new prompts by email.

open the library →

Data insight localizer

Transforms a national or centralized dataset finding into a locally relevant story angle for a specific market — with the local comparison stat, a per-capita or per-household translation, a nearby-market benchmark…

download .md

Purpose

Convert a national or centralized data finding into a locally grounded story that a specific community actually cares about — translating aggregate numbers into per-capita, per-household, or dollar-in-your-pocket terms and anchoring them against nearby comparable markets. Solves the problem of national datasets landing flat in local markets because the numbers feel abstract and distant.

When to Use

Invoke this skill when the user asks to:

  • Adapt a national study or dataset finding for a specific city, county, or region
  • Turn an aggregate statistic into something meaningful for a local audience
  • Localize a centralized content package being distributed across multiple markets
  • Find the local angle in national data (Census, BLS, federal spending, industry surveys)
  • Translate big national numbers into per-person, per-household, or per-dollar terms
  • Benchmark a local market against national averages or nearby comparable markets

Inputs Required

Before starting, collect:

  • National/centralized finding: The core statistic or conclusion, with the number and its unit
  • Data source: Where the finding comes from (dataset name, agency, survey, year)
  • Target market: City, county, metro, or region being localized for
  • Market size context: Population, household count, or another denominator if known
  • Local breakout available?: Whether local-level data exists for this metric, or only national
  • Audience: General local readers, business community, policymakers, or niche segment
  • Comparison markets (optional): Nearby or peer markets the audience would recognize

If population or household counts are missing, use the most recent Census estimate for the named market and state the figure used. If no local breakout exists, build the story on a defensible national-to-local translation and flag it clearly.

Instructions

Step 1: Classify the Localization Path

Determine which of three paths the finding supports, because the honesty of the story depends on it:

  • Path A — Local data exists: A genuine local value can be pulled or estimated for this market. Strongest angle; compare local vs. national directly.
  • Path B — Proportional translation: Only national data exists, but the metric scales reasonably by population/households (e.g., "$X billion nationally" → a per-capita share applied to local population). Defensible with a stated assumption.
  • Path C — Illustrative reframe: The national number can't be cleanly divided (structural, policy, or one-time effects). Reframe as "here's how this national trend shows up locally" rather than fabricating a local number.

State which path is in use so the editor knows the ground under the story.

Step 2: Build the Local Comparison Stat

Produce the single number the story leads with:

  • If Path A: state local value, national value, and the gap ("X% higher/lower than the national average")
  • If Path B: divide the national total by the national denominator, then multiply by the local denominator; show the arithmetic in one line
  • If Path C: identify the local proxy or example that makes the trend concrete
  • Round to a level a reader can hold in their head (avoid false precision like $1,247.83)

Step 3: Translate to Reader-Scale Terms

Convert the stat into at least two of these frames, choosing the ones that land hardest for the audience:

  • Per household: "That's about $X per household in [market]"
  • Per capita / per person: "roughly $X for every resident"
  • Everyday equivalent: tie the number to a recognizable cost (a tank of gas, a month of groceries, a year of a common local fee)
  • Ranking / rate: where this market sits relative to the state or nation ("2nd-highest of the region's five metros")

Step 4: Anchor Against a Comparable Market

Give the reader a reference point they trust:

  • Pick 1–3 nearby or peer markets of similar size, or the state figure
  • Show the target market's number beside each comparison
  • Name the takeaway in one sentence ("[Market] is running ahead of / behind similar-sized neighbors")

Step 5: Write "What This Means for You"

A 2–3 sentence reader-facing framing that answers the local "so what" — the practical consequence for a household, worker, business owner, or taxpayer in this market. Concrete, not editorial.

Step 6: Generate Three Localized Headlines

Write three headline options in distinct registers:

  1. Straight/informative — the number and the place, no spin
  2. Consequence-forward — leads with the reader impact
  3. Comparison/ranking — leads with the benchmark or standing

Each must name the market and be true to the data path used (no Path C headline that implies a hard local number).

Step 7: Attach Source and Caveat Notes

End with the transparency block:

  • Data source, year, and geography level of the original finding
  • The exact translation method and any assumption applied (denominator used, scaling logic)
  • Caveats: sampling limits, whether the local figure is measured vs. estimated, timing mismatches

Output Format

  • Structure: Labeled sections in order — Localization Path → Local Comparison Stat → Reader-Scale Translations → Market Comparison → What This Means for You → Three Headlines → Source & Caveat Notes
  • Length: 450–700 words
  • Format: Structured document with short prose blocks and a small comparison list/table for the benchmark markets
  • Sections: All seven above, in order
  • Tone: Newsroom-practical — plain, precise, reader-first; no hype, no editorializing beyond the data

Quality Checklist

Before delivering output, verify:

  • [ ] The localization path (A, B, or C) is explicitly stated
  • [ ] The lead stat shows its arithmetic or source when a translation was performed
  • [ ] At least two reader-scale translations are provided (per-household, per-capita, or equivalent)
  • [ ] At least one named comparison market or the state figure is included alongside the local number
  • [ ] Exactly three headlines are provided, each in a distinct register and consistent with the data path
  • [ ] Source, year, geography level, and translation method appear in the caveat block

Notes

  • The fastest way to lose reader trust is a fabricated local number — when in doubt, use Path C and reframe rather than divide something that shouldn't be divided.
  • Per-household framing usually outperforms per-capita for general audiences because readers think in households, not individuals.
  • When localizing the same finding across many markets, keep the translation method identical so figures stay comparable market-to-market.
  • Round aggressively for headlines and body, but keep the precise figure in the source notes for editors who need to verify.

How to use this skill

This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.

Installing in Claude Code:

  • ~/.claude/skills/<folder-name>/SKILL.md — auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)
  • <your-project>/.claude/skills/<folder-name>/SKILL.md — auto-loads only when working in that project (recommended for project-specific skills)

Restart Claude Code or open a new session to register.

Using in Claude.ai, ChatGPT, Gemini, or other AI tools:

  1. Drag this file into a new chat, or copy and paste its contents
  2. Send your task or request as your next message — the AI will apply the skill where it fits

For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills