01 · Foundation
Foundation
Steradian's measurement approach builds on peer-reviewed research on strategic consensus, vertical and horizontal alignment, and shared mental models. The framework adopts the two-dimensional alignment structure introduced by Kathuria, Joshi and Porth (2007) and refines it using subsequent meta-analytic and contingency research.
The eight-domain capability lens also sits within broader strategy research on resource positioning and organizational learning modes (Barney, 1991; March, 1991). Steradian does not estimate resource economic value or exploration–exploitation rates; those sources anchor why multi-domain leadership perception is a standard object of inquiry.
We distinguish between three things: what peer-reviewed research establishes, what Steradian operationalizes as engineering choices, and what Steradian has not yet validated. This page covers all three.
02 · Constructs
Core measurement constructs
Steradian measures five constructs. Each is defined in the Terminology Dictionary; two feed the Steradian Score directly.
What the organization is able to do, measured as the average of leader self-assessments across eight strategic domains.
- Corporate Strategy and Vision
- Industry Landscape
- Innovation Culture & Mindset
- Operations & Delivery
- Product & Innovation Development
- Sales, Branding & Marketing
- Talent Management
- Digital Transformation & Automation
Vertical alignment. How similarly leaders at different organizational levels — Board, CEO, C-Suite, Middle Management — perceive capabilities across all domains. Higher scores indicate levels share the same picture of strengths and gaps.
Horizontal alignment. How much individual leaders agree with each other on a specific strategic domain (reported per domain). Higher scores indicate convergent perception within a domain.
A 50/50 composite of Cross-Level and Domain Alignment, used as the alignment input to the Steradian Score.
Execution readiness. The product of Capability and Combined Alignment — capability × (combined alignment ÷ 100). Reflects the moderating role of alignment on capability-to-performance translation.
03 · Diagnostic
Diagnostic constructs
Some metrics appear on reports for interpretation and data quality; they are not components of the Steradian Score formula.
Team Coherence
What it measures: internal consistency of an individual respondent's answers across questions within a domain. Used as a data-quality indicator.
What it does not measure: alignment with other respondents. Team Coherence is a within-respondent reliability check, not a between-respondent agreement metric.
Use: when Team Coherence is low for a respondent, their data is flagged as potentially low-engagement (straight-lining, contradiction, or inattention). Their responses are still included unless severe.
Important: Team Coherence is NOT included in the Steradian Score formula. It is a data-quality flag only.
Decision Intelligence
Decision Intelligence is a separate response-spread diagnostic from Domain Alignment. It summarizes how much responses vary at the question level and aggregates spread to the domain grain for interpretation. Any internal "alignment" index produced inside Decision Intelligence for calibration is not equivalent to Domain Alignment as defined on this page (within-level, per-domain max(0, 100 − 4σ) on normalized scores).
The Capability×Alignment matrix uses Domain Alignment as its alignment axis where within-level data support computation. When those data are insufficient (for example, fewer than three respondents at the relevant organizational level for a domain), the matrix shows an em dash (—) for that cell and explains the requirement rather than substituting a different construct.
04 · Operationalization
Operationalization choices
The constructs above are research-grounded. Their specific implementations are Steradian's engineering choices. We name them explicitly here.
Likert-to-100 normalization
5-point Likert responses are normalized to 0–100 using (v − 1) / 4 × 100. This is a standard psychometric transformation.
Vertical alignment formula
For each pair of organizational levels with sufficient data, we compute the normalized Euclidean distance between mean capability profiles across shared domains, then rescale to 0–100 using max(0, 100 − d). The choice of Euclidean distance over alternatives (Mahalanobis, correlation-based) is Steradian's; alternatives are research-defensible.
Horizontal alignment formula
Within an organizational level, we compute population standard deviation across leaders for each domain, then rescale using max(0, 100 − 4σ). The 4σ scaling assumes that σ of approximately 25 on a 0–100 scale represents the maximum practical disagreement; this is a normalization heuristic, not a research-validated constant.
Industry Landscape and Porter's Five Forces
The Industry Landscape domain includes 16 items: the legacy industry-snapshot set plus six stems aligned to Porter's Five Forces — rivalry among existing competitors, bargaining power of suppliers, bargaining power of buyers, threat of new entrants, threat of substitutes — each written as leader-observable industry conditions without requiring framework vocabulary, plus a final synthesis item on structural attractiveness for sustained profitability. Industry structure is a distinct lens from internal cross-leader alignment; we cite Porter's foundational treatment and the updated HBR statement of the model (Porter, 1980; Porter, 2008).
Composite weighting
Combined Alignment uses 50/50 weighting between vertical and horizontal. The equal weighting reflects the literature's position that both dimensions are necessary; the specific equal-weight choice is Steradian's null. Alternative weightings are research-defensible and may be revisited as benchmark data accumulates.
Multiplicative composite
The Steradian Score uses Capability × Alignment rather than additive combination. Multiplicative form reflects the moderating role of alignment in the literature: high capability with low alignment mathematically penalizes to a low readiness score, exposing execution risk that summed approaches would mask.
05 · Data quality
Sample size and data quality
We do not generate reports below n = 3. At or above that floor, every report is produced to the same standard: scores are computed against a calibrated reference cohort and reported as the median alongside the inter-quartile spread — the spread carries as much signal as the median. The diagnostic measures the leadership team that responded rather than extrapolating to a larger population, so a report that clears the floor is board-appropriate on its own terms.
When data is missing for a metric, the metric is reported as "insufficient data." We do not substitute default values for missing measurements.
Cross-Level Alignment headline scores require at least two organizational levels with at least one respondent each. Domain Alignment (horizontal) headline rules still require at least three respondents at an organizational level where measured.
We surface four response-quality diagnostics on every report:
- Acquiescence Bias — tendency to agree
- Extreme Responding — use of endpoints
- Central Tendency — avoidance of endpoints
- Straight-lining — consecutive identical responses
When central tendency exceeds 70%, alignment scores are flagged on page 1 of the report as potentially inflated. When acquiescence exceeds 65%, true disagreement may be higher than reported. When any individual respondent shows straight-lining above 25%, their data is flagged for review. At trial-tier sample sizes (n = 3–4), bias flags are suppressed and replaced with an explanatory note that bias diagnostics require 5+ responses for reliable interpretation.
06 · Evidence
What the research establishes
The two-dimensional alignment framework
Kathuria, Joshi and Porth (2007) review four decades of alignment research and articulate vertical and horizontal alignment as the two dimensions of organizational alignment. Both dimensions are associated with performance outcomes. This framework is the foundation of Steradian's measurement structure.
Strategic consensus and execution
The earlier conceptual paper by Kellermanns, Walter, Lechner and Floyd (2005) in Journal of Management introduced the locus-and-content framework that informs how Steradian measures consensus across organizational levels.
Shared mental models and team performance
Mathieu, Heffner, Goodwin, Salas and Cannon-Bowers (2000) demonstrate that shared mental models among team members correlate with team process quality and performance. This supports our treatment of within-level perception convergence as a meaningful execution variable.
Top management team dynamics and cross-leader comparison
Research on top management teams emphasizes how executives surface and integrate conflicting views under time pressure (Eisenhardt, 1989) and how teams can structure disagreement so it remains task-focused rather than personal (Eisenhardt, Kahwajy, & Bourgeois, 1997). Steradian's Cross-Level and Domain Alignment metrics are statistical summaries of perception across leaders — not transcripts of decision processes — but they are motivated by the same practical question: whether the leadership system sees the same reality.
Walter, Kellermanns, Floyd, Veiga and Matherne (2013) in Strategic Organization found that the consensus–performance relationship is stronger when external strategic alignment is low and weaker when it is high. In other words: when an organization's strategy clearly fits its environment, internal consensus matters less. When that fit is uncertain, consensus matters more.
This is an important boundary condition for interpreting Steradian scores. A high alignment score in an organization with poor environmental fit indicates the team agrees on a strategy that may not work. A high alignment score with strong environmental fit reinforces a strategy that does work.
Steradian currently measures internal consensus; it does not yet measure external strategic fit. See Methodology Roadmap below.
Productive divergence and groupthink
Janis (1972), Amason (1996), Simons and Peterson (2000), and Edmondson (1999) collectively establish that some disagreement is productive and that complete consensus can degrade decision quality. Steradian's diagnostic surfaces variance as information rather than treating it as inherently negative — consistent with this research tradition. Recent synthesis sharpens the point that alignment is not groupthink and that well-structured disagreement can support learning (Bonomi, 2025).
07 · Limits
What Steradian has not yet validated
We name these explicitly because honesty about scope is itself a methodological choice.
Test-retest reliability
We have not yet conducted a published study measuring score stability when the same team retakes the assessment. This is a gap we plan to close with a research partnership.
Convergent validity with established scales
We have not yet correlated Steradian scores with established alignment instruments. We adopt the constructs from this literature; we have not empirically demonstrated our measures correlate with theirs.
Predictive validity for organizational outcomes
The literature broadly supports a positive alignment–performance relationship. We have not yet conducted longitudinal research linking Steradian scores specifically to subsequent organizational outcomes. Score bands shown in reports are interpretive guidance, not validated performance thresholds.
Anti-gaming defense
The current instrument does not include attention-check items, reverse-coded items, or response-style controls beyond the bias diagnostics shown. We treat this as a known gap and a near-term methodological priority.
08 · Roadmap
Methodology roadmap
The following extensions are planned but not currently implemented. We name them publicly because they represent acknowledged limitations of the current product, not gaps we have hidden.
External Strategic Fit
Walter et al. (2013) demonstrate that internal consensus interacts with environmental fit. Steradian currently measures internal consensus; an External Strategic Fit module would measure whether stated priorities respond to actual environmental conditions.
Decision-Rights Clarity
Beer and Eisenstat (2000), Kaplan and Norton (2005), and the practitioner literature on transformation execution treat decision-rights ambiguity as a separate execution failure mode. A Decision-Rights Clarity module is a planned product extension.
Strategic Commitment vs. Understanding
Ates, Tarakci, Porck, van Knippenberg and Groenen (2020) distinguish shared understanding from shared commitment. The current instrument captures understanding (perception scoring) but not commitment (willingness to act). A future revision may add commitment items.
Independent psychometric validation
We are pursuing a research partnership with a peer-reviewed-published methodologist to conduct independent reliability and validity studies on the Steradian instrument. Until that work is complete, we describe our methodology as research-grounded but not independently validated.
09 · References