One thing most readers will know about large Government IT projects is that many end up being labelled ‘failures’, leading to delays in the implementation of policy and write-offs of millions of pounds of taxpayers’ money. Much previous academic work has suggested that such failures are the result of weak management or poor application of technology (e.g.Nawi et al, 2011). Yet significant efforts have been made to rectify problems like these, especially within the UK Government. I was intrigued by this apparent paradox and wanted to understand the detail of how such projects played out and how we could account for this paradoxical situation. I interviewed project managers, project reviewers and auditors with responsibility for delivering and evaluating these projects and for compiling recommendations when projects were seen to fail. I also analysed audit reports (published by the National Audit Office) and project review reports (normally confidential but a few have been put in the public domain following Freedom-of-Information requests). By analyzing this material from a narrative perspective (Czarniawska, 2007), I was able to identify how and why project narratives were dynamically and interactively constructed in context to produce the label ‘success’ or ‘failure’ (Fincham, 2002) and how this label then became accepted ‘fact’. In this blog I provide an example of how this kind of analysis can account for the continuing failure of some Governmental IT projects.

“Digital by Default” and “Digital by Design”

The UK Government had the ambition to take the majority of interactions between the general public and Government on-line. In a world where you could bank, shop and even get therapy over the internet, why would you want to telephone a Government Department, or even worse, fill in a paper form? The proponents of ‘Digital by Default’ in the Government Digital Service (GDS) wanted to radically change the way that Government ran IT projects. Mostly recruited from outside Government, these ‘digital evangelists’ (Kanter, 2011) dressed differently, spoke differently and importantly behaved differently to ‘conventional’ civil servants. Their network of power relations included a senior Government Minister who acted as gatekeeper for project go-ahead.
Some new IT projects were designated ‘Digital exemplars’ and received special support. These included a project called Common Agricultural Payments-Delivery (CAP-D), which implemented a new regime of farm payments. Concerns were raised by users and policy owners that the new system would not work in practice. Farmers were not the most digitally literate constituency, and the poor connectivity offered by rural broadband was not conducive to the transmission and manipulation of digital maps. More prosaically, the new system would crash with a high workload. As the hard deadline (set by the EU) for deployment approached, the development was abandoned and a backup system was implemented by the Rural Payments Agency (RPA) using paper maps, which were posted to farmers, hand annotated and posted back. These were then interpreted by eye before subsidies were paid out.

The perceived failure of the digital project was the subject of a Public Accounts Committee inquiry. The senior GDS official involved was criticized for having “digital dreams” (PAC, 2015 Q 62). Analysis of competing narratives around this project reveal a long battle for its control. While GDS was developing the software, the RPA had complete control of communications with the users; farmers and rural landowners. In these communications the RPA made it clear that the new software was ‘owned’ by GDS and soon the farming press labelled “Digital by Default” as ‘Dogma. By promoting a counter-narrative “Digital by Design”, the RPA relabelled the CAP-D project as a success, as the backup system had saved the policy and farmers had been paid.

Conclusions

I interpret these narratives and counternarratives as tools in a system of power relations (Dawson and Buchanan, 2005) and indicative of a struggle for control of the project. Different parties used ‘success’ and ‘failure’ as rhetorical terms (Fincham, 2002) in order to justify different courses of action. The term ’project failure’ is frequently deployed when a struggle reaches a conclusion, and is used to justify a change in approach, a reduction in scope or a decrease in funding. In 2012 GDS used the ‘failure’ of previous Government IT projects as part of the rationale to introduce “Digital by Default”. In 2015 other actors used the ‘failure’ of CAP-D to discredit “Digital by Default” and regain control of the project, and later label their intervention as a ‘success’. The labelling of a project as ‘failed’ is an indicator of the intensity of the conflicts surrounding them. Yet formal reviews and audits of Government IT projects omit all mention of political activity. Projects are discussed as rational, objective engineering activities and the struggles for control over scope, funding and approach are glossed over. Until such a perspective can be overtly acknowledged, it is not possible to fully understand ‘failing’ Government IT projects or learn from them.

References

Dawson, P. and Buchanan, D., 2005. The way it really happened: Competing narratives in the political process of technological change. Human Relations, 58(7), pp.845-865.

Fincham, R. (2002) Narratives of Success and Failure in Systems Development. British Journal of Management, 13, 1–14. Available online at: http://doi.wiley.com/10.1111/1467-8551.00219.

Kanter, R. M. (2011) Kanter Evolve ( Again ). Harvard Business Review, 89(7–8), 36. Available online at: hbr.org.