A fishbone diagram (also called an Ishikawa or cause-and-effect diagram) helps teams brainstorm every potential cause of a problem before jumping to conclusions. Below are seven complete fishbone diagram examples drawn from real industries — each with filled-in categories, identified root causes, and the corrective actions that followed.
Whether you work in manufacturing, healthcare, software, or any other field, these examples show you exactly how to structure your own fishbone diagram. Each example uses the classic category framework suited to its industry, with 3–5 causes per branch to demonstrate the level of detail that makes this tool effective.
1. Manufacturing: High defect rate on assembly line
Industry context
An automotive parts manufacturer saw its defect rate climb from 1.2% to 4.8% over two months. The team used a fishbone diagram to systematically identify all possible contributing factors before investigating each one.
Problem / Effect: Assembly line defect rate increased to 4.8% (target < 1.5%)
| Category (6M) | Potential Causes |
|---|---|
| Man (People) | New operators lack training on updated specs; night shift has 40% less experience; high overtime causing fatigue |
| Machine | Calibration drift on CNC press #7; worn tooling not replaced on schedule; vibration in conveyor belt |
| Method | Work instructions not updated after design change; inspection sampling rate too low; no standardized handoff between shifts |
| Material | New supplier resin has different viscosity; incoming material inspection skipped for rush orders; batch-to-batch variation |
| Measurement | Go/no-go gauge tolerance too wide; visual inspection criteria subjective; CMM not recalibrated after move |
| Mother Nature | Summer humidity affecting adhesive cure time; temperature swings between shifts; dust from nearby construction |
Root cause identified: Work instructions were not updated after a design revision three months prior. Operators followed the old tolerances, and the wider go/no-go gauge did not catch the resulting deviations.
Corrective action: Implemented a mandatory document control step linking every engineering change order (ECO) to a work instruction review within 48 hours. Recalibrated gauges to match the new specification. Defect rate returned to 1.1% within four weeks.
2. Healthcare: Patient falls on medical-surgical unit
Industry context
A 200-bed hospital noticed a 35% increase in patient falls on its medical-surgical floor over one quarter. The quality team assembled nurses, aides, and physicians for a fishbone brainstorming session.
Problem / Effect: Patient fall rate increased from 2.1 to 3.4 per 1,000 patient-days
| Category | Potential Causes |
|---|---|
| People | Agency nurses unfamiliar with fall protocol; patient non-compliance with call light use; insufficient 1:1 sitter coverage |
| Process | Fall risk assessment not repeated after medication changes; hourly rounding inconsistent on nights; bed alarm response time > 3 min |
| Environment | Wet floors near bathroom; poor lighting in hallways at night; cluttered rooms limit walker access |
| Equipment | Bed alarms malfunctioning on 8 beds; non-slip socks out of stock for 2 weeks; call light cords too short |
| Policy | No mandatory post-sedation reassessment; fall bundle compliance not audited monthly; restraint policy too restrictive to allow bed rails |
| Measurement | Near-miss falls not tracked; fall risk score cutoff too high; incident reports filed late, delaying trend analysis |
Root cause identified: Fall risk assessments were not being repeated after medication changes. Patients started on new sedatives or opioids were not re-scored, so high-risk patients did not receive enhanced precautions.
Corrective action: Added an automatic trigger in the EHR that flags a fall risk reassessment whenever a sedating medication is ordered. Combined with a nursing education blitz, the fall rate dropped to 1.8 per 1,000 patient-days within 60 days.
3. Software: Production outage lasting 4 hours
Industry context
A SaaS platform serving 50,000 users experienced a four-hour outage during peak business hours. The engineering team conducted a fishbone-based postmortem to identify all contributing factors.
Problem / Effect: Complete service outage for 4 hours, affecting all customers
| Category | Potential Causes |
|---|---|
| People | On-call engineer was new and unfamiliar with runbooks; deploy was performed by one person with no reviewer; insufficient cross-training on database layer |
| Process | No canary deployment; rollback procedure untested; change management window ignored for "small" changes |
| Technology | Database migration script had no rollback path; monitoring did not alert until 100% error rate; load balancer health check too lenient |
| Environment | Staging environment did not match production data volume; deploy happened during peak traffic; third-party DNS propagation delay |
| Policy | No freeze window policy during peak hours; deploy approval not required for schema changes; incident severity classification too slow |
Root cause identified: A database migration script lacked a rollback path and was deployed without canary testing during peak hours. The staging environment had only 1% of production data volume, so the migration completed in seconds there but locked tables for hours in production.
Corrective action: Mandated that all database migrations include a tested rollback script. Implemented canary deployments (1% traffic for 30 minutes before full rollout). Added a production-scale staging dataset refreshed weekly. Created a deploy freeze window during peak hours (9 AM – 12 PM).
Build your own fishbone diagram
Use our free online tool to create, collaborate, and export your Ishikawa diagram in minutes.
Start Free Fishbone Diagram →4. Restaurant: Food safety complaints
Industry context
A regional restaurant chain received three customer illness complaints in one month, all traced to a single location. Management assembled kitchen staff, the general manager, and a food safety consultant for a fishbone session.
Problem / Effect: Three foodborne illness complaints in 30 days at one location
| Category | Potential Causes |
|---|---|
| People | Two new cooks skipped food handler certification; handwashing compliance low during rush; manager not present during evening shift |
| Process | FIFO rotation not followed in walk-in cooler; temperature logs filled in retroactively; cross-contamination during prep (same cutting boards) |
| Materials | Chicken delivered above 41°F on two occasions; produce supplier changed without quality audit; sanitizer concentration too low |
| Equipment | Walk-in cooler thermostat reading 2°F higher than actual; dishwasher not reaching sanitizing temperature; probe thermometer not calibrated |
| Environment | Summer heat causing dock temperature spikes during delivery; fly strip near prep area full and ineffective; floor drain backing up near cooler |
Root cause identified: The walk-in cooler thermostat was reading 2°F higher than the actual temperature, meaning food stored at a displayed 38°F was actually at 40°F. Combined with FIFO violations, older proteins were held in the danger zone long enough for bacterial growth.
Corrective action: Replaced the thermostat and installed an independent digital temperature monitor with SMS alerts. Implemented daily FIFO audits with a checklist. Required all new hires to complete food handler certification before their first solo shift. Zero complaints in the following 90 days.
5. Logistics: Delivery delays exceeding SLA
Industry context
A regional parcel delivery company saw its on-time delivery rate fall from 96% to 82% over six weeks. The operations team used a fishbone diagram to map out every potential cause before assigning investigation priorities.
Problem / Effect: On-time delivery rate dropped to 82% (SLA target: 95%)
| Category | Potential Causes |
|---|---|
| People | Driver shortage due to seasonal demand; new drivers unfamiliar with routes; dispatcher overloaded with manual route planning |
| Process | Route optimization not updated for new residential developments; no real-time rerouting for traffic; package sorting errors causing reloads |
| Equipment | Three trucks down for maintenance simultaneously; GPS units outdated on 20% of fleet; handheld scanners crashing mid-route |
| Materials | Oversized packages not flagged at intake, causing truck capacity issues; incorrect address labels from shipper; fragile items requiring extra handling time |
| Environment | Winter road closures; construction detours on two major routes; holiday volume surge 30% above forecast |
| Measurement | Delivery timestamp recorded at scan, not at door; SLA clock starts at wrong event; customer notification system delayed |
Root cause identified: Route optimization had not been updated to account for three new residential developments. Drivers were spending 15–20 extra minutes per route navigating unmarked roads. Combined with the holiday volume surge, these delays cascaded across all afternoon deliveries.
Corrective action: Updated the routing software with current map data and added a monthly map data refresh cycle. Hired four seasonal drivers to absorb the volume increase. On-time rate recovered to 94% within three weeks and reached 97% by end of quarter.
6. Marketing: Campaign underperformance
Industry context
A B2B software company launched a product campaign expecting 500 qualified leads but generated only 180 after four weeks. The marketing team used a fishbone diagram to avoid finger-pointing and systematically analyze every factor.
Problem / Effect: Q1 product launch campaign generated 64% fewer leads than projected
| Category | Potential Causes |
|---|---|
| People | Content writer unfamiliar with target persona; sales team not briefed on campaign messaging; no dedicated campaign manager |
| Process | Landing page launched without A/B testing; lead scoring model not aligned with sales; no nurture sequence for non-converters |
| Message | Value proposition unclear; too much jargon; CTA language weak ("Learn more" vs. "Get your free audit") |
| Channel | LinkedIn ads targeted too broad an audience; email list included 30% inactive contacts; webinar scheduled at low-attendance time |
| Technology | Form had 9 fields causing abandonment; page load time 5.2s on mobile; tracking pixel misconfigured for 10 days |
| Measurement | Lead projection based on outdated benchmarks; attribution model over-counted organic; no mid-campaign performance review scheduled |
Root cause identified: The landing page form had nine fields, causing a 72% abandonment rate (industry average for B2B is 3–5 fields). The tracking pixel misconfiguration meant the team did not see this data for the first 10 days, delaying corrective action.
Corrective action: Reduced form fields to four (name, email, company, role). Fixed the tracking pixel immediately. Added a mid-campaign review checkpoint at day 7 for all future campaigns. Lead generation rate tripled within two weeks of the form change.
7. HR: Employee turnover exceeding industry average
Industry context
A mid-size tech company saw annual voluntary turnover rise from 14% to 26%, well above the industry average of 18%. HR partnered with department heads to run a fishbone analysis using exit interview data and engagement survey results.
Problem / Effect: Annual voluntary turnover at 26% (target < 18%)
| Category | Potential Causes |
|---|---|
| People | Managers not trained in coaching or feedback; no mentorship program for mid-career staff; hiring for skills but not culture fit |
| Compensation | Base pay 8% below market median; equity refresh grants not competitive; no spot bonus program for high performers |
| Career Growth | No clear promotion criteria; individual development plans not followed up; lateral moves discouraged by managers |
| Culture | Remote workers feel excluded from decisions; meetings dominated by senior voices; recognition only happens at annual review |
| Workload | Chronic understaffing in engineering; on-call rotations too frequent; unrealistic sprint commitments |
| Environment | Office renovation causing noise; inflexible return-to-office mandate; poor tooling slowing daily work |
Root cause identified: Exit interviews revealed that 68% of departing employees cited lack of clear promotion criteria and stalled career growth as their primary reason for leaving. Managers had no structured framework for discussing career progression, and development plans created during annual reviews were never revisited.
Corrective action: Published transparent promotion criteria for every level. Required quarterly career development check-ins between managers and direct reports. Launched a formal mentorship program. Within two quarters, voluntary turnover dropped to 19% and engagement scores improved by 22 points.
How to apply these examples to your own analysis
These seven examples share a common pattern you can replicate for any problem:
- State the problem precisely. Every example above starts with a measurable effect, not a vague complaint. Instead of "quality is bad," write "defect rate increased to 4.8% (target < 1.5%)."
- Choose the right categories. Manufacturing problems fit the 6M framework. Service and IT problems often need custom categories like Channel, Message, or Career Growth. Use categories that reflect where causes actually live in your organization.
- List 3–5 causes per category. Fewer than three often means the team did not brainstorm deeply enough. More than five per branch can become overwhelming — break the category into sub-categories instead.
- Validate before acting. The fishbone identifies possible causes, not confirmed ones. Every example above required data analysis or investigation before pinpointing the root cause. Use 5 Whys to drill deeper into your top candidates.
- Define corrective actions with owners and deadlines. A fishbone without follow-through is just a wall decoration. Assign each action to a person, give it a deadline, and schedule a verification check.
If you want step-by-step instructions for building your own diagram from scratch, see our guide on how to create a fishbone diagram.
Frequently asked questions
What is a good example of a fishbone diagram?
A good fishbone diagram example starts with a clear, measurable problem statement and maps causes across 4–6 categories relevant to the context. Each branch lists specific, observable causes rather than vague labels. For instance, instead of writing "bad training," write "new operators lack training on updated specification rev. 4.2." The more specific, the more actionable.
What industries use fishbone diagrams?
Fishbone diagrams are used across virtually every industry. They originated in manufacturing but are now common in healthcare, software engineering, food service, logistics, marketing, and human resources. Any problem with multiple potential causes benefits from the structured brainstorming a fishbone provides.
How many categories should a fishbone diagram have?
Most effective fishbone diagrams use 4 to 6 categories. Manufacturing typically uses the 6M framework (Man, Machine, Method, Material, Measurement, Mother Nature). Service industries often use custom categories. Fewer than 4 usually means you are missing perspectives, and more than 8 becomes unwieldy for a single brainstorming session.
How do I identify the root cause from a fishbone diagram?
After brainstorming, use dot voting to identify the most likely causes. Then validate with data: check logs, run experiments, or analyze metrics. Many teams follow up with a 5 Whys analysis on the top 2–3 candidates. The root cause is the one whose removal actually prevents the problem from recurring.
Can I use a fishbone diagram for non-manufacturing problems?
Absolutely. While fishbone diagrams originated in manufacturing, they work for any problem with multiple potential causes. Simply adjust the category labels to fit your domain. Software teams use categories like People, Process, Technology, and Environment. Marketing teams might use Message, Channel, Audience, and Timing.
Recommended Reading
- The Lean Six Sigma Pocket Toolbook — George et al. — Comprehensive reference covering fishbone diagrams, 5 Whys, and dozens of other quality tools
- Root Cause Analysis: Simplified Tools and Techniques — Bjorn Andersen & Tom Fagerhaug — Practical guide with templates and worked examples
- The Quality Toolbox — Nancy R. Tague — Encyclopedic reference for every major quality improvement tool
Related resources
- Free Online Fishbone Diagram Tool — build your own Ishikawa diagram in minutes
- How to Create a Fishbone Diagram: Step-by-Step Guide
- 5 Whys vs. Fishbone Diagram: When to Use Which
- Root Cause Analysis Examples Library
- Free 5 Whys Online Tool — drill deeper into your fishbone findings