Go to top of page

Experimenting and engaging to innovate

There is a persistent view that characterises the public sector globally as conservative, bureaucratic and reluctant to change.13 This view has been challenged, in particular in relation to service innovation.14 The public and media scrutiny under which the public sector needs to innovate is, however, significantly higher than in the private sector, and often the focus is confined to achieving efficiencies in the way public services are delivered. One author characterised the environment in this way15:

Any failure … leads to inefficiency, which of course is bad … Innovation, along with uncertainty, entrepreneurship; imagination, experimentation, competitive enterprise and technological and structure change are excluded, by definition.

While this is an extreme characterisation, the reality of innovation in the public sector involves complex political and stakeholder considerations that lead to variable assessments of risk tolerance. Open innovation, based on the idea of ‘free-revealing’ (providing others with early insights into thinking or data) to engage users and stakeholders in concept and design, requires a different risk appetite and new techniques. For example, open innovation is characterised by early involvement of stakeholders in iterative, interactive probe-and-learn approaches to understanding a problem and developing solutions.16 These approaches challenge the more structured approaches to innovation that have characterised traditional inside-out approaches.

The above notwithstanding, there are good examples in the APS of where open innovation approaches have been successfully employed. One initiative by the then Department of Families and Community Services is the HILDA survey, which has been running since 2001. This is one of only a small number of well-established, large, nationally-representative household panel studies in the world.17 The contract to design and manage the HILDA survey was awarded to the Melbourne Institute of Applied Economic and Social Research (at the University of Melbourne). The data collected through the HILDA survey is available as a confidential longitudinal dataset on application from the Department of Social Services (DSS). HILDA survey data has been used by academics, government and other research bodies to achieve a range of outcomes. Its website tracks these uses and outcomes in a bibliography.18 As at September 2014, this bibliography included the details of more than 540 journal articles, 34 books or chapters in books, and a large range of other publications, including student essays and dissertations and conference, research and discussion papers.

The approach to innovation at the Department of Human Services (DHS) is also noteworthy. The iDHS programme was launched in October 2012 and by May 2014, more than 1,100 ideas had been submitted, with more than 900 posted and discussed through online employee forums.

The process, which has seen more than 4,000 posts from approximately 1,300 different commentators, effectively allows for employees to moderate and shape the evolution of innovation proposals against the strategic objectives of providing easy and efficient access to services. Time-limited challenges also provide regular open forums for employees focusing on specific business initiatives or issues. These mirror the extensive efforts that DHS already undertakes in engaging customers and community representatives in co-designing services.

Additionally, DHS holds administrative (mainly payment) data from the Centrelink, Medicare and Child Support master programmes. Final responsibility for external release of that data lies with the responsible policy agencies (DSS for Centrelink and Child Support data, and the Department of Health for Medicare).

External requests for data from these databases can be made to statistics [at] humanservices.gov.au which is administered by DHS on behalf of policy departments. Release of data depends on the request meeting legislative and privacy provisions (which includes a guarantee of anonymity and the purpose of the research being in the public good), plus the scale of the request being within the scope of resources available to meet it. This determination is made by the External Requests Evaluation Committee (EREC) for Medicare data requests.

During January to August 2014, 203 external applications for requests for information were received and assessed by the EREC, with 20 of these (9.85%) declined. There were 86 external requests for Centrelink data and 15 requests for Child Support data in the same period, all of which were met.

In the same period, DHS received nine requests for data for data linkage projects, all of which were approved. These are assessed under the guidelines for data integration involving Australian Government data. In addition, DSS placed a large number of statistical datasets on data.gov.au which are available in machine-readable format.

DesignGov: Exploring business and government interaction

Last year's State of the Service report noted the establishment of DesignGov as an 18-month experimental project to explore new methods in solution formulation, development and delivery. DesignGov employed approaches and techniques consistent with the probe-and-learn approach to innovation.

DesignGov's major project was to demonstrate how design principles and techniques might be applied to improving interactions between business and government. The guiding question for the project was: How might we dramatically improve business and government interactions?

Fifteen small to medium-size business owners and operators in manufacturing, information technology-enabled services and hospitality/retail sectors, together with six intermediaries and 18 federal and state public servants, engaged in the project. Additional information, insights and ideas were obtained from workshops, online collaboration platforms, meetings, desktop research and surveys.

DesignGov gathered insights from the businesses, intermediaries and public servants involved in the project (shown below) and was able to highlight that the complexity in the relationship between business and government arises from each using different processes, different language and having different expectations and requirements of each other.19

DesignGov: Insights

Small businesses

Small businesses feel marooned and cut off from champions in government or industry. They feel alone, different from big business and not understood. They find it too hard to navigate government for the right information. They are frustrated by the constant change in policy and compliance requirements and often fear they are unwittingly doing the wrong thing.

Small businesses do not necessarily universally regard regulation as an unnecessary burden. While some specific requirements and regulations are onerous (particularly so if others are seen to avoid compliance), many interactions are seen as part and parcel of running a business. These businesses believe they operate on trust, respect, deep knowledge and stable personal relationships with their customers and other associations—this is lacking in their relationship with government. The cumulative impact of government attitudes and requirements on businesses and intermediaries is underestimated, not understood or may not be considered significant by much of the public sector.


Intermediaries—especially industry and professional—package and translate information from all levels of government relevant to their members. They are a vital connector. Intermediaries, whether informal (such as friends or other business colleagues) or formal (such as industry associations) are most critical to interactions at times of change or as insurance against ignorance of specific regulatory change and its implications. Consultation processes often fail all parties for reasons such as too little time, lack of clarity of scope and suspicion that the result is pre-determined.

Public servants

Public servants are aware of many of the issues but feel unable to fix them. Public servants may experience a conflict in expectations between the role of compliance enforcer and industry partner. At times, the public sector finds it difficult to deal with its own red tape, impersonal relationships, lack of information sharing and its complex operating environment. As a result, creativity is often directed to work-arounds. Public sector agencies have already introduced or are administering a number of formal and informal initiatives that seek to improve business and government interactions. Further improvements from government will likely require breaking siloed or lead agency approaches, as many issues cut across agency responsibilities or are at the intersection of multiple agencies or jurisdictions.

The first stage of the project (completed in September 2013) involved understanding the problem and key issues and then conceiving of them in a new way. The second stage (completed in December 2013) focused on testing concepts that could reveal further insights into business and government interactions, and shape practical options to dramatically improve these interactions.

The second stage used a design-led approach to identify five concepts that might be prototyped to deliver improvements in business and government interactions, along with number of action areas for individual agencies to consider. It was anticipated that by improving the quality of the interaction between business and government, there would be a reduction in frustration, wasted effort and inefficiencies leading to better engagement, reduced misunderstanding and better outcomes for government, businesses and the community at large. These five proposed concepts are shown in Figure 7.2.

Figure 7.2. DesignGov: Co-creating practical solutions to improve interactions between business and government—a framework for action
Facet of interaction Identified problem Proposed prototype
Source: Adapted from DesignGov 2013, Co-creating practical solutions to improve interactions between business and government, Commonwealth of Australia, Canberra.
Understand me and my context The context for interactions between business and government keeps changing Emerging issues detection
A systemic approach for sharing emerging issues that might have relevance for business and government interactions, either generally or specifically
Help me find out what I need to know Finding meaningful answers is hard Peer-to-peer crowd-support platform
Platform where people can ask questions and provide answers about navigating government and government processes.
Before making decisions recognise my experiences, my skills and my needs Businesses perceive consultation as ‘nonsultation’ and engagement as limited OneConsultation.gov.au
A single place for conducting consultations, supplemented by a suite of consultation and engagement tools and options to suit a variety of stakeholder and agency needs
Work with me to unravel problems There is a lack of clear channels for resolving specific problems that span agencies and jurisdictions, and that affect business Fix-it squads
A form of tiger team made up of seconded public servants from relevant agencies with responsibilities relating to a specific problem area identified by business and agreed by government, charged with developing options for resolving the issue
Treat me with respect and consideration The preconditions for high standard service provision in the APS are not consistently available Service by design
A common framework of service design principles and customer service infrastructure for trial and implementation

DesignGov tested the idea of design ‘hubs’ connecting existing APS design capability and expertise. These hubs would be used by agencies to access current practice in design-related methodologies. The business and government interaction project demonstrates how design thinking can be applied to a complex problem, spanning multiple agencies and jurisdictions.


Some open innovation approaches rely on crowdsourcing, which has been described as a way to ‘leverage the collective intelligence of online communities to serve business goals, improve public participation in governance, design products and solve problems’.20 Prominent crowdsourcing activities are the contributions made to Wikipedia and the development of open source software (for example Linux). The key principle of crowdsourcing is participation through input with the objective of reaching a large and diverse pool of people.

Public sector outsourcing activities are increasingly common. The first recorded example is from 2007 when the New Zealand Police Commissioner put its 1958 Police Act online as a wiki and invited people to edit it.21 The US Patent and Trademark Office is using crowdsourcing to evaluate pending patents to reduce the number of poor quality patents and subsequent lawsuits. Finland is soliciting input from citizens online to develop new laws.22 It has been reported that the California Probate Code will be the first instance of legislation being purely crowdsourced. The proposed changes will be created through an online platform similar to Wikipedia.23

National Library of Australia: Trove

November 2014 marks the fifth anniversary of the launch of Trove, a national discovery service built and managed by the National Library of Australia. Trove provides access to the holdings of more than 1,000 Australian libraries, archives, museums, galleries, government agencies, and universities, as well as more than 13 million pages of digitised Australian newspapers. Trove has gained significant traction in the Australian community, with more than 22 million unique visitors in 2013–14. Trove's digitised newspaper content accounts for 80% of service use.

Microfilmed newspapers are digitised by external contractors, who also process the resulting images, applying Optical Character Recognition (OCR) to generate text from them. OCR accuracy depends on the quality of the original newspapers and text rendered from images is therefore imperfect, reducing search accuracy. The Library developed a world-leading crowdsourced approach to correcting this imperfect text. Its aim was to invite the public to assist with a task too large for the Library and to improve the Trove discovery experience for all by improving search accuracy.

Trove allows volunteers to simultaneously view the image of a digitised article and the text resulting from OCR processes. Volunteers spotting errors ‘fix this text’ using a simple online tool. The Library intentionally took a light approach to this activity. Volunteers are not required to register, but gain additional benefits if they do so, including a running tally of lines corrected and a place in the Trove Hall of Fame. Text-correctors communicate with each other through the Trove forum, and have collectively developed a set of guidelines addressing the many issues encountered with newspaper text dating back to the 1803 Sydney Gazette. Text correction activity occurs 24 hours a day, 365 days a year. A typical day would see some 100,000 lines of text corrected by volunteers working in their own homes and offices. Instances of user misbehaviour are extremely rare, and the volunteer community is quick to report such issues to Library employees.

At the end of August 2014, more than 132 million lines of newspaper text had been corrected, significantly improving searchability of the newspaper corpus. This effort is the equivalent of 353 work years, has been valued at more than $23 million, and far exceeds any expectations the Library had about take-up of this volunteer opportunity. Text correctors choose their own work, and have adopted many approaches to the task: some correct all articles relevant to their family history; some choose specific topics, time periods or events; some groups have set themselves targets such as fully correcting their local newspaper. Academic research projects have tasked volunteers with particular projects. For example, the Eureka prize-winning South Eastern Australian Recent Climate History project encouraged digital volunteers to find and correct articles relating to 19th century weather conditions.

Text correctors tell the Library that text correcting is easy, relaxing, addictive, gives them a strong sense of giving back to the community, and allows them to express their gratitude for a service that has revolutionised their access to this important documentary heritage. A number of Trove's top text correctors are retired, and say they regard Trove as their ‘work’ and a way to keep their minds active in retirement.

For the Library, this crowdsourced activity has assisted with a task beyond the Library's resourcing capacity. More importantly, the programme has built a new, large and dedicated audience keen to support the Library and acts to promote the Library's collections, services and mission. The programme has also allowed the Library to reach millions of Australians who may never visit the Library in person, but who now know they have a National Library, and a national collection that belongs to them.


13 Borins, S 2002, ‘Leadership and innovation in the public sector’, Leadership & Organization Development Journal, vol. 23, no. 8, pp. 467–476; Mulgan, G & Albury, D 2003, Innovation in the public sector, working paper version 1.9, Strategy Unit, United Kingdom Cabinet Office, London, viewed 19 September 2014, http://webarchive.nationalarchives.gov.uk/20130128101412/http:/www.cabinetoffice.gov.uk/strategy/work_areas/innovation.aspx.

14 Gray, A, Broadbent, J & Hartley, J (eds) 2005 ‘Editorial: The state of public management—improvement and innovation’, Public Money and Management, vol. 25, no. 1, pp. 7–8; Walker, RM 2008, ‘An empirical evaluation of innovation types and organisational and environmental characteristics: Towards a configuration framework’, Journal of Public Administration Research and Theory, vol. 18, no. 4, pp. 591–615.

15 Potts, J 2009, ‘The innovation deficit in public services: The curious problem of too much efficiency and not enough waste and failure’, Innovation: Management Policy and Practice, vol. 11, no. 1, p. 36.

16 Gassmann, O, Enkel, E & Chesbrough, H 2010, ‘The future of open innovation’, R&D Management, vol. 40, no. 3, pp. 213–221.

17 Watson, N & Wooden, M 2012, ‘The HILDA survey: A case study in the design and development of a successful household panel study’, Longitudinal and Life Course Studies, vol. 3, no. 3, pp. 369–381, viewed 2 October 2014, http://www.llcsjournal.org/index.php/llcs/article/view/208.

18 Melbourne Institute of Applied Economic and Social Research, The Household, Income and Labour Dynamics in Australia (HILDA) Survey, University of Melbourne, Melbourne, viewed 2 October 2014, http://www.melbourneinstitute.com/hilda.

19 DesignGov 2013, Business and Government Interactions Project, DesignGov, viewed 19 September 2014.

20 Brabham, D 2013, Crowdsourcing, MIT Press, Cambridge, MA, p. 14.

21 Furnas, A 2012, Can We Harness the Internet to Collaboratively Write Better Laws?, The Atlantic, Washington, DC, viewed 22 September 2014, http://www.theatlantic.com/technology/archive/2012/02/can-we-harness-the-internet-to-collaboratively-write-better-laws/253445.

22 Charalabidis, Y, Triantafillou, A, Karkaletsis, V & Loukis, E 2012, ‘Public policy formulation through non moderated crowdsourcing in social media’, in E Tambouris, A Macintosh & Ø Sæbø (eds), Electronic Participation, vol. 7444, Springer Berlin, Heidelberg, pp. 156–169; Clark, BY & Logan, J 2011, A Government of the People: How Crowdsourcing Can Transform Government, ResearchGate, Cambridge, MA, viewed 22 September 2014, http://www.researchgate.net/publication/228159911.

23 Hickey, K 2014, Wiki to crowdsource changes to Calif. probate code, GCN, Vienna, VA, viewed 22 September 2014, http://gcn.com/articles/2014/01/21/crowdsource-probate-bill.aspx.

Last reviewed: 
8 June 2018