Skip to content
Tallyfy

Measuring process performance with sigma levels

What are sigma levels and why do they matter?

Sigma levels measure process quality as defects per million opportunities (DPMO). This gives you a universal metric to compare different processes - invoice accuracy, on-time delivery, anything - and set targets based on world-class benchmarks instead of arbitrary percentages.

Most organizations sit between 2 and 4 sigma. At 3 sigma (93.3% defect-free), performance sounds good. But that’s 66,807 defects per million opportunities. For a hospital, that’s unacceptable medication errors. For a bank, thousands of incorrect transactions.

Understanding the sigma scale

The sigma scale uses standard deviations to measure how well a process meets customer requirements:

  • 1 Sigma: 31% successful (690,000 DPMO) - Barely functional
  • 2 Sigma: 69% successful (308,537 DPMO) - Significant errors
  • 3 Sigma: 93.3% successful (66,807 DPMO) - Typical performance
  • 4 Sigma: 99.38% successful (6,210 DPMO) - Good performance
  • 5 Sigma: 99.977% successful (233 DPMO) - Excellent performance
  • 6 Sigma: 99.99966% successful (3.4 DPMO) - World class

The real impact of sigma levels

At 99% quality (3.8 sigma):

  • 20,000 lost articles of mail per hour
  • 5,000 incorrect surgical operations per week
  • 2 short or long landings at major airports daily

At 99.99966% quality (6 sigma):

  • 7 lost articles of mail per hour
  • 1.7 incorrect surgical operations per week
  • 1 short or long landing every 5 years

How to calculate your process sigma

You need three concepts:

  1. Defect opportunities: Each customer requirement is a chance for defects. An invoice might have 5 opportunities - correct amount, right address, accurate items, proper formatting, timely delivery.

  2. Defects vs. defectives: A defective unit can have multiple defects. One incorrect invoice (defective) might have wrong amount AND wrong address (two defects).

  3. Sample size matters: Your data should represent typical performance, not best-case or worst-case scenarios.

The calculation:

  1. Count total defects in your sample
  2. Multiply units processed × opportunities per unit
  3. Calculate DPMO: (Defects ÷ Total Opportunities) × 1,000,000
  4. Convert DPMO to sigma using a conversion table

Example: Processing 500 insurance claims with 4 requirements each (completeness, accuracy, timeliness, proper documentation) = 2,000 opportunities. Finding 40 defects gives DPMO of 20,000, approximately 3.4 sigma.

Using Tallyfy to track sigma performance

Set up measurement:

  • Define defect opportunities as required fields in task forms
  • Use validation rules to catch defects at the source
  • Track rework tasks as defect indicators

Monitor performance:

  • Analytics calculate cycle times and completion rates automatically
  • Process health indicators show trends over time
  • Export data for detailed sigma calculations

Drive improvement:

  • Comments capture why defects happen
  • Pattern analysis reveals common failure points
  • A/B test process changes to raise sigma levels

Setting meaningful targets

Don’t set arbitrary goals like “reduce errors by 50%.” Use sigma levels for context-appropriate targets instead:

Life-critical processes (healthcare, aviation): Target 5-6 sigma

  • Even small error rates have severe consequences
  • Near-perfection pays off in lives saved

Financial processes (billing, payroll): Target 4-5 sigma

  • Errors directly hurt customer trust and regulatory compliance
  • Prevention costs less than correction

Internal processes (expense reports, meeting scheduling): Target 3-4 sigma

  • Balance improvement costs with business impact
  • Prioritize customer-facing processes first

Common pitfalls in sigma measurement

Measuring activities, not outcomes: Tracking “emails sent” rather than “customer issues resolved” misses the point. Measure what customers value.

Ignoring hidden factories: Rework often hides in unmeasured activities. A “quick fix” culture masks true sigma performance. Make rework visible.

Cherry-picking data: Measuring only your best performers or easiest cases inflates sigma levels. Include all typical work for accurate baselines.

Overlooking customer requirements: Internal quality standards may not match customer expectations. A perfectly formatted report delivered late still fails the customer.

Beyond the numbers

Sigma levels are useful, but keep perspective:

  • Context matters: 4 sigma might be excellent for one process, inadequate for another
  • Costs escalate: Moving from 3 to 4 sigma typically costs far less than 5 to 6
  • Culture matters: Higher sigma levels demand systematic thinking, not heroic efforts

The goal isn’t perfection everywhere - it’s the right quality level for each process. Use sigma levels to decide where improvement investment will have the most customer impact.

Process Improvement > Introduction to DMAIC

DMAIC (Define Measure Analyze Improve Control) is a five-phase methodology that turns process improvement from guesswork into data-driven change by targeting root causes and maintaining long-term performance gains.

How To > Process improvement

Learn process improvement methods like DMAIC, Lean, and Kaizen - with practical techniques for identifying customer needs, eliminating waste, and using Tallyfy analytics and AI suggestions to sustain continuous improvements.

How To > Improve processes effectively

Tallyfy enables continuous process improvement by collecting team and customer feedback analyzing performance bottlenecks deploying instant template updates and implementing incremental changes while balancing standardization with flexibility to improve workflows and boost customer satisfaction.
Get in touch
About Tallyfy