*3.2.2 Stage two: extracting sustainability data*

In stage two, coding rules relating to the three pillars of sustainability (i.e., economic, environmental, and social) were developed using the global reporting initiative (GRI) G4 reporting criteria as a guide [50]. The GRI is an international independent non-profit organisation founded in 1997 in association with the United Nations that has published a comprehensive set of economic, environmental and social sustainability performance standards considered globally to be the basis for best-practice sustainability reporting [51]. Due to its function as a leading


**Table 1.** *Summary of sample firms.*


#### **Table 2.**

*Knowledge management variables and coding rules.*

international benchmark, the GRI has served as the basis for content analysis coding firm sustainability strategies and activities in a range of academic research disciplines [52–54]. All annual report data gathered in Stage One were entered into a codified database using the GRI definitions. Coding refers to the process of converting data into a contextual value for the purposes of storage and thematic analysis [54]. The coding was facilitated using the NVIVO (version 12) software package, which is a computer-aided qualitative data analysis tool that researchers use to organise data and conduct thematic analysis. The data were coded in a 'line-by-line' manner according to the process prescribed by [55].

### *3.2.3 Stage three: applying a knowledge management lens*

In stage three, coding rules for detecting data relevant to KM were developed (see **Table 2**) by adapting the operational definitions of KM elements provided by [28], p. 319 that were described in the literature review section. The data were then entered into a codified database using the same techniques as described above for stage two. The data emanating from stage three of this content analysis process for the basis of the findings that follow.
