Lean Information: When Bad Data Becomes Waste

Everyone knows lean management from industry. But what if the biggest quality problem lies not in production, but in the information itself? A concept, a prototype – and a browser button.

From Toyota Assembly Line to Inbox

Lean Management refers to the entirety of thinking principles and methods for efficiently designing the entire value chain. Its origins lie in the 1950s with Japanese automaker Toyota – more precisely: in the Toyota Production System by Taiichi Ohno. The goal is: creating value without waste. All activities necessary for value creation should be optimally coordinated, and unnecessary activities – Japanese "Muda" – should be avoided.

What Toyota developed for assembly lines became a worldwide leadership philosophy. Today, Lean Management is applied in hospitals, administrations, and software companies. Lean Management is no longer limited to manufacturing processes, but includes other business areas – from maintenance to order processing.

However, one area remained outside for a long time: information itself.

8 Types of Waste – and the Ninth

The classic Lean model recognizes eight types of waste: overproduction, waiting times, transport, unnecessary processes, inventory, motion, defects – and unused employee potential.

Ron Hyams, founder of LeanInformation.com and member of the Swiss Knowledge Management Forum, identifies a ninth, systematically overlooked type: wasteful information. In his article "Plain Language and Lean Information", published in the PLAIN eJournal Vol. 8 (2026), he writes:

"Information is lean's overlooked waste."

His core argument: employees generate the largest portion of information in an organization – reports, emails, spreadsheets, data records. When this information is poor, not only efficiency suffers. In the era of artificial intelligence, the AI systems that train on or work with this data also suffer: Garbage in, garbage out.

What is Lean Information?

Lean Information is not simply "less information." It is the minimum amount of information an audience needs to understand and act – no more, no less. Too much is waste. Too little is also waste.

Ron Hyams' framework defines four quality attributes – two sides of the same coin:

These four pairs are universal – they apply to a sentence in a newsletter just as much as to a column in a database, to a subject line just as much as to an API result.

TRIM: The Tool for Daily Use

Theory is one thing. But how should employees actually provide feedback on wasteful information in their daily work – without great effort, without a separate system, directly when the problem occurs?

This is exactly where TRIM comes in – the Tool for Reporting Information Mediocrity (or simply: the tool that makes bad information visible).

TRIM is a browser extension for Chrome and Firefox. Operation follows a single workflow:

1. Mark text – on any website, in the intranet, in a PDF
2. Click ⚑ TRIM button – appears automatically when marking
3. Step 1: Choose waste type – Meaningless / Ambiguous / Incorrect / Incomplete
4. Step 2: Choose solution path – from the Solution Grid (Needed, Plain, Structured, Verified, Governed, …)
5. Submit – feedback goes to the organization, including URL and task context

The result: a structured, traceable signal. Not a complaint, but a quality impulse.

The Prototype: lean.clarus.news

A first functional prototype emerged from the collaboration between Ron Hyams (Lean Information Framework) and Andreas Binggeli (technical architecture, clarus.news).

lean.clarus.news is a first prototype – v0.1 – and demonstrates the technical feasibility of the concept. The stack:

• Browser Extension (Manifest V3, Chrome + Firefox) – the visible part for employees
• ASP.NET Core Backend (C#) with PostgreSQL database – receives and stores captures
• MCP integration with Claude Sonnet (Anthropic) – automatically classifies each capture by Lean waste type, severity level and suggests action recommendations
• Apache Reverse Proxy on Ubuntu 24.04 with Cloudflare integration

The prototype is open source and can be loaded directly from the browser without store installation.

Why This is Relevant

Organizations invest millions in data quality – in cleansing, enrichment, governance. What they do far less often: make the people who work daily with this information into information stewards. Into people who actively provide quality signals, not just receive quality guidelines.

Lean Information connects three movements that previously ran alongside each other:

• Plain Language – understandable, clear language
• Lean Management – systematically eliminate waste
• AI Readiness – clean data as prerequisite for reliable AI systems

The browser button is small. The idea behind it is not.

Prototype and technical documentation: lean.clarus.news Framework: leaninformation.com – Ron Hyams, CC BY-ND 4.0 Technical implementation: Andreas Binggeli, clarus.news