What is a root cause tree?
A root cause tree is a branching diagram that starts with a problem and repeatedly asks why, splitting into several causes wherever more than one factor contributed. Unlike a linear 5 Whys, it captures the reality that most quality events and incidents have several interacting causes. Unlike a fault tree, it does not need formal logic gates or probabilities, which makes it a practical format for deviation, nonconformance and incident investigations.
Problem, causal factors and root causes
- Problem: the event being investigated, described with facts.
- Causal factors: the conditions, failures or actions that directly led to the problem. Removing one would have prevented or mitigated it.
- Intermediate causes: the reasons each causal factor existed.
- Root causes: the most basic causes that management can control and fix, usually weaknesses in procedures, design, training or oversight.
Keep going until the cause is something the organization controls. "Analyst made a mistake" is not a root cause; "the method allows two interpretations of the dilution step" is.
Root cause categories
Classifying root causes makes them comparable across investigations. This tool uses these categories, which map to most corporate and regulatory taxonomies:
| Category | Examples |
|---|---|
| Equipment / design | Inadequate design, missing interlock, calibration drift, maintenance gaps |
| Procedures | Missing, unclear, wrong or not-followed instructions; ranges not aligned with validation |
| Training / qualification | No training, ineffective training, qualification not verified |
| Human performance | Slips and lapses enabled by workload, distractions or poor ergonomics |
| Management & oversight | Inadequate change control, supervision, resources or risk assessment |
| Communication | Shift handover gaps, unclear responsibilities, information not shared |
| Materials / suppliers | Raw material variability, supplier changes, incoming inspection gaps |
| Environment / external | Temperature, humidity, utilities, external events |
Root cause tree example
The example investigates a batch that failed dissolution testing. Two causal factors were considered. A laboratory error was ruled out by a retest. Tablet hardness above the limit was verified and traced to two branches: a compression force set too high, because the batch record allows a wider range than the validated one, and an over-dried granulation, caused by a time-based dryer endpoint and a drifting temperature probe. The result is three root causes in two categories, each needing its own corrective action.
Tips for a strong root cause tree
- Back every cause with evidence: records, data, interviews or observations.
- Keep ruled-out branches to show that alternative hypotheses were considered.
- Check each level with "therefore": reading from the root cause back up should make logical sense.
- Start broad with a fishbone diagram if the team is unsure where to look.
How to use this Root Cause Tree tool
- State the problem. Write the problem at the root of the tree, with the facts that define it: what, where, when and how much.
- Add causal factors. Add the direct contributors to the problem, the conditions or actions without which it would not have happened.
- Ask why for each factor. Use the plus button to add the causes of each factor, branching wherever there is more than one.
- Record evidence and status. Note the evidence for each cause and mark it verified, suspected or ruled out as the investigation progresses.
- Flag and classify root causes. Tick "Root cause" on the causes that represent system weaknesses and choose a category for trending.
- Export for your investigation report. Download the tree and root cause summary as a PDF, PNG or Excel file for your deviation, CAPA or incident record.
Frequently asked questions
What is a root cause tree?
A root cause tree is a hierarchical diagram that breaks a problem down into causal factors, then into deeper causes, until it reaches the root causes: the underlying system weaknesses that, once corrected, prevent recurrence. It is sometimes called a why-why diagram, cause tree or root cause analysis tree.
How is a root cause tree different from 5 Whys?
The 5 Whys follows one linear chain. A root cause tree branches: every cause can have several contributing causes, so it captures problems with multiple root causes and shows how they relate. Each branch can still be developed by asking why.
How is a root cause tree different from a fault tree?
A fault tree uses formal AND/OR logic gates and probabilities, usually to analyze how failures combine. A root cause tree is more flexible: it records causes, evidence and verification status, and classifies root causes into categories for trending.
Why classify root causes into categories?
Categories such as procedures, training, equipment design or management oversight let you trend root causes across investigations. If most root causes in a year are procedure-related, the fix is systemic, not another one-off correction.
What does "ruled out" mean in the tree?
A ruled-out cause was investigated and shown not to contribute, for example a lab error excluded by a retest. Keeping ruled-out branches documents the investigation and shows inspectors that alternative causes were considered.
Can I export the root cause tree?
Yes. Export the diagram as PNG or as a PDF report including the root cause summary and the full cause list, or export the tables to Excel or CSV. You can also save a .json file to reopen later.