Sally Burford, Faculty of Communication and International Studies, University of Canberra, ACT, Australia, 2601. sally.burford@canberra.edu.au
The design and structure of the information on any web site, that is, its Information Architecture (IA), is an important criterion in the successful use of the web. The online information ‘space' should allow intuitive and easily navigable access to useful information. Whilst there is now a well defined process for designing the information structures on large information-rich web sites, few management models exist for achieving this within organisations. This paper outlines a research approach to identify the organisational factors and strategies that support web IA processes. Grounded theory and multi case study methodologies are used to study the complex, social environments in which web IA outcomes are achieved. In constructing a conceptual picture of how organisations are practicing web IA, this research will establish a deeper understanding of the contextual processes involved and how they might be employed more effectively. The research findings to date locate responsibility for IA in organisations, identify IA as new work and propose that IA be considered as knowledge work.
In recent years, organisational use of the web for providing vast quantities of information has become mainstream and much attention has focused on optimizing a process for the design of online information spaces (Morville and Rosenfeld 2006; Garrett 2003). As a result, a discipline of web Information Architecture has emerged – to the extent that the Australian Government Information Management Office now provides best practice guidelines for practicing web IA titled Information Architecture for Websites Better Practice Checklist (2004). Whilst best practice in IA methodology is sometimes known in organisations, few management models exist for achieving it. IA enabling factors such as skill sets, best practice in staffing, organisational structure and resourcing are generally not embedded in organizational corporate knowledge.
Against a backdrop of rhetoric about the importance of the web as major media for information provision for the clients of large organisations, it is relevant to ask why some organisations are not building web sites with effective information architectures. What are the issues that disconnect an organisation from achieving an effective IA for their web site? Conversely, what is the optimal environment in which web IA might flourish? What factors contribute to an organisation's capacity to develop effective web Information Architecture for public- facing, information-rich web sites? And how do organizations facilitate the use of best practice guidelines and processes for web IA? These questions provide a backdrop and impetus for this research enquiry.
This paper describes a research approach that explores this recent and important phenomenon that has, to date, been little researched and goes on to present preliminary findings. Grounded theory and multiple case studies are used for this exploration and the emerging framework of a more comprehensive understanding of the way in which organisations achieve an effective web IA is outlined.
The formalisation of the practice of web IA and its documentation have been driven largely by practitioners in the area (Fast 2006; Campbell 2007). An abundance of short papers supporting and reporting the work of IA exist in online magazines such as Boxes and Arrows and Digital Web Magazine . Rosenfeld and Morvilles' book Information Architecture for the WWW, now in its 3 rd edition, has proved a pivotal work in defining an IA methodology and a seminal textbook in IA education. Whilst the process of developing web IA is well documented by professionals in the field, the recommended practice is often disconnected from an organisational context and research is scant. In 2006, Fast (p. 1 of 3) considered ‘that IA is characterized by its practice: not by its research'. Research literature (eg. Sinha and Boutelle 2004) is beginning to emerge but is piecemeal and usually devoted to aspects of the IA process rather than an organisation's ability to enable it.
In introducing a special edition of the Journal of the American Society for Information Science and Technology on IA, Dillon (2002, p.821) proposed a broad definition of IA. ‘IA is the term used to describe the process of designing, implementing and evaluating information spaces that are humanly and socially acceptable to their intended stakeholders'. Dillon begins the consideration of IA in the organisation and sees business and organisational aspects of IA encompassed by ‘human and social acceptance'.
Using the term IA in a much more expansive way, Evernden and Evernden (2003) consider all of an organisation's electronic information as information in need of architecting and report the field of information architecture as one that has been in existence for some decades. Evernden and Evernden (2003, p.24) then present the term ‘the webbist movement' to situate IA for the web in this larger picture.
Evernden and Evernden (2003) ask us to consider that every web site does indeed have an information structure – be it optimal for its audience or a frustrating and unsuccessful experience that will detract from the organisation's business goals. Deliberate attention to the crafting of information structures to meet the information seeking behaviour of the business user is what the process of information architecture is all about. Evernden and Evernden (2003) challenge organisations to accept that doing IA requires expertise and a mindfulness of the hidden costs in an unplanned, impromptu approach to IA, as other competitive organisations build capability and expertise in architecting information.
Morville and Rosenfeld (2006, p.333) exhort practicing IAs to ‘make the case' for IA within organizations and admit that IA is often not acknowledged because it is ‘abstract, intangible and new, and each situation demands a unique solution'. The authors suggest that the promotion of IA will be easier as organisations feel more and more ‘information pain' and note that it is a process and skill set that suffers from invisibility.
Eschenfelder (2003) examines the conflict involved in developing a web IA for a large organization. She focuses on a significant IA component – web classification schemes (WCS). Acknowledging the importance of organisational web sites as interface to customers and integral to achieving business goals, Eschenfelder notes that a site must often serve multiple customer profiles with differing needs and expectations. This leads to goal conflict between organisational sub units with different target customers. This study provides some understanding of the organisational conflict involved in the development of an IA for a public-facing web site and offers the ‘beginning of a practice-based methodology to predict and proactively manage conflict in website design and management' (p.435) claiming that with planning, conflict could be used to enhance the process of IA in a large organization.
A short study of the work practices of IA was carried out by Busch-Geertsema et al. (2005). The authors' rationale for the study was to understand and clarify IA working processes from a practitioner's point of view and to contribute to a model of the work of information architects. A contextual analysis of the work of IA was reported in the form of two fictitious practitioners and their diverse approaches to IA. The authors point out that the work of IAs has been shaped by the environment in which they practice and, implicit in the scenarios presented, they in turn shape that organisation's approach to IA. User centred design is less prevalent in the low budget, low IA focused development of web sites. This study highlights the impact of organisational readiness to acknowledge, fund and use the best practice of IA.
Evernden and Evernden (2003) present a conceptual framework for the work of IA in organisations. They present a set of eight factors that need to be taken into account when developing any information architecture. These eight factors are: Categories, Understanding, Presentation, Evolution, Knowledge, Responsibility, Process, Meta Levels. Any or all of these factors can be applied to a particular IA, argue the authors, and thus become a ‘dimension' of that IA. Whilst Evernden and Evernden's eight factors address the need for organisations to own, manage and know IA, they are presented as ‘an integrating frame thought' (p.40) without strong process or methodology and provide a parallel conceptual world to the practical, procedural approach of Rosenfeld and Morville (2006). A synthesis of these two approaches to support organisational web IA is needed.
Evernden and Evernden's eight factors or framework for enterprise IA in organisational context makes valuable input to the practice of web IA in organisations but is not empirically based. Evernden and Evernden have proposed these eight factors from experience in architecting information and theories from diverse disciplines – they are not the result of research.
Grounded theory – a term that is used for both the mode of inquiry and the outcome of the research (Charmaz 2005) is used as the research approach for this inquiry. Grounded theory was first ‘discovered' by Glaser and Strauss in 1967 as a research approach that insisted that the theory put forward by qualitative researchers was firmly grounded in the data collected during the investigative processes. There is a strong emphasis on theory development in this methodology ie. the creation of theory rather than testing of theory. A grounded theory is one based on concepts rather than description. It provides a systematic and explicit process for conceptualisation from data.
Charmaz (2005, p.508) describes grounded theory thus:
A grounded theory approach encourages the researcher to remain close to their studied worlds and to develop an integrated set of theoretical concepts from their empirical materials that not only synthesize and interpret them but also show processual relationships.
This study of web IA in organisations has followed a constructivist epistemology and will be positioned in Charmaz's 21 st century ‘constructivist grounded theory'. Constructivist grounded theory has enabled the study of IA processes within the enterprise to be taken into a social realm – involving people in their organisations – as well as the complex interactions involved in achieving an appropriate IA for an organisation's web site. Glaser and Strauss (1967, p.3) argue that grounded theories are more likely to ‘fit and work' than theories that might be proposed and tested. For this research into a recent phenomenon, there is a greater chance that an emergent theory will ‘fit and work' and prove useful to those responsible for creating web IA in organisations.
The grounded theory methodology has been tightly integrated with a case study approach. A number of large organisations that use the web as a significant information delivery platform to communicate with and service their clients will be investigated as cases or units of analysis. The criteria for selecting an organisation for this research are that it has a web site that is information-rich and public-facing. In order to establish that the organisation is of a considerable size, only those with over four hundred employees will included.
The case study is defined by Yin (2003) as an empirical inquiry that investigates a contemporary phenomenon within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident. ‘In general, case studies are the preferred strategy when ‘how' or ‘why' questions are being posed, when the investigator has little control over events, and when the focus is on a contemporary phenomenon within some real-life context' Yin (2003, p.21).
This research meets all of these theoretical conditions. The investigation is of a contemporary nature – the phenomenon of organisations providing information on the web is less then two decades old, during which it has grown in importance and continues to do so. The focus on the skill and professionalism of the process of web IA is even more recent. The ‘real-life context' of web IA is the key research focus – how IA is actually done in organisations is explored as opposed to an idealistic or best practice process for doing Web IA. The way that IA is done is tightly coupled to an organisational context – many organisational factors and variables that are not known, defined nor controllable will be explored in this research. ‘Case studies can be useful for exploring new or emerging processes or behaviours' and understanding ‘how behaviour and/or processes are influenced by, and influence context' (Hartley 2005, p.323). Complex interacting constructs come into play in the conduct of Web IA in an organisation – politics, resourcing, awareness, skills to name but a few.
In merging the methodologies of grounded theory and multi case study, this investigation generates theory from the analysis of the data collected in each organisation. Hartley (2005) supports this approach saying that ‘case studies have an important function in generating hypotheses and building theory' (p.325). Eschenfelder (2004) applied an exploratory multi-case study approach and a grounded theory methodology to the study of the production of textual content for American government web sites using the organisation as the unit of analysis.
The development of theory during the research process is described by Strauss and Corbin (1990) to be a continuous interplay between data analysis and data collection. Analysis begins after the first data set has been collected and from that point the ‘ grounded theory methods consist of simultaneous data collection and analysis, with each informing and focusing the other throughout the research process. As grounded theorists, we begin our analyses early to help us focus further data collection. In turn, we use these focused data to refine our emerging analyses .' (Charmaz 2005, p.508)
This research approaches data collection and analysis using this key tenet of grounded theory, t heoretical sampling (Strauss and Corbin 1990) . Theoretical sampling directs the process of continual collection of data throughout the investigation. As concepts and relationships between those concepts emerge from the data, the investigator seeks more specific and focused data to flesh out, refine and validate the emerging concepts, categories and relationships. Collection of data and its analysis continue until theoretical saturation (Strauss and Corbin 1990) occurs ie. until the analysis of new data no longer informs the emergent theory.
Analysis of the qualitative data collected in this research began with coding. Initial coding is open and involves very close and detailed scrutiny of the data (Charmaz 2006). Codes emerge as the data and its meaning is studied. During open coding, the researcher must remain open to any and all concepts that emerge from the data – the analyst must not be influenced by preconceived theoretical directions. ‘Initial codes are provisional, comparative and grounded in the data' (Charmaz 2006, p.48).
With the analytic direction from open coding, the second form of coding, focused coding, takes place by examining the most significant and frequently used open codes. These are tested and used across a large data set to examine their ability ‘to categorise the data incisively and completely' (Charmaz 2006, p.57). It is here that the process of constant comparison (Glaser and Strauss 1967; Strauss and Corbin 1990) of data against data and concepts against data occurs. If the purpose of open coding is to fracture the data, focused coding is an analytical pursuit that begins to draw a coherent picture from the pieces (Charmaz 2006).
A qualitative approach to data collection is used throughout the research. The first method employed was to ask those people with responsibility and involvement in IA to tell the story of how it is done within their organisation. These group narratives were analysed and followed up with relevant semi structured interview questions to further the development of theory. The data (the narrative and interview transcripts) were coded to allow concepts and their relationships to be established. The research also examines web IA documentation that exists within the organisation – be it policy, process or best practice documents.
The authority and responsibility for web IA can be ill defined in large organisations. It is complicated by the different components of IA on a web site and the sense of ownership of content structures by the business stakeholders. The work of IA carried out by central web teams in this study can be divided into two categories. The first is where the central web team has responsibility and some authority. This was found to be the global utility navigation (eg. search engine access, site indexes and privacy statements) and the page layout that generally appears on every page of the organisation's site. In two of the three organisations studied, these IA design decisions are made by the web team without the backing of governance or policy. Authority and responsibility for these global aspects of the IA of an organisation's web site are enacted technically in all of the organisations studied in the use of a content management system or static page templates that prevent changes to global utility navigation and page layout.
This study reveals that, as the status and role of the web team matures within an organisation, the team has gradually assumed responsibility for the global navigation and standard page layout of the organisation's web site. Much of the authority for IA has been claimed, grasped or earned by the web team over a period of time. In discussing the web team's responsibility and authority for global IA one web manager made the following comments that exude a certain confidence as well as an awareness of the risk that is involved in the assumed authority and action.
We actually just designed it and just put it out there, we didn't do the consultation thing and everyone's happy - as far as we can tell.
Yes we just did it, yes. I think partly because we've had a change of manager in this group in the intervening period the previous manager was more about consultation – I'm more about lets just get it done and apologise later.
We are left to ourselves, we strangely enough make sensible decisions most of the time.
In all organisations studied, there was an overall sense of responsibility for the organisation's homepage by the central web team but full authority had not been assumed and changes in the content structures required negotiation or approval with business stakeholders and in the case of one organisation, a formal approval process. This negotiation was more of a political nature than the design process of IA.
The second proposed category of IA work is the unique, ambiguous taxonomy that must be constructed to organise the content for each sub site of the organisational web presence – a site that is ‘owned' by a business unit. In the three organisations studied, central authority and responsibility did not exist for the development of content structures of web sites that serve the needs of devolved business units and their clients. A business unit is typically unskilled in developing effective content structures for their web site and an organisation will often employ a person whose title is information architect . Yet an information architect cannot work in isolation - ‘a silo of expertise' (Cox 2007a, p.776) will achieve very little. Centrally employed information architects must interact with business units to achieve effective outcomes. It is here that the IA capability of the web team must find a collaborative and integrative approach, working alongside the business units who are the knowledgeable owners of the information that is provided on the site. The central IA expertise forms part of a team in the redevelopment of a sub site.
Centrally employed information architects are often in a position of working collaboratively, encouraging, transferring their skills and mentoring the people in business units to do the work of IA as effectively as possible. This is consistent with the findings of Cox (2007a) who writes that ‘this work was carried out in a context where there is little direct formal power' (p.776) and required cajoling and winning consent. One web information architect who was interviewed stated:
I t is all about persuasion, horse trading skills. It is not like they say in the textbooks where you can go away and do this research and come out with some wireframes and then that is kind of it. That is the easy bit really.
And his manager agrees with this approach:
That is the reason why we employed him – seductiveness.
Yeah I think the approach should be consultation first, and confrontation as the last resort, and we probably wouldn't bother unless it was something that seriously embarrassed the organisation.
A strong message from this research is that the business must be ready and willing to participate in redevelopment of content structures and that projects that are driven from a central initiative are likely to fail. In the organisational context, IA and content development go hand it hand and one interviewee points out that this lesson has been learned the hard way.
And the lessons we learnt from trying to do large redevelopments of themes is that unless the line area, the people who own the content are on-board you'll never get it to happen, and so we're sort of evolving from doing large scale redevelopments into smaller scale redevelopments and restructuring of websites and parts of the web, of the main website, according to the enthusiasm of the line area because we're not successful otherwise and we had, we had a person, who came in and had skills and knowledge about IA and about user testing and attempted to, to drive from our section a redevelopment of a whole theme and it, it literally took, well, a year and then nothing happened.
Managers above the level of web manager are generally disconnected and ignorant of the need for web IA. This is recognised by the web managers that were interviewed who all accepted a responsibility for knowing and leading the processes of IA in their organisation as this web manager indicates:
Well we are it………so does my manager know that [about IA] specifically? probably not. She is not a web person. Does the general manager know that specifically? Again probably not, that is not her job. So whoever is at my level would have the responsibility for owning that knowledge within the organisation .… so the people with that responsibility do know it. The people who would depend on the advice of people like me, if you asked them that question they might even say: what's information architecture?
Higher level managers are still being persuaded and educated that IA is important and necessary. ‘Web managers often saw themselves as bridges or ambassadors' (Cox 2007a, p.776) with a strong desire to enhance web site usability and to influence senior management in order to achieve this outcome (Cox 2007a). One facet of the management of a web site is to provide an effective IA and it is here that we see the importance of the role of web site manager. Web managers must educate and lobby higher management for resources for this skill set within the organisation as these managers describe:
…until recently we've probably haven't had a lot of support from higher up, so it's like you know it's only really lately that our branch head has actually realised that we are doing good things…. but you know he, lets, you know, we really need to prove to him that this is the best way and all that sort of stuff so that is quite a battle anyway to convince him that, you know, he should stick up for us but still he does seem to more lately.
…we deliberately included important stakeholders like our branch head, so I did some card sorting with him individually because I wanted to educate him about what it was, why it was important, how it, how it meant that what he wanted wasn't the thing that was going to happen but it was going to be a thing which was what the users needed to have as their structure.
This study reveals that leadership for web IA in an organisation exists at the level and role of the web manager. Web managers take on the role of ‘gatekeeper' or leader of how IA for the web is done within the organisation – a powerful and responsible position. The web manager decides and enacts what is needed and how it should be obtained. They are the point at which learning, valuing and knowing about web IA occurs – hence the value of IA to the organisation is largely the value that the web manager places on it. One web manager says of IA:
…it's not been an interest of the organisation, it's not been an interest of the executive…
For IA capability to be employed in any form, in-house or consultant, the need must first have been signalled by the web manager. Significant lobbying of senior management for resources, recruitment and other HR processes etc. are needed to affect the availability of web IA expertise. Strong commitment and determination on the part of the web site manager are needed to procure the skills necessary for effective information structures on the web site of a large organisation.
Cox (2007b) points out that the web and its growth opens up need and opportunity for new expertise and explores the work of those with full time roles in the production of university web sites, with a particular focus on the web manager. The position of organisational web manager remains an emerging new profession with ill defined roles, expectations and place in the organisation (Cox 2007b) – that of the web information architect is even more recent and nebulous.
‘Liminality' refers to ‘the condition of being betwixt and between, of existing at the limits of existing structure' (Tempest and Starkey 2004, p. 507). Those working in new or temporary positions can lack the assurance of a well defined place in an organisational structure and of a traditional and established profession – they are ‘inhabitants of new territory' (Gornall 1999, p.48). Gornall goes on to describe those in newly developed professions as ‘threshold people who fall on or between the boundaries of categories' of workers in an organisation (p.48). New organisational roles and work, such as that of web IA, can be viewed through a lens of liminality. Gornall (1999) applies the concept of liminality to the relatively new profession of educational technologists in universities and Cox (2007b) extends it to those involved in web site production and management. Liminality has also been applied to the more flexible contract and consulting modes of working (Barley and Kunda 2004, Garsten 1999).
This paper argues that the professional role of web information architects in many organisations is one with a ‘liminal' status. There are three aspects of IA work that contribute to the liminality of the work.
The ambiguous nature of liminal roles is noted by Barley and Kunda (2004), Tempest and Starkey (2004) and Garsten (1999). A liminal status, carries both risk and opportunity and implications of both marginalisation and power – ‘the capability of upsetting normative orders and of transcending institutional boundaries' (Garsten 1999). Positively, the liminal worker can be seen as a catalyst for change and creativity, ‘a position from which new models, symbols and institutions arise' (Garsten 1999, p. 615) renewing and instigating change in organisations. Yet marginalisation, vulnerability and invisibility exist for the liminal new professional. There is inertia, ownership and established order and tradition to combat in the workplace. Cox (2007b) warns that for some individuals involved in web site production, the vulnerabilities are real. The complexity, the struggle and the uncertainty of the work can fail if not ‘connected up to a valued organisational purpose' (p. 166).
The work of web IA is a newly emerging professional role and must find not only its place of best fit within the organisation but also a way of working that is powerful and productive. These research findings point, for the development of content structures, to that being a mode of collaboration with web site owners. One information architect described it as a ‘velvet glove' approach to achieving effective IA. This aspect of liminality in the work of an information architect will not be transitory. Barley and Kunda (2004, p.176) note that ‘for contractors….liminality is a continual condition, indeed a way of life'. Corporate IA professionals will continue to need to work in a ‘consultancy' role with devolved business units and establish transitory working relationships that enable the design method of web IA to proceed effectively.
The vulnerability or marginalisation of the new role is partially overcome with acceptance and improved credibility within the organisation, often borne of improved web IA outcomes. Cox (2007a, p. 776) considers this process of establishing a new professional role as the ‘legitimization' of a profession. Cox (2007a) points to the legitimization of the work involved in building and maintaining web sites and that legitimacy can be applied specifically to the work of web IA. Legitimizing of this profession requires information architects and web teams in general to build credibility and reputation in organisations. These stories demonstrate the growing credibility of the IA work carried out by central experts and gradual acceptance of this new work within an organisation.
I basically just took on jobs with, almost acted like an internal consultant, I went around and saw people who were interested in doing anything about the web and, and got them to work with me to improve the stuff and then, by word of mouth, that would expand to other people saying “I heard you did this job for so and so, can you come and help us do that?”.
It's been recently, because of the work SE's been doing, they've actually recognised, well there's also I guess we're a corporate area and we're not a line area we, there's, I think it's less now but people worry that they're the experts of the content and if we did something whether we'd change it or do something wrong with it and all that sort of thing but from the things that we've been doing people realise we don't, we actually improve it so then again it's giving us more and more work on top of the normal, “here's a new publication, mark it up and stick it on the website”.
In some ways it is inaccurate to say that information architecture is new work. Every web site has an information structure (Evernden and Evernden 2003) and information has been organised and placed on the web for almost two decades. In early days and frequently in current practice, information structures are designed by unskilled but enthusiastic workers with no training in IA. Cox (2007b) labels those practising web site production without training as ‘hobbyists'. Professionalising this work and wrestling it back from amateurs or ‘hobbyists' is a gradual and ongoing process and one that needs organisational acceptance of the business value of doing web IA well and recognition of IA as skilled design activity. The following stories are of strategies, sometimes unintended, for the legitimisation and professionalising of the work of web IA.
Yeah because we, we went along to the web training, the training for writing for the web and we said “blah, blah, blah, blah, blah, this is what we can do for you and this is how we do it” and they thought “okay, right they've got this professional person in teaching us writing”, they seemed to be respecting these people, these people who know what they're doing and so they started to think that we were professionals rather than, you know.
And that is why it goes back to standards. People feel kind of reassured. If I just go up to somebody and go, “Ah I think I might just look at the business goals or whatever”. But if I say, “Oh we are using ISO 13407” loosely, people are very reassured.
… it's getting to the point now where that's becoming a positive thing and we're getting too much work, but we're getting, we're getting champions who are looking for money in their budgets to be able to do real projects with a real amount of money to employ us as sort of semi project manager to manage the process and the liaison with the contractors.
The process of legitimising IA work is complicated by the counter arguments to the centralisation of the design of information for web sites, such as the sheer quantity of information in some large organisations and the desire of organisational sub units to own and maintain their own information (Cox 2007a).
The process of designing a structure for the information on an organisation's public-facing web site is unarguably knowledge work. It is about making organisational knowledge and information explicit, publicly available and easily accessible via the internet. It is logical then to suggest that this work would be enriched by an environment that is aware of the enabling processes and principles of Knowledge Management (KM). It is not the intention of this research to examine the KM strategy or maturity of organisations, but to identify any strategies, intentional or intuitive, that foster the knowledge and work of web IA and to consider their fit within a KM framework. To further the clarification and understanding of web IA in organisations this paper will seek, where possible, to map the IA patterns and practices that are being carried out in organisations to KM theoretical good practice.
High level KM/Information Management (IM) strategy or policy documents were not acknowledged by research participants as influencing the work of web IA to any large extent:
There is no free standing policy for want of a better term that governs an IA as opposed to anything else. It is more bundled in with the operational concerns – here is a tool kit, if you want to take some short cuts, we will help you take them, but there is no sort of overall document. The only sort of thing that exist at that level are things like information management plan which don't go down to the level of what the IA should look like; they simply outline some high level priorities about the web as a delivery mechanism for information.
And there is a set of principles about information management. There should be one version, one authoritative version that should be re-used, so that helps to guide some advising consultancy that we do. Yeah, very much high level principles rather than….
In two of the three organisations studied, there was an explicit existence of a KM/IM organisational unit and the web IA function was situated within that KM/IM umbrella. Those two organisations were conscious of the need for web IA skills and were proactive in acquiring ongoing expertise in IA. Likewise, in these two organisations, there was strong support for the in-house positioning of IA expertise There was an awareness that employing an IA consultant for a finite term to produce an IA ‘product' was capturing very little learning about web IA within the organisation. Rather, these two organisations preferred to employ IA capability in-house in an effort to foster IA as a process to be supported in an ongoing manner within a specific cultural context. This preference is expressed clearly by research participants in those two organisations:
When you use a consultancy for an IA type person as well, you get them in, they come in and they go “yep, blah, blah, blah” and they come up, usually they have a wireframe of how pages should look and a structure diagram, “this is how stuff goes” and then they go away and so then it comes to us and we start implementing it, so we start going “okay you start mapping the existing content with the new content, working with the gaps and all that sort of stuff” and you realise that this section here which has got five or six subsections is really, there's no content and there's not going to be content and so you then collapse it back.
And we end up changing the IA as well… The thing that we're finding, if you were to go out and have a consultancy, you really do need the IA skills inside because that thing that they leave you with is not set and its, it changes, so you need to have those skills again to do this sort of stuff,
And our previous experience was that when we had consultants come in, do a quick job and go, they never, they never understood the culture of the department and quite often they didn't have sufficient sector experience to understand the tensions or the subject matter, the business processes or the nature of the organisation..
By way of contrast, the third organisation investigated did not have a KM/IM organisational unit, did not employ in-house expertise in web IA and did not consider it a necessity. An IA consultancy was under consideration for a future project. From these preliminary research findings it is worth considering the relationship between how effectively organisations enable the work of web IA and the KM/IM vision, culture, organisational structure and maturity. It is relevant to reflect on whether the two information architects located in web teams in KM/IM units were better positioned to provide effective IA for their organisation's web sites.
Web IA , described by Hider et al. (2008) as a design process or method, is predominantly ‘know-how' knowledge work. Brown and Duguid (1999) tell us that ‘know-how' and ‘know-what' knowledge are both components of a core competency in organisational knowledge generation. The practice of web IA is the enactment of knowledge about the design method of IA – it is knowledge in action. It is in keeping with Brown and Duguid's (1999) claim that not only is ‘know-how' revealed in practice, it is ‘also created out of practice' – ‘the product of experience and the tacit insights that experience provides' (p. 31). This research has revealed that content structures of organisational web sites are being developed in a working relationship between IA experts and business stakeholders. Organisational knowledge of web IA, at the very least the need for the IA design method, is created from the insights gained by the collective doing of IA. We see too, the heavily social character of organisational knowledge that Brown and Duguid (1999) have described and evidence of the communities of practice that Wenger and Snyder (2000) have outlined.
Snowden (2002) claims that KM is now in its third generation – one that views KM as flow and an ‘ephemeral, active process of relating' (Stacy 2001, p. 4). Rather than leave behind the focus of the second generation, which characterised knowledge as ‘thing', Snowden prefers to view knowledge as a paradox, both thing and flow. He also points to the benefits of considering complex adaptive systems theory for the management of knowledge within an organisation – a complex system is made up of many interacting agents and direct cause and effect are not knowable. Snowden (2002) proposes that identification of and attention to patterns of interactivity, enables a manager to influence these patterns, disrupting or encouraging as appropriate.
IA knowledge work within organisations in this study occurred in at least three patterns:
These patterns, once identified, are open to the influences of management for refinement or reshaping. Opportunities exist for managers to ‘steer' the work of IA in a specific organisational context
Tempest and Starkey (2004) apply the liminality lens to organisational learning that occurs at the periphery of organisational boundaries – to better understand ‘the benefits and challenges of learning between or at the margins of organisations' (p.508). Such a lens could also be applied to the new knowledge work, team work and learning that occurs at the internal dividing lines of large organisational structures. It is an appropriate frame for conceptualising the work of web information architects as they consult and collaborate with business units across an entire organisation to carry out the design method of web IA.
In the two organisations with formal IA expertise in this study, we see the person in that role working with business units to design information structures for sub sites. We see here the new skill set and expertise, working ‘betwixt and between' (Garsten 1999) organisational divisions, creating temporary teams and working relationships, transferring skills of IA processes and enabling informal learning within the devolved business unit. Tempest and Starkey suggest that teams and networks involving mobile workers may create ‘particularly rich learning environments that increase the possibility of innovative knowledge combinations' (p.512). This learning opportunity for IA occurs as web information architects work with diverse business units on specific web projects.
One of the metaphors that Inkson et al. (2001) present to understand and describe the way that mobile, liminal workers relate to the various organisations in which they work is that of a ‘bee'. A mobile worker is capable of pollinating the organisation in which they work with their expert knowledge - as does a bee in its work and movement. As mobile workers move from work location to work location they collect knowledge and skill and transmit them to people in the next situation. Although working within a single organisation, this metaphor can be comfortably applied to the way that the information architects in this study work with dispersed business units. An information architect's knowledge and expertise, always being developed by his/her activity, is ‘serially pollinating organisations with learning' (Tempest and Starkey 2004, p. 276) in IA. Knowledge of structuring web content is transmitted to where it is needed in an organisation, as the organisation gradually and successively learns to do effective web IA.
Many of the work practices of web IA identified in these case studies, can be fitted to KM principles, concepts and discourse. Yet the processes and practices for doing IA in organisations have more of an experimental and emergent nature and are not seen to be influenced by KM. Much IA practice has resulted from trial and error and sometimes been constructed using an intuitive approach with much adaptation along the way. Web managers and information architects have strong experiential knowledge of how best to practice IA in their organisations but are not supported by an explicit KM frame of reference. A clearer perception and explicit acknowledgement of the work of IA as knowledge work may hasten the developing maturity, effectiveness and legitimisation of the doing of IA in organisations. With this acknowledgement of IA as knowledge work, there comes the chance and benefits of applying a KM framework to examine web IA at a higher level of abstraction.
Snowden (2002) and Brown and Duguid (1999) encourage organisational managers to develop awareness and reflective processes in their responsibility for knowledge work. There is opportunity then for those responsible for managing the work of IA in organisations to explicitly examine the patterns of interactivity and processes of relating that have developed, possibly without intention or reflection and to manage the way that IA is achieved by disrupting or promoting these patterns. There is scope to consider the IA learning that is occurring across the organisation and how it might be captured at the organisational level. Shifts are needed to increase the corporate ownership of web IA within organisations – so that the practice of IA can truly be considered an organisational knowledge asset.
In the findings of this ongoing research, we see the emergence and identification of some of the factors, patterns and strategies affecting the successful practice of web IA in large organisations. The delineation of responsibility and authority for web IA is served by the categories of global utility navigation and page layout, and content structures that organise business information on the web. The global IA components are increasingly and productively the responsibility of central IA expertise whereas the latter category requires collaborative, team work with business stakeholders within an organization. The work of IA can be further fostered by acknowledging IA as new work and as knowledge work. Understanding IA through these lenses enables IA work patterns to be examined and managed at a higher level of abstraction.
The need to better understand the practice of web IA in large organisations, the design of this research approach and the emerging findings of this research have been reported. This research presents a developing conceptual understanding of the way that web IA is done in large organisations by an in-depth examination of three case studies. The concepts presented are IA as ‘new work', IA as ‘knowledge work' and how authority and responsibility for IA have evolved and exist in organisations.
This paper will add to the existing knowledge and literature base in the area of web IA by providing emerging concepts that organisations can use to make positive changes in the way that they manage the practice of web IA. Positioned within the complexity of an organisational context, i t will begin to fill the current research and knowledge gap in the organisational practice and management of IA processes. It points to the need for further case study research that will continue to build concepts and theory that can be used to underpin the effective and accessible provision of information on the web.
The author wishes to thank Dr Philip Hider, Associate Professor Trish Milne and Dr Stuart Ferguson for their input and advice in this research and Ms Carole Gerts for her assistance in literature searching.
Australian Government Information Management Organisation 2004, Information Architecture for Websites:Better Practice Checklist, AGIMO, ACT.
Barley, S.R. & Kunda, G. 2004, Gurus, Hired Guns, and Warm Bodies:Itinerant Experts in a Knowledge Economy , Princeton University Press, Princeton and Oxford .
Brown, J.S. & Duguid, P. 1999, "Organising Knowledge", Reflections, vol. 1, no. 2, pp. 22-44.
Busch-Geertsema, B., Balbo, S., Murphy, J. & Davey, S. 2005, "Towards a Framework to Analyse Information Architecture Work Practices", Proceedings of OZCHI 2005 CHISIG, Canberra , Australia .
Campbell, D.G. 2007, "IA Research: The Future State of the Art", Bulletin of the American Society of Information Science and Technology, vol. 33, no. 5, pp. 9-10.
Charmaz, K. 2006, Constructing Grounded Theory, Sage, London .
Charmaz, K. 2005, "Grounded Theory in the 21st century: A qualitative method for advancing social justice research." in Handbook of Qualitative Research, eds. N. Denzin & N. Lincoln, 3rd edn, Sage, Thousand Oaks, CA, pp. 507-535.
Cox, A.M. 2007b, "The power and vulnerability of the "new professional": web management in UK universities", Electronic Library and Information Systems, vol. 41, no. 2, pp. 148-169.
Cox, A.M. 2007a, "Beyond information - factors in participation in networks of practice: A case study of web management in UK higher education", Journal of Documentation, vol. 63, no. 5, pp. 765-787.
Dillon, A. 2002, "Information Architecture in JASIST: Just Where Did We Come From?", Journal of the American Society for Information Science and Technology, vol. 53, no. 10, pp.821-823.
Eschenfelder, K. 2004, "Behind the Web Site: An Inside Look at the Production of Web-based textual Government Information", Government Information Quarterly, vol. 21, no. 2004, pp. 337-358.
Eschenfelder, K. 2003, "The customer is always right, but whose customer is more important?: Conflict and Web site Classification Schemes.", Information Technology and People, [Online], vol. 16, no. 4, 01/06/2006.
Evernden, R. & Evernden, E. 2003, Information First: Integrating Knowledge and Information Architecture for Business Advantage, Butterworth-Heinemann, Oxford .
Fast, K. 2006, "IA Column: The Confluence of Research and Practice in Information Architecture", Bulletin of the American Society of Information Science and Technology, vol. 32, no. 4, p.27.
Garrett, J.J. 2003, The Elements of User Experience: User-Centered Design for the Web, New Riders Publishing, Indiana .
Garsten, C. 1999, "Betwixt and Between: Temporary Employees as Liminal Subjects in Flexible Organisations", Organisational Studies, vol. 20, no. 4, pp. 601-617.
Glaser, B. & Strauss, A. 1967, The Discovery of Grounded Theory, Aldine, Chicago .
Gornall, L. 1999, "'New professional': change and occupational roles in higher education", Perspective, vol. 3, no. 2, pp. 44-49.
Hartley, J. 2004, "Case Study Research" in Essential Guide to Qualitative Methods in Organisational Research, eds. C. Cassell & G. Symon, Sage, London, pp. 323-333.
Hider, P., Burford, S. & Ferguson, S. 2008, "The Use of Information Architecture Guidelines by Australian Libraries", VALA 2008: Libraries / changing spaces, virtual places: 14th Biennial Conference & Exhibition, 5-7 February 2008, Melbourne Convention Centre, Australia : Conference proceedings. Melbourne : VALA, 2008.
Inkson, K., Heising, A. & Rousseau, D. 2001, "The interim Manager: Prototype of the 21st-century worker?", Human Relations, vol. 54, no. 3, pp. 259-284.
Morville, P. & Rosenfeld, L. 2006, Information Architecture for the World Wide Web, 3rd edn, O'Reilly and Associates, USA.
Sinha, R. & Boutelle, J. 2004, "Rapid Information Architecture Prototyping", Proceedings of the 5th conference on Designing Interactive Systems, eds. D. Benyon, P. Moody, D. Gruen & I. McAra-McWilliam, ACM, Cambridge, MA, USA, pp. 349.
Snowden, D. 2002, "Complex Acts of Knowing: Paradox and Descriptive Self-awareness", Journal of Knowledge Management, vol. 6, no. 2, pp. 100-111.
Stacey, R.D. 2001, Complex Responsive Processes in Organisations: Learning and Knowledge Creation, Routledge, London .
Strauss, A. & Corbin, J. 1990, Basics of Qualitative Research, Sage Publications, Newbury Park , California .
Tempest, S. & Starkey, K. 2004, "The Effects of Liminality on Individual and Organisational Learning", Organisational Studies, vol. 25, no. 4, pp. 507-527.
Yin, R.K. 2003, Case Study Research: Design and Methods, 3rd edn, Sage, California .