What PLS-SEM does
Partial least squares structural equation modelling is a multivariate statistical technique used to test structural relationships among latent constructs. It is widely used in marketing strategy research and appropriate for ordinal (Likert-scale) data and complex multi-construct models. PLS-SEM produces path coefficients (the strength of structural relationships), R-squared values (the variance explained), and significance tests for each.
The RevOps Value Chain was tested using ordinal PLS-SEM specifically, which is the appropriate variant for the Likert-scale data the stakeholder survey produced. Ordinal alpha reliability scores exceeded 0.89 across all constructs — well above the conventional 0.7 threshold for acceptable construct reliability.
The principal coefficients
The standardised path coefficient from Resources to Drivers is 0.752. This means that a one standard deviation improvement in the Resources construct is associated with a 0.752 standard deviation improvement in the Drivers construct. This is a strong positive structural relationship.
The standardised path coefficient from Drivers to Outcomes is 0.818. A one standard deviation improvement in Drivers is associated with a 0.818 standard deviation improvement in Outcomes. This is a very strong positive structural relationship.
Both coefficients are highly statistically significant — well beyond the conventional p < 0.05 threshold. The statistical strength matters, but the magnitude matters more: these are large effect sizes for perceptual social-science measures of organisational mechanisms.
R-squared and the indirect effect
Resources explain 56.5% of the variance in Drivers. Drivers explain 67% of the variance in Outcomes. The remaining variance is explained by factors outside the model — market conditions, executive sponsorship, organisational maturity, measurement noise.
The indirect effect of Resources on Outcomes through Drivers is 0.615 — the product of the two path coefficients. This validates the theoretical claim that Resources affect Outcomes structurally rather than directly. Investing in Resources without strengthening Drivers produces weaker Outcomes than the indirect effect would suggest.