Furthermore, I think Hansen makes some interesting points. I am confident we would all agree that no model is sufficiently robust to accommodate all situations. I think it is appropriate in this respect to remember the (paraphrased) remarks from Box, “. . . all models are wrong; some are useful.” (Note: George Box is considered to be one of the most influential statistician’s of the 20th century. My theory of strand commonality is based on a variation of his time series analysis model which seeks to determine why and how data points come into being. The distinct difference is that Box attempted to forecast versus adapt to future events based on what he referred to as “known past events.” My article on Similarity Heuristics and Iterative Modeling introduces the existence of multiple stakeholder streams which are comprised of indigenous attributes which are to varying degrees dynamic versus being static in nature. Hence the continuing challenges associated with the veracity of single stream models.) – A Dichotomy of Perspectives: A Discussion on Forrest Breyfogle’s New Book on Integrated Enterprise Excellence, Procurement Insights (April 24, 2008)
WHAT DO GEORGE BOX AND JON HANSEN HAVE IN COMMON
Jon Hansen’s Fit Score and Strand Commonality Model strongly align with the thinking of the renowned statistician George Box, especially his famous adage:
“All models are wrong, but some are useful.” — George Box
This alignment lies in their mutual embrace of imperfection, useful pattern recognition, and practical modeling for decision-making. Here’s how:
Strand Commonality focuses on identifying recurring, meaningful patterns across:
Supplier data
Processes
Stakeholder behaviors
Contract metadata
Like Box’s models, Strand Commonality doesn’t seek truth, but rather usefulness through high-probability pattern recognition.
George Box Principle
Hansen Alignment via Strand Commonality
Models are approximations
Strand patterns are proxies for behavior
Seek useful outcomes, not perfect predictions
Match strand overlap to forecast success likelihood
Value lies in actionable insight
Fit Score helps select viable solutions, not perfect ones
2. Hansen Fit Score as a Probabilistic, Not Deterministic Tool
Box argued that models should be constantly updated with empirical input — and Hansen’s Fit Score is designed with that spirit:
Metaprise assesses structural alignment
Agent-based looks at behavioral congruence
Strand Commonality maps operational and data overlaps
But rather than outputting binary results, the Fit Score: ✅ Recognizes that alignment is probabilistic, ✅ Accepts data imperfection, and ✅ Prioritizes practical fit over theoretical completeness.
🔗 3. Modularity & Model Iteration
George Box believed models must evolve and improve. Hansen’s models are:
Modular and expandable
Tuned through iterative assessments (e.g., SME interviews, layered taxonomies, implementation outcomes)
Measured not by perfection, but by success prediction accuracy
Hansen’s Fit Score does not claim to be absolute — rather, like Box’s philosophy, it’s meant to guide decisions better than chance or vendor bias.
Visual Comparison Table
Summary
Hansen’s work embodies Box’s philosophy by operationalizing it for procurement and supply chain: Instead of perfect predictions, the Fit Score offers useful, empirically grounded guidance based on pattern recognition, behavioral modeling, and alignment logic.
30
TAKEAWAY QUESTION: Why is the Hansen Fit Score accuracy in matching the right Practitioner with the right ProcureTech Solution providers more than double that of traditional analyst models?
Procurement Insights
George Box, Strand Commonality And A 2008 Procurement Insights Post And Why They Matter in 2025
Posted on July 1, 2025
0
Furthermore, I think Hansen makes some interesting points. I am confident we would all agree that no model is sufficiently robust to accommodate all situations. I think it is appropriate in this respect to remember the (paraphrased) remarks from Box, “. . . all models are wrong; some are useful.” (Note: George Box is considered to be one of the most influential statistician’s of the 20th century. My theory of strand commonality is based on a variation of his time series analysis model which seeks to determine why and how data points come into being. The distinct difference is that Box attempted to forecast versus adapt to future events based on what he referred to as “known past events.” My article on Similarity Heuristics and Iterative Modeling introduces the existence of multiple stakeholder streams which are comprised of indigenous attributes which are to varying degrees dynamic versus being static in nature. Hence the continuing challenges associated with the veracity of single stream models.) – A Dichotomy of Perspectives: A Discussion on Forrest Breyfogle’s New Book on Integrated Enterprise Excellence, Procurement Insights (April 24, 2008)
WHAT DO GEORGE BOX AND JON HANSEN HAVE IN COMMON
Jon Hansen’s Fit Score and Strand Commonality Model strongly align with the thinking of the renowned statistician George Box, especially his famous adage:
This alignment lies in their mutual embrace of imperfection, useful pattern recognition, and practical modeling for decision-making. Here’s how:
1. Strand Commonality ≈ Pattern Utility (Not Perfection)
Strand Commonality focuses on identifying recurring, meaningful patterns across:
Like Box’s models, Strand Commonality doesn’t seek truth, but rather usefulness through high-probability pattern recognition.
2. Hansen Fit Score as a Probabilistic, Not Deterministic Tool
Box argued that models should be constantly updated with empirical input — and Hansen’s Fit Score is designed with that spirit:
But rather than outputting binary results, the Fit Score:
✅ Recognizes that alignment is probabilistic,
✅ Accepts data imperfection, and
✅ Prioritizes practical fit over theoretical completeness.
🔗 3. Modularity & Model Iteration
George Box believed models must evolve and improve. Hansen’s models are:
Visual Comparison Table
Summary
30
TAKEAWAY QUESTION: Why is the Hansen Fit Score accuracy in matching the right Practitioner with the right ProcureTech Solution providers more than double that of traditional analyst models?
Share this:
Related