Analytical Framework | Business Process Management
This is a case study of the Australian Bureau of Statistics’ 2016 online census project as a portrait of complete management failure by government, the ABS, and its principal contractor, IBM. The case study is split over several pages because of its length. The table of contents below offers quick navigation to the various sections. Scroll down to read the contents of this section.
The goal of this analysis is to begin with the particular and work towards a progressively more strategic overview to create as deep an understanding as is possible of interrelated groups, aims, methods, and outcomes relating to the case study.
Figure 2 presents a graphical representation of these interrelationships. The coloured horizontal bars represent management dimensions divided into several specialisations, which will be individually considered in the following pages, narratively ordered from the top down, as indicated by the icons to the right, matching those on the contents section.
The specialisations are sub-sets of strategic management, and sometimes overlapping in their application to overall strategy. They are aligned to reflect a management continuum with strategic planning at the left and operational management to the right.
The areas of specialisation are examined in the context of the principal actors in the online census project, shown by the rectangles outlined with dashed borders: the government and parliamentary processes; the public service generally but the ABS in particular; and the private sector generally as contractors to government, but IBM in particular in this case study.
The management performance of the principal actors is examined from the three stakeholder perspectives described in Table 1, and shown in Figure 2 as shaded vertical oblongs.
Not every one of these perspectives is applicable to all factors, and no single perspective is presented as superior to the others.
|TABLE 1: Stakeholder Perspectives|
|Informed Citizen (IC)||The nominal sponsor, beneficiary, consumer and owner of all public services.||Concerned about value for taxation, rationality of policy, integrity and competence of public officials, fair play in the private sector. Influenced by public opinion expressed through news media, but inclined to do some original online and social investigation into current affairs.|
|Professional (P)||Expertise in various professional disciplines to evaluate the processes and actions of the Commonwealth Government of Australia, the Australian Bureau of Statistics, and vendors (especially IBM).||Contemporary professional assessment based on standards, canon of academic and practical knowledge and experience. Particularly as they relate to –
|Self-interested-Shareholder (SIS)||A model of government as a business, and citizens as shareholders is a paradigm favoured by many proponents of laissez faire market capitalism, including the Coalition itself.||Looking for sound strategy, management competence, and results that reflect well articulated outcomes that create a return on investment or some other quantifiable value.|
The goal is to identify how differing points of view may focus on different aspects of the case study, with differing criteria for evaluating success or failure.
Business Process Management
The interactions and processes that underpin the operation of the ABS generally, and the online census project specifically, can be seen as intractably complex and opaque to most public observers. To assist in simplifying this perspective, tools from business process management permit a visual representation of a high-level abstraction as a starting point.
Business process management is a commonly used method with its own visual language, business process model and notation (BPMN), that uses a set of common symbols to chart process flows. Its principal function in many organisations is to streamline activities for cost savings and other efficiencies, but it can be useful in organisational re-design by helping to understand existing processes to design improvements, or to design entirely new processes.
Using BPMN conventions, a strategic overview of the ‘choreography’ or interaction between the government, the ABS, and its vendors is presented in Figure 3.
An explanation of the diagram is given below:
1. Ultimate authority and responsibility for budget processes, ministerial responsibilities and conduct rests with the Prime Minister and his Cabinet. From the SIS perspective this is analogous to the authority and responsibility of an Executive Chairman plus a board of executive directors with responsibilities for corporate divisions. If budget shortfalls or negligent ministerial oversight can be traced as root sources for any subsequent failures, it is reasonable to argue that the responsibility rests here. The red dotted lines throughout the diagram represent an ‘association’ between actors and processes whose nature is not entirely clear because it is opaque to public scrutiny (sometimes deliberately secretive). Arrows pointing in both directions indicate a two-way interplay between actors and processes.
2. From an interested citizen perspective the budget process is highly politicised, and subject to rhetorics that are often difficult to reconcile with proposed policy. In the case of budget cuts for the public service, it becomes uncertain whether departments have failed to perform adequately if they do not possess the necessary resources. The SIS perspective might then be that the census was always seen as an undesirable line of business, with funding cuts imposed to wither that business away. Alternatively, if the census was seen as an essential service, the Chairman and Board failed in its duty of care to ensure adequate funding and support for that line of business.
The professional perspective could be that budget allocations reflect priorities, and the census was not high on that list. Whether this is a mistake or not depends on political evaluation of the full scope of extant priorities.
3. A ministerial oversight vacuum after the last Federal election and immediately prior to the census is clearly the responsibility of the Prime Minister and his Cabinet.
It could be argued the responsibility rests solely with a Prime Minister who has exercised significant executive authority to make ministerial appointments.
Nevertheless, a minister cannot claim to be absolved from responsibility without admitting potential incompetence or implying being misled by department heads.
Citizen, professional, and shareholder observers are likely to conclude that there is reason to question the competence and diligence of the executive board for neglecting the census and then seeking to pin the blame on their subordinates.
4. The actors and processes enclosed by the yellow dotted line are largely ‘black boxes’, containing sub-processes and choreographies invisible to the public. Given the political dimension and largely unassailable impunity of politicians, questions relating to sector 4 of the diagram become highly political and subjective. The only certainty is that what occurred in sector 4 had significant downstream effects on the online census system planning and implementation process, particularly on the viability of the project’s business case.
5. ABS stage planning involved outsourcing work packages to various vendors, indicated by the dotted amber lines, but with uncertain lines of supervision and responsibility, indicated by the absence of arrows for the dotted amber lines. It is therefore unclear whether an ABS project manager had oversight of the outsourced work packages and their architectural, risk, or quality dimensions. If that is the case, there are grounds for concern particularly from professional and shareholder perspectives about lacking competence and governance by the ABS project management team.
That team would logically include the chief executive David Kalish, chief operating officer Jonathan Palmer, chief information officer Patrick Hadley (and possibly a project manager responsible to him), general manager Chris Libreri, and Program Manager Duncan Young (see the ABS organisational chart).
If it were agreed that this management structure constituted a point of failure, it would suggest there is a structural and/or capability problem within the ABS management hierarchy. A structural problem might point to the necessity for changes to the legislation governing the ABS, changes to resourcing necessary to attract or foster missing capability, or to the need for a management shakeup.
An alternate view, from the citizen perspective, is that IBM should have addressed all these shortcomings, accepting a consulting responsibility as the principal contractor to identify and address client structural difficulties.
6. The entire layer of outsourced services is a black box, indicated by the yellow dotted rounded rectangle drawn around what remain speculative sub-processes. Speculative because the precise nature of the processes and actors remains unknown. Architecturally this black box represents a problem in that it is not certain what level of oversight and control was available to the ABS, and exercised during the project.
7. Project execution is a nexus for many red lines simply because it is unclear whether the ABS was in control of architecture and project management, whether these functions were outsourced, or whether they were neglected. If outsourced, it is unclear what level of oversight and control remained within the hands of the ABS. If the ABS lost control of the project at this juncture, it could be regarded as a serious management failing. The red dotted lines with arrows pointing to and from the vendor layer indicate that we know there was some communication between the actors after problems were encountered, but the lines are rendered in red to indicate that this communication did not resolve the problems encountered.
Stakeholder perspectives are likely to be similar to those mentioned for point five above.
8. Given the nature of the technical problems encountered, there are now legitimate questions about whether data exfiltration occurred, whether data was contaminated, and whether significant numbers of citizens deliberately falsified answers as a protest against demands from the ABS to provide personal information.
Without definitive answers to these questions, the reliability of the entire census, online and offline, becomes uncertain because there could be concerns about significant misleading data. In that light, it is unclear what would constitute a successful project completion.
Reaction from all three stakeholder groups is likely to be that uncertainty about census outcomes is due to leadership and management failure at all levels.
9. Leading on from point 8, given the nature of the unanswered questions mentioned above, there appears to be no logical project close condition or event, except for a finding of project failure.
Political process may not be subject to professional business process management, but there are agenda and policy platforms, and there are discontinuities between public rhetorics and outcomes.
From the shareholder perspective for example, if the intention behind budget cuts to the ABS is seen as fostering greater efficiency, but no other complementary measures are taken, the stated agenda can be seen as not aligned to a serious commitment to achieve efficiency outcomes. It is a visible absence of developing adequate processes to support the stated goals.
Opportunities existed, for example, to amend the legislation in order to change the old fashioned silo structure of the ABS as a precursor to encouraging a move away from the top-down command-and-control paradigm to some other, more flexible and agile teamwork structure (see the section on strategic management).
Within the ABS itself, there appears to be a conspicuous absence of processes to develop or better explain clear business cases for the bureau’s statistical products. These would have helped with more convincing budget justifications, and perhaps also to produce roadmaps for inter-departmental cooperation to more efficiently leverage existing public service resources.
Looking at government agenda on their own, it is easy to conclude that budget cuts in isolation from other policies were not aimed at creating greater efficiency, but rather at the withering away of the bureau’s capabilities. That outcome is consistent with the present government’s stated aims to shrink the functions of the state by outsourcing or privatising them. Privatising state functions has had limited public support for its history of increasing costs without evidence of greater efficiency or service quality (this is most evident in electricity retail and supply, telecommunications, and the unemployment industry). To make the case for such privatisation, therefore, it would be useful to be able to point to a massive public sector failure.
As for the ABS, it may not be able to sway government policy, but its leadership did not demonstrate a capability to better argue its funding needs on the basis of practical and convincing plans.
The business process management (BPM) method is much more versatile than shown in this analysis. Figure 4 illustrates a complete BPM lifecycle, indicating that it can turn high level strategy into operational routine, and assist with continuous improvement in all areas of operation. That implies a potential for efficiency gains, cost reductions, and the development of marketable new products in many other ABS areas of activity. Such activities are beyond the scope of this analysis. However, it is pertinent to note the potential for business process management to assist government and the public sector in clarifying existing modes of operation, and to develop desired future modes of operation from that base.