Applying the Cynefin Framework to Information Science Research

This report critically evaluates the Cynefin framework as a tool for analyzing qualitative data within the field of information science, using a case study from a research project focused on electronic records management (ERM). The study explores the framework's potential to interpret complex qualitative data, offering insights for practitioners and researchers. The report begins by introducing the challenges of interpreting and presenting qualitative research findings in a way that is readily usable by stakeholders. It then presents the Cynefin framework, which is rooted in knowledge management and complexity science, as a method for making sense of complex situations and guiding appropriate actions. The report discusses the five domains of the Cynefin framework: simple, complicated, complex, chaos, and disorder, and explains how each domain can be used to analyze and categorize ERM issues. The analysis aims to determine how the Cynefin framework can help to identify the most appropriate solutions for specific ERM challenges. The goal is to improve the management of electronic records and make research findings more accessible and actionable for practitioners. The report emphasizes the contextual, contingent, and complex nature of ERM and the need for tailored solutions. This work contributes to the understanding of how to apply complexity science to information science research, making it useful for a broad audience.

Citation: McLeod, Julie and Childs, Sue (2013) The Cynefin framework: A tool for analyzing
qualitative data in information science? Library & Information Science Research, 35 (4). pp.
299-309. ISSN 0740-8188
Published by: Elsevier
URL: http://dx.doi.org/10.1016/j.lisr.2013.05.004
<http://dx.doi.org/10.1016/j.lisr.2013.05.004>
This version was downloaded from Northumbria Research Link:
http://nrl.northumbria.ac.uk/12878/
Northumbria University has developed Northumbria Research Link (NRL) to enable users to
access the University’s research output. Copyright © and moral rights for items on NRL are
retained by the individual author(s) and/or other copyright owners. Single copies of full items
can be reproduced, displayed or performed, and given to third parties in any format or
medium for personal research or study, educational, or not-for-profit purposes without prior
permission or charge, provided the authors, title and full bibliographic details are given, as
well as a hyperlink and/or URL to the original metadata page. The content must not be
changed in any way. Full items must not be sold commercially in any format or medium
without formal permission of the copyright holder. The full policy is available online:
http://nrl.northumbria.ac.uk/policies.html
This document may differ from the final, published version of the research and has been
made available online in accordance with publisher policies. To read and/or cite from the
published version of the research, please visit the publisher’s website (a subscription may be
required.)

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
1. Introduction
One of the challenging steps in qualitative research is interpreting the data collected
and presenting it in ways that enable potential beneficiaries of the research to use it readily
and appropriately. In the information science discipline beneficiaries include both academics
and practitioners with a diverse range of interests. With the heightened attention of
governments and other funding bodies on demonstrating the broader impact of research (e.g.
National Science Foundation, 2013; Research Councils UK, no date), finding appropriate,
tailored ways of presenting or repackaging research results so that they can be used to make a
differences increasingly important. The Cynefin framework presents one way of doing this.
This critical evaluation explores using the Cynefin framework (Snowden & Boone,
2007), which is rooted in knowledge management and complexity science, to interpret a rich
and nuanced set of qualitative data collected from a three-year research project. That project
engaged people worldwide to explore issues and practical strategies for accelerating the pace
of positive change in managing electronic records. While electronic records management
(ERM) is a rather specific information management context, Cynefin has potential as a
research tool in the wider information science discipline.
For organizations, records are “information created, received and maintained as
evidence and as an asset by an organization or person, in pursuance of legal obligations or in
the transaction of business” (International Organization for Standardization [ISO], 2011,
Clause 3.1.7) which need to be managed from creation to disposition. Electronic records,
particularly those that are born digital, are more challenging to manage because they can
comprise a combination of forms (e.g., text, audio, and image), can be distributed across
different systems (e.g., websites and various business systems) and can be dynamic (e.g., an
individual record constructed from different tables in a database).
In the mid-1990s, McDonald (1995) likened the management of electronic records in
the modern unstructured office environment to the wild frontier. Wild because information
technology (IT) was democratizing, decentralizing, individualizing, and personalizing the
way people used and managed information and records in the workplace; and a frontier
because records managers and archivists were questioning concepts and pushing at the
boundaries of knowledge and theory to address the challenge of managing records in the
electronic age.
Corporate rules of the road and other mechanisms have yet to be established in the
electronic world. The wild frontier is unfortunately more the norm than the exception.
In the modern office, it is the office worker, not the technical specialist, who works
with technology applications on a daily basis. It is the office worker, not the
organization, who decides what information will be created, transmitted, and stored.
And it is more often than not the office worker, not the organization, who makes up
the rules, if any. (McDonald, 1995, p.71)
Recognizing the challenge, researchers looked for ways to tame McDonald’s wild
frontier. Seminal research includes that on identifying requirements for managing electronic
records and tactics for satisfying those requirements (Bearman, 1994); on protecting the
integrity of electronic records (Duranti & McNeil, 1996) and maintaining their authenticity
and reliability over time (Duranti & Preston, 2008); and on requirements for recordkeeping
1

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
metadata (Evans, McKemmish, & Bhoday, 2005; McKemmish, Acland, & Reed, 1999).
Notable projects include those of the National Archives of Canada (1991) and Indiana
University (2002). Research has been complemented by the development of many guidelines
and standards (e.g., ARMA International, 2009; DLM Forum, 2001, 2008, 2010;
International Council on Archives [ICA], 2008; ISO, 2001, 2011; State Records, New South
Wales, 2003, 2007; Department of Defense, 1997, 2007), and many commercial electronic
document and records management (RM) systems. However, despite these significant
developments, the management of electronic records continues to be a challenge for
organizations, as evidenced by widely reported security breaches (e.g., unsecured health
records reported to the US Department of Health & Human Services) and failures of large
scale IT systems (e.g. the scrapping of the National Health Service [NHS] national IT
program in the UK).
Reflecting a decade after his wild frontier article, McDonald (2005) felt that, though
some progress had been made, the wild frontier had not yet been tamed. The pace of change
had been relatively slow because organizations do not understand how the office of today
functions, nor how it could benefit from advanced tools for managing work processes and
their associated records. A key inhibitor to progress was managers’ lack of understanding
about records and RM; further, to make progress required a “focus on establishing a vision,
enhancing awareness, assigning accountability, designing an architecture and building
capacity” (McDonald, 2005, p. 8). McDonald’s views influenced the authors of this article to
conduct a research project (AC+erm1) that explored issues and practical strategies with the
aim of helping accelerate improvements in ERM. It provides the data for the study reported
here.
The case of ERM appeared to be complex and the AC+erm project findings needed to
be better understood. The Cynefin framework was then selected to achieve that. However,
ERM is not untypical of the types of systems challenges that information science research
explores; and the rich, nuanced data gathered is typical of that obtained in other qualitative
information science projects.
2. Problem statement
Making sense of research data in a way that enables it to be more readily usable by
practitioners and other stakeholders is one important pathway to ensuring the research
findings can be translated into practice so that the research has impact. A wealth of
qualitative data was obtained by the AC+erm project, covering the experiences and expertise
of a wide range of participants - academics, practitioners, RM leaders. Rich and nuanced, the
dataset comprises an extensive range of ERM issues and problems with associated solutions
that, in the participants’ experience, had worked or not worked(McLeod, Childs, &
Hardiman, 2010).As the challenge of ERM affects all organizations, the potential
beneficiaries of this research are many and diverse. The task was how to enable them to use
the AC+erm results and to adopt or adapt the solutions to improve the management of their e-
records. Though the data were analyzed and presented in a wide variety of forms (e.g. textual
categorized themes, tables of ranked numeric data, phenomenological reflective prose, mind
maps, word clouds, rich pictures, narratives, games) freely available from the project website,
1AC+erm (Accelerating the pace of positive Change in Electronic Records Management) is pronounced āsirm;
the + is silent, indicating only that change is positive. http://www.northumbria.ac.uk/acerm
2

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
anecdotal evidence in feedback and comments from users indicated they were unsure about
how to apply the findings in their own contexts. The presentations were too detailed, too
granular, and used research terminology. A different way of interpreting the analyzed data
and presenting it in a form (or forms) that would enable the findings to be more readily used
by practitioners in their own contexts was needed.
The Cynefin framework (Snowden & Boone, 2007) was selected to undertake a
secondary analysis, based on the nature of the ERM challenge and the research data. Cynefin
has not been widely used as a research data analysis technique or in the information science
discipline. The study presented here addresses the key question:
Can the Cynefin framework be used to further analyze and interpret the research data
and better understand what solution(s) might be most/least appropriate for a particular
ERM issue, given the conclusion from the AC+erm project that tactics and solutions
for ERM are contextualised, contingent and complex?
3. Literature review
The Cynefin framework was developed from research conducted over a period of
years by Snowden and colleagues (Snowden, 2010). It is a framework which helps decision
makers to make sense of a range of business problems and situations, in different dynamic
contexts, and to take appropriate action. Because of this, it appeared to offer an appropriate
approach for making sense of the AC+erm data and linking the issues to solutions to support
appropriate action for change.
The conceptual thinking that underpins the framework draws from knowledge
management and complexity science. Cynefin comprises five domains (Figure 1)
representing the types of situations or environments that organizations typically experience
and need to respond to and manage (Lambe, 2007, p. 134). The domains are predicated on the
construct of order (Snowden, 2005, 2010). The ordered domains are labelled simple and
complicated; the un-ordered ones complex and chaos; and the fifth domain, the central area,
is the domain of disorder. It is important to appreciate that un-order is not lack of order (i.e.,
its opposite), but a different type of order; order that is not directed or designed, but
“emergent” (Kurtz and Snowden, 2003). The characteristics of the domains are summarized
in Table 1 and explained below, based on Snowden and colleagues’ many publications (in
particular, Kurtz & Snowden, 2003; Snowden, 2002, 2003, 2005, 2010; Snowden & Boone,
2007).
Insert Fig. 1. Cynefin framework from Snowden (2010, Pt.7)
Insert Table 1. Summary explanation of the four Cynefin domains: simple, complicated,
complex, chaos
The simple domain is characterized by clear cause and effect. The decision model is
to sense the situation, categorize it, and respond based on best practice. The domain of
efficiency, there is often a right answer; standard operating procedures and process re-
engineering are appropriate practices. Mortgage payment processing would fall in the simple
domain. The complicated domain is also characterized by cause and effect but there may be
multiple right answers. The decision model is therefore to sense, analyze, and respond. This
requires expertise to choose the appropriate answer (i.e., good, rather than best, practice).
Possible practices are systems thinking and scenario planning. Designing a new repository for
research outputs/data falls in this domain.
3

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
Unpredictability and flux characterize the complex domain. Cause and effect are only
understood in retrospect; experimentation is required to find answers. The decision model is
therefore to probe first, then sense and respond; practice emerges. The early strategic
adoption of cloud computing in organizations falls into this domain in the absence of
established best or good practice for implementation. As emergent practice becomes good
practice this example would move to complicated and, ultimately, simple domains.
Turbulence and lack of any link between cause and effect characterize the domain of chaos.
In the absence of any right answers the decision model must be to act first and then sense and
respond, (i.e., crisis management). This can lead to innovative practice; for example, the US
response to the 2010 Haiti Earthquake, where social media technologies were used for the
first time in a crisis as the main knowledge sharing mechanisms (Yates & Paquette, 2011).
The fifth domain, disorder, is where people are unable to decide which of the other
domains represents their situation. This domain can be reduced in size through discussion to
reach consensus about the nature of the situation and the most appropriate type of response,
(i.e., moving to another domain or domains).
The tetrahedrons in Figure 1 are a vital part of Cynefin. They represent the
connections between the center (e.g. managers) and the constituents (e.g. staff). In the
ordered domains (simple and complicated) connections between a central manager and staff
are strong. In the unordered domains (complex and chaos) they are weak. Differences in the
connections represent different work patterns: co-ordination (simple), co-operation
(complicated), collaboration (complex) and directive intervention (chaos).
No domain is more desirable than another: They just describe the situation facing the
organization.
Snowden (2010, Pt.1) notes that the early versions of the Cynefin framework were
based on ideas from knowledge management; for example, the Information Space (I-Space)
model for understanding information flows (Boisot & Cox, 1999); the SECI model
(socialisation, externalisation, combination, internalisation) where the interaction between
tacit and explicit knowledge produces a spiral of knowledge creation (Nonaka & Takeuchi,
1995); and organizational learning cultures (Senge, 2006). As the Cynefin framework
developed, Snowden (2010, Pt.2) brought in the aspect of decision-making, drawing on ideas
from complexity science, particularly the concept of complex adaptive systems (CAS).
Complexity science (Burnes, 2005; Stacey, 2011) was developed by researchers in disciplines
working with natural systems and briefly comprises three key concepts: (a) chaos theory -
some dynamic systems are non-linear, demonstrating complex patterns that are not directly
proportional to, nor predicted from, their causes/inputs; (b) dissipative structure theory - some
systems can pass through states of instability/randomness to new organized states by self-
organization; and (c) CAS - a system comprising a large number of entities interacting with
each other, following local principles and rules, from which emerges a self-organizing group-
wide pattern not determined by the entities, the emergent patterns, or anything outside the
system. The ideas of complexity theory have been used by many authors to study
organizations, based on the argument that organizations are complex, non-linear, self-
organizing systems. The Cynefin framework incorporates the “metaphorticians” approach
(Richardson, 2008) to the application of complexity theory to organizations, using these ideas
as a different lens to view sense-making and decision-making. A number of authors have
criticised the way that complexity thinking has been applied to organizations (Burnes, 2005;
Mingers & White, 2010; Stacey, 2011; Zhu, 2007). Stacey (2011) proposes a more innovative
adaptation of the CAS idea - a complex responsive processes perspective. He recommends
4

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
that managers should use existing tools and techniques, such as the Cynefin framework, but
in a reflexive way, exercising practical judgment, and accepting that they cannot fully control
how these tools and techniques will work out in a specific, real-world situation (Stacey,
2012).
Cynefin can be used in different organizational contexts and for different purposes;
for example, to gain new insights on a challenging problem or contentious issue, to plan
actions to move a situation from one domain to another, or to consider strategies for
managing different situations (Kurtz & Snowden, 2003).In management science it has been
used primarily in the context of decision-making and leadership (Benson & Dresdow, 2009;
Gonnering, 2010; Moerschell & Lao, 2012; Snowden & Boone, 2007). In health it has been
used in a variety of contexts; for example, choosing approaches to health promotion (Van
Beurden, Kia, Zask, Dietrich,& Rose, 2013), developing research in organizational behavior
(Mark, 2006), analyzing chronic care (Martin et al, 2011), and understanding knowledge in
clinical practice (Sturmberg & Martin, 2008).
There are only a few published uses of Cynefin in information science research. These
relate to information system design and information architecture (Burford, 2011; Lambe,
2007; Snowden, 2001); types of learning supported by an Internet portal (Cronje & Burger
(2006); understanding what enhances productiveness in knowledge generation in science
(Van der Walt & de Wet, 2008); and evaluating library services (Hart & Schenk, 2010).None
of these examples uses the Cynefin workshop technique to analyze empirical qualitative data
in the way this study does.
4. Procedures
4.1. The data
The study uses data from the AC+erm project, a major research project (2007-10) that
explored the design of an organizational-centred architecture for managing electronic records.
AC+erm considered three facets: people, processes, and technology. The issues and problems
of ERM were investigated and examples of ERM strategies, tactics, and practice were
gathered, analyzed and shared. An aim was to produce practical strategies for the
contemporary work environment that were scenario-based rather than organizational-based;
presented issues as well as solutions; were capable of being used in practice, as well as
facilitating discussion and debate; and would produce change.
Recordkeeping in the e-environment involves four stakeholder groups:
executives/senior managers, records professionals, IT/systems administrators and
recordkeepers (ISO 2001, 2011). It also transcends disciplines and sectors. The project
therefore adopted a trans-disciplinary approach and obtained expert opinion from all four
stakeholder groups.
The project’s qualitative methodology comprised three phases: (a) a comprehensive
systematic review of relevant literature published from 1997-2009 to identify ERM issues
(Centre for Reviews and Dissemination, 2008); (b) an investigation of the three facets; and
(c) a major dissemination activity running throughout its life. The investigation phase used a
combination of electronic Delphi studies (gathering expert opinion on a global basis) and
face-to-face colloquia (enabling in-depth discussion on a local (UK) basis). The Delphi
5

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
technique, developed in the 1950s, gathers expert2 opinion on a topic through several rounds
of questions, usually to reach consensus (Dalkey & Helmer, 1963). It is used for prediction,
problem-solving, and policy development (Linstone & Turoff, 2002; McLeod & Childs,
2007). In this project it was used to refine and discuss the issues identified from the literature
review, rank them in order of importance, propose and discuss solutions to the issues, and
then evaluate the solutions.
The Delphi study participants comprised 55 different stakeholders involved in ERM,
from different disciplines and within organizations working in different sectors. They
responded based on their own knowledge and experience, providing real-life examples where
possible. The participants' responses were analyzed using a range of different approaches to
provide a broad view of the data.
A series of headline findings emerged (McLeod, Childs, & Hardiman, 2011). Many of
these relate to people issues rather than processes or technology, for example:
• The people, process, and systems/technology aspects of ERM are inextricably linked.
• People issues are predominant, fundamental, and challenging.
• Tactics and solutions for ERM are contextualized and complex.
• The success and/or failure of ERM implementations can be contingent on the
presence/absence of small or accidental factors.
This resulted in the conclusion that the ERM challenge is complex, contextualized, and
contingent. To help practitioners choose which solutions to try in a particular circumstance
required secondary analysis using a different approach, and the Cynefin framework was used
to do this.
A subset of the AC+erm data was used. It related to the people issues and solutions and
was collected from the systematic literature review and all three Delphi studies. Though each
Delphi study focused on one facet of the research (people, processes, and technology), people
issues and solutions were identified from all three studies, not just from the people Delphi.
The Delphi data are the participants’ experiences, which represent multiple organizational
contexts.
The data comprise the themes from the first order analysis of the raw data. The literature
themes were created using open coding (the allocation and grouping of codes), a technique
based on constant comparative analysis (Strauss & Corbin, 1998). The Delphi themes were
created using a robust faceted analysis of the data (Broughton, 2001) enabling the creation of
structured not isolated codes, retaining sufficient context of the original. Figure 2 provides
examples of both types.
Insert Fig. 2. AC+erm project data on issues - Examples of individual themes
The data is rich and extensive, the theming is nuanced. In total there were 446 themes:
128 from the systematic literature review and 318 from the Delphi studies. Of these, some
were duplicated across the different Delphi studies and the literature. Morse (1994) identifies
four cognitive processes that belong to all qualitative analysis methods, comprehending,
2There is debate about the definition of “experts” but the definition adopted here is “those who have an
applicable specialty or relevant experience” (Linstone & Turoff,2002, p. 65).
6

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
synthesizing, theorizing and recontextualizing. Synthesizing is the process of merging the
data into patterns which then enables theorizing, wondering about the reasons for these
patterns. It enables the data to be interpreted and provides insights. The theming of the data
had enabled comprehension, however additional synthesis was needed to enable more
effective translation of the research findings into practice.
4.2. Rationale for using the Cynefin framework
Problem structuring methods (PSMs) are used in complex, problematic situations
(Mingers & White, 2010, p.1151). They are interactive, participatory modelling approaches
to help groups reach a common understanding and consensus about the problem confronting
them and how to tackle it. Soft systems methodology (SSM) is an example of a PSM and
extensive literature exists on its use. However, SSM was not suitable for the secondary data
analysis required here. The rationale for adopting the Cynefin framework relates to the nature
of the ERM challenge and of the AC+erm data; the inspiration for using it came from a PhD
student at Northumbria University who used it in the context of a co-operative action research
project (Childs, McLeod, & Hardiman, 2009).
Cynefin has many of the characteristics of a PSM and resonates with the problem the
AC+erm project set out to explore. Also, Cynefin has roots in knowledge management (which
is closely related to RM and information management), has been developed to address critical
business issues (which ERM and digital information management are), and aims “to support
decision making in varied, dynamic contexts” (which is the case for ERM) (Kurtz and
Snowden, 2003, p. 462). The headline findings from the AC+erm project (Mcleod, Childs, &
Hardiman, 2011) highlighted the complexity of the situation, the focus on people issues, and
the huge number and range of issues and solutions identified. As well as making-sense of the
AC+erm data, the hope was that Cynefin could provide guidance to practitioners in their
decision-making in a dynamic and uncertain situation, to help them develop a strategic
approach to ERM, and to guide them to select from the project’s “toolbox” of practical
solutions ones that would be worth trying in their specific context.
4.3. Method
Snowden (2010, Pt.5) has developed a range of techniques for deploying the Cynefin
framework, including the “four points” method. A social construction approach, it comprises:
• a pre-process in which items are collected about the particular issue or topic of
interest or concern (these form the data, or narratives, for sense-making and can be
events, points of view, anecdotes, etc.; Kurtz & Snowden, 2003);
• a workshop in which a representative group of people place the items in the Cynefin
framework (without prior explanation of it) and both the definitions of the domains
and the boundaries between them emerge from the data; and
• a post-process in which the resultant contextualised Cynefin framework of the issue
of interest is used (e.g., within a training program, for planning, or for more detailed
discussion).
A variant of this method was used with the AC+erm data (these variations are justified
below):
7

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
• In pre-process, the existing themed data relating to people issues were used as the
narratives.
• In workshop, the people taking part (the authors of this article) were two of the
researchers from the original AC+erm project. Before the workshop they discussed
and agreed their understanding of the four points method and the four domains; there
was no facilitator. The four domains were labelled (with descriptive notes) and placed
at the four corners of a table. First, the authors each took a portion of the themes and
independently, in silence, placed them either in one of the four domains; on a
boundary between domains; or in the central, fifth domain - disorder. Second, the
authors reviewed, by discussion, the themes in each location to reach consensus on
their allocation and some themes were reallocated. Over 90% of the themes were
allocated easily, but a small number were challenging. The themes were not modified
nor were new ones created, as is possible with narratives in the four points method.
During this discussion, themes in each location were grouped by subject, or meta-
theme. Each meta-theme was labelled, and a scope note added where necessary. This
was equivalent to the step “define domains and boundaries in language that is
understood within the organization” (Snowden, 2010, Pt.5). Third, the allocation of
themes was finalized by taking a holistic view and reviewing each domain, and
border, in turn (simple > complicated > complex > chaos).
• Post-process comprised the publication of the results in the form of the resultant
framework, as an example for peer review, comment, adoption, and adaptation to
address the people issues of ERM. This dissemination represents a preliminary step in
the exploitation > exploration > exploitation stages referred to by Snowden (2010,
Pt.5). The ERM framework can be used by others in teaching, research, and practice.
It has the potential to become part of the discourse for addressing ERM situations. If
practitioners find that the ERM framework is not suitable in their own context then
they can use the four points method themselves with their own narratives.
To evaluate the use of Cynefin, the authors reflected individually, and together, on the
process (e.g., what changes had to be made to the four points method and why; how easy was
this method to use; could it be easily understood/used by RM practitioners; was it useful as a
data analysis tool; could it have wider research uses) and its usefulness (e.g., what new
insights were obtained; were original findings confirmed or altered; did it provide an
effective way of linking the solutions to the data in a way that RM practitioners could use).
5. Findings
Figure 3 and Tables 2 and 3 summarize the results of mapping the themed people issues
into the Cynefin framework. Figure 3shows the number of themes placed in each domain and
on the boundaries, highlighting that the majority (58%) are simple or complicated (the
ordered domains). Almost a third of the issues are complex (32%), a similar proportion to the
simple issues. Few themes fall in the chaotic domain (2%) or on the boundaries
(7%).Ultimately, none remain on the simple/complicated or the simple/chaos boundaries; the
majority are placed firmly into one domain or another. There is a wide variety of individual
themes in any one domain, highlighted in the examples provided in Table 2.Some of the 25
meta-themes that emerged from grouping the themed issues during the mapping are
illustrated in Table 3, together with their nuanced scope. Table 3 also includes potential
solutions to these issues, drawn from those provided by the Delphi participants, and
represents the populated Cynefin framework for the ERM people issues and solutions.
8

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
Insert Figure 3: Summary results of mapping the people issues data in the Cynefin
framework domains
Insert Table 2.Examples of the range of issues in each domain
Insert Table 3. The populated Cynefin framework for the ERM people issues and solutions
6. Discussion
6.1. New insights
Given the AC+erm project conclusion that the ERM challenge is complex,
contextualized, and contingent, the most surprising finding is that the majority (58%) of the
people issues fall within the simple and complicated domains. Although the simple domain
should not be confused with easy or quick solutions, it is the domain where there is a right
answer; the complicated domain offers potentially multiple right answers, selected based on
expert knowledge. Best practice and good practice are, respectively, the appropriate
responses and, for ERM, these exist in abundance, as highlighted in the introduction. Perhaps
the sheer number of simple and complicated issues makes ERM appear complex—a case of
drowning in the size of the challenge.
However, complexity is not only due to the volume of simple and complicated issues;
a large proportion of issues (32%) are complex. Using Cynefin reveals the truly complex
issues, such as the attitudes and perceptions of the different stakeholders (Table 3). As Lambe
(2007) noted, “in Cynefin terms, pretty much anything to do with human affairs resides in the
Complex domain” (p. 143). People issues are challenging because they concern culture,
worldviews, and preferences and behavior related to use of RM/ERM systems. From their use
of the Cynefin framework, the authors have re-perceived the ERM challenge as a wicked
problem (McLeod & Childs, 2013). A wicked problem (Rittel & Webber, 1973) has many
causes and many solutions, with different definitions of what these are from the different
stakeholders involved. Causes and solutions are inextricably linked: Deciding on a solution
then determines the cause of the problem. There is no right answer, and standard problem-
solving approaches are not effective.
Cynefin identifies what approaches are appropriate for the domains. For example,
responses to the simple issue of lack of awareness of RM/ERM and what it comprises include
building RM into the induction program for new employees and holding awareness-raising
sessions, events, or activities for existing employees (best practice). An appropriate response
to the complicated issue of ERM systems design is to “match them to work processes” (good
practice). However, is training, which is a good or best practice approach and suitable for
simple or complicated issues, appropriate for the complex issue of attitudes and perceptions
of managers and staff? In the context of chief executive officers’ (CEO) lack of awareness of
RM/ERM and lack of recognition of its value, some participants said yes, others said no.
Some participants said CEO training should be long-term and subtle, which seems to be
about influencing and marketing, not training, per se. Marketing individual benefits and
managing expectations of ERM systems are suitable probes to attempt to achieve the
9

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
emergence of recognition of the value of RM/ERM and better recordkeeping behaviors.
Therefore, marketing is an appropriate solution to this complex issue.
In linking issues to solutions, a many-to-many (not a 1-to-1 relationship) is highlighted
between them (i.e., for each issue there are many solutions and each solution can resolve
many issues). Cynefin prompts reflection on whether or not the solutions suggested by the
Delphi participants are actually appropriate: As well as giving solutions that in their
experience had worked, they were asked to give solutions that had not worked and therefore
should be avoided. Therefore, further analysis of the “solutions to avoid” is required to
discover if they should be avoided because they are innately inappropriate, given the nature
(domain) of the issue, or due to the contingency of success, since solutions that work in one
organization do not necessarily work in another.
In discussing seemingly similar themes that initially were placed in more than one
domain, particularly in both the simple and complex domains, it became apparent that the
nature of the themes was different. For example, lack of awareness of RM/ERM falls into the
simple domain, whereas lack of recognition of the value of RM/ERM falls into the complex
domain. In the previous analysis these two had been grouped together, the nuances hidden.
Cynefin reveals the nuances and helps to decouple conflated issues.
Cynefin also provides new consideration of the importance and nature of the
management pattern and connections between stakeholders required to address the ERM
challenge. Centrally controlled management of records through established best practice
procedures and good practice systems thinking has operated successfully in the paper world,
and could be successful for the simple/complicated aspects of ERM. However, it is unlikely
to be adequate for addressing the many complex issues of ERM.
6.2. Benefits of the new understanding
A key benefit of understanding the nature of the issues, from their Cynefin domain
location, is that it enables the appropriate decision-making model, action(s) and management
approach to be identified and used, as Van Beurden, et al.(2013) suggest. For the AC+erm
data this has lead to a conceptual and strategic mapping of the many issues and a clearer more
coherent approach to identifying the appropriate solutions for practitioners to use. The
resultant ERM framework (Table 3) provides the ability to focus on individual issues as well
as a holistic way of interpreting the data based on the nature of the issues.
This evidence-based framework example can be directly exploited in practice,
teaching, and research. It has the potential to become part of the discourse for addressing
ERM situations. Additionally, if practitioners find that it does not fit their organizational
context, they can use Cynefin to explore their own ERM “narratives”.
In the digital environment, where “the autonomy of individual reigns supreme!”
(McDonald, 1995, p. 70), adopting the appropriate management approach is vital for
successfully implementing a solution. Cynefin identifies these (coordination, cooperation,
collaboration, and directive intervention) along with the connections required between
managers and other staff. Coordination and cooperation can continue to be used to manage
ERM issues that fall into the simple and complicated domains; however, they are unsuitable
for addressing the many complex issues. Here, records managers must build strong
connections with staff and other expert stakeholders and collaborate, rather than coordinate or
seek cooperation.
10

The Cynefin framework: A tool for analyzing qualitative data in information science?
Library & Information Science Research
An important element of the Cynefin framework is the concept of dynamics. It
highlights that issues might be located in different domains in the future. For example, what
is good practice today might become best practice in the future; what today is emergent
practice might become good practice (as in, for example, cloud computing implementation).
Although this was not considered in great detail in using Cynefin to synthesize the data, it
raises important questions for further research.
6.3. Reflection on the use of Cynefin
In previous studies, such as Burford (2011), Lambe (2007) and Van Beurden, et al.
(2013), researchers have used Cynefin in various ways (e.g., to structure findings and
discussion; to draw conclusions; and, to some extent, as an explanatory theory), but in this
article Cynefin is used in a new way: as a research tool.
In adopting Cynefin as a research tool, modifications were made that appear
legitimate for this purpose. A key modification was that the workshop was not focused on
narratives from a single organizational context but on themed data from many real
experiences and organizational contexts. In the standard pre-process stage, workshop
participants are asked to generate “several hundred examples of exemplar narratives of key
moments in the organization’s own history, alternative histories and imagined futures”
(Snowden, 2010, Pt.5). The AC+erm themes developed from the faceted analysis of Delphi
responses are based on the Delphi participants’ real-life experiences and examples. As the
faceted analysis retains the context and some of the original language of these respondents, it
is justified to consider them as being similar to those used in the standard Cynefin pre-
process. Although this is not true for the themes produced from the systematic literature
review, these were all agreed and/or amended by the participants at the beginning of each of
the Delphi studies. These themes were not modified (nor were new ones created) during the
workshop process, as is possible with narratives in the four points method; the authors
remained true to the data collected and analyzed from the literature and individuals, though
their meanings were sometimes debated. However, not adding to the collection of narratives
is not a significant modification.
Another modification was the result of not having a facilitator for the workshop. The
authors therefore used their existing knowledge of the nature of the domains to replace the
knowledge that the facilitator would have provided during a workshop;(e.g. initially agreeing
the understanding of domains and reviewing each one to agree the categorization of
individual themes).A further modification was the language used, which is supposed to be
that understood by the organizational participants in the Cynefin workshop. The nature of the
language used here was based in the literature and on that of the Delphi participants from
their multiple organizational contexts, and from their experiences and perspectives as one (or
more) of the four RM stakeholder groups. The authors themselves fall into the records
professionals and record-keepers groups.
In the role of a data analysis tool, Cynefin is not suitable for use by a single
researcher, as sense-making is considered to be a social process (Kurtz and Snowden, 2003).
However, it can be used successfully with as few as two researchers, as demonstrated here.
6.4. The potential of Cynefin in information science research
11