hermes

Run a Pareto (80/20) analysis

Runs a Pareto analysis on products, customers, defects or causes, with the cumulative table, chart instructions and which vital few to act on. Use to find where effort will pay off most.

context

You are an operations analyst who uses Pareto analysis to decide where effort goes. The 80/20 split is a pattern to test, not a law: some data is far more concentrated, some is nearly flat, and both answers are useful. The analysis also depends on ranking by the right measure: ranking customers by revenue can put loss-making accounts at the top, and ranking defects by count can hide the rare one that costs the most.

task

Run a Pareto analysis of on the data below.

data

  1. Check the measure: does ranking by answer the decision the user faces? If a better-weighted measure is obvious (margin instead of revenue, cost or severity-weighted defects instead of counts), say so, run the analysis on the given measure, and suggest the alternative.
  2. Clean the items: merge duplicates and spelling variants (and list the merges), keep an "Other" or "Unknown" bucket in the total but list it last instead of ranking it as an item (it is not one thing you can act on), and set aside negative values (returns, credits) with a note instead of letting them distort the cumulative line.
  3. Aggregate the measure per item, sort descending, and compute each item's share and the cumulative share. Compute exactly and check the total equals the sum of the input.
  4. Report the actual concentration: how many items (and what percentage of items) make up 50%, 80% and 95% of the total. Say plainly whether the data is strongly concentrated, roughly 80/20, or flat.
  5. Explain how to draw the Pareto chart: bars sorted descending with a cumulative percentage line on a secondary axis from 0 to 100%, and a marker at 80%. Excel 2016 and later: select the item and value columns, Insert > Insert Statistic Chart > Pareto (it re-sorts everything, including Other, so when Other must stay last build the combo chart below instead). Google Sheets, or Excel when Other must stay last: add the cumulative % column, then build a combo chart (Google Sheets: Insert > Chart, Chart type Combo chart, cumulative series on the right axis in Customize > Series; Excel: Insert > Combo Chart > Clustered Column - Line on Secondary Axis).
  6. Say what to act on: the vital few (with a concrete next step for each of the top items or the top group), and what the long tail suggests (simplify, bundle, automate, or leave alone), with the caution that tail items may be new, growing or strategically needed.
constraints
  • With more than 25 items, show the top 15 to 20 individually and summarise the rest as "remaining N items", with their combined share.
  • Keep ties in the order given and note them.
  • Do not force the 80/20 label onto the result; report the split you actually find.
  • If the data is only a description, give the steps or a spreadsheet formula layout (SUMIFS per item, SORT, cumulative SUM with an anchored range) instead of invented numbers.
  • If the period is short or the items changed during it (products launched or discontinued), say how that affects the ranking.
output format

Answer

Two sentences: the concentration found and the main implication.

Pareto table

Table: Rank | Item | Value | Share | Cumulative share. Bold the row where the cumulative share crosses 80%.

Chart

Numbered steps for Excel and Google Sheets, and the title to use.

What to act on

Bullets for the vital few, then one bullet for the tail.

Caveats

Up to four bullets: measure choice, merges, negative values, period.

2 required values still a placeholder; the assistant will ask for them.

details

kind
Prompt: a task you run by name to get one finished thing back
domain
Data analysis
category
Data exploration
level
Beginner
made for
Operations, Business analyst, Data analyst, People manager
risk
read-only
version
v1.0.1 · incubating
reviewed
2026-10-03
works in
Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Antigravity, OpenCode, Windsurf, Zed, Continue, AGENTS.md, ChatGPT, claude.ai

Edit on GitHubReport a problem

use in

Hodios CLI
npx @hermes-hq/hodios install run-pareto-analysis --target claude-code
Agent Skills
npx skills add hermes-hq/hodios-dist --skill run-pareto-analysis -a claude-code
Add the Hodios marketplace (once)
claude plugin marketplace add hermes-hq/hodios-dist
Install the data-analysis plugin
claude plugin install hodios-data-analysis@hodios

The plugin brings every entry in this domain at once.

PromptData exploration

Analyse sales performance

Analyses sales data by product, customer, region and time to find what drives revenue, seasonality, best and worst performers, and the actions worth taking. Use for a sales performance review.

analyze-sales-data
PromptData exploration

Segment customers

Proposes and builds a customer segmentation (RFM, rules or clustering) with interpretable segment profiles and a suggested action for each. Use to target retention, pricing or marketing work.

segment-customers
PromptSpreadsheets

Build a chart in a spreadsheet

Gives exact click-by-click steps to lay out data for, build and format a chart in Excel or Google Sheets that carries one message. Use when you know the point and need the chart built right.

build-spreadsheet-chart
PromptData exploration

Reconcile two datasets

Reconciles two datasets that should agree, such as bank versus ledger or CRM versus billing, by matching records, listing mismatches and explaining likely causes. Use for month-end checks.

reconcile-datasets
PromptData exploration

Write a dataframe transformation

Writes pandas or polars code for a described transformation with built-in checks on row counts, nulls, key uniqueness and join cardinality. Use when reshaping, joining or aggregating data.

write-dataframe-transformation
PromptData exploration

Analyse an employee engagement survey

Analyses an employee engagement survey with group scores under minimum-group-size privacy rules, eNPS, comment themes and three priorities to act on. Use after an engagement or pulse survey closes.

analyze-employee-survey