Using LeanProjax with work data

Last updated: August 28, 2026.

You found LeanProjax, you want to use it on a real process at work — and somewhere in the back of your mind is the real question: am I allowed to put company data in here? This page answers it honestly. The short version: there are three ways to get full value from LeanProjax that range from “no company data leaves the building at all” to “bring everything” — pick the one that matches your situation, and when in doubt, ask first. We even wrote the memo for you to ask with.

Path 1 — Summary statistics only (no raw data needed)

Most people assume a statistics tool needs their raw records. LeanProjax doesn’t. The platform computes process capability, sigma levels, and baselines from summary figures: a mean, a standard deviation, a sample size, or simple counts (units processed, units defective). Those numbers describe how a process performs without containing a single customer record, transaction, or employee name — they are the kind of figures that already appear in management reports.

• The guided project wizard builds your baseline from counts you type in — no upload involved.
• Process capability accepts summary statistics (mean, standard deviation, n) as a first-class input; raw measurements are optional.
• The public calculators run entirely in your browser with no account at all — a fine way to evaluate before deciding anything.

Path 2 — Anonymised or genericised data

When you do want row-level analysis (run charts, Pareto, hypothesis tests), you can usually strip what makes data sensitive before it leaves your spreadsheet: remove identifiers, replace names with codes, drop columns that aren’t part of the analysis. A cycle time is a number; it stops being sensitive the moment it isn’t attached to a customer. Most DMAIC analysis needs the measurement and a category or two — not the record behind it.

Path 3 — The practice sandbox (no work data at all)

Every workspace can create practice projects — clearly marked training sandboxes with realistic scenarios. They are the right home for learning the method, building a certification project, or evaluating the platform end to end before any real data is involved. Nothing in a practice project touches your employer’s information.

When to ask first

If your company has a data-classification or approved-tools policy, operational data — defect rates, cycle times, cost figures, complaint text — may fall under it even when it feels harmless. We would rather you ask than guess: an improvement project that gets pulled for a policy breach helps nobody, least of all the process. Two honest rules of thumb:

• Working at a small company where you own the improvement task? You are probably the person who decides — read the security page and make the call.
• Working somewhere with an IT approval process? Start on Path 1 or 3 today, and hand your IT or security team the memo below to unlock Path 2 properly.

The one-page memo for your IT team → It answers the questions a reviewer actually asks — where data lives, how workspaces are isolated, what leaves the platform, how you get everything out — in one printable page.

What we do with your data

• Isolation: every workspace is fenced in three independent layers, down to database row-level security that fails closed. The security & trust page describes the controls — and the gaps — in plain terms.
• No training on your content: we do not train AI models on your project content, and we never sell or share it for advertising. Details in the privacy policy.
• Your exit is built in: any project exports as a complete package — workbook, full data, and PDF reports — so your work is never hostage to the platform.

Still unsure about your specific situation? Ask us directly at info@leanprojax.com — you will get a straight answer, including “don’t upload that” when that is the right answer.