These are interesting times for those of us involved with public sector multi-sourcing. The Government Digital Service (GDS) earlier this year gave a strong indication that the Government is eyeing a sea-change in how it approaches multi-sourcing. The recent election of a majority government would also seem to provide some certainty in the near term that the policies championed over the last five years will continue apace, including encouraging small and medium-sized enterprises to play a greater role in the public sector.
The Tower Model
In the IT industry ‘multi-sourcing’ is generally taken to mean a customer delegating projects or services to two or more external vendors who must often co-operate to achieve desired objectives. Central government, in common with many businesses, clearly believes multi-sourcing can often deliver costs savings and improved services as against the alternatives of either doing everything in-house or using a single vendor.
Over recent years, many commentators became accustomed to treating the so-called ‘Tower Model’ as the dominant central government approach to multi-sourcing. That Model evolved in response to the multi-sourcing approach advocated in the Government’s ICT Strategy published in March 2011 as part of the desire to move to smaller-scale public sector IT contracts, with a best of breed approach, the re-use of systems and software, and the promotion of small and medium-sized IT contractors.
The classic Tower Model involves a public sector programme being broken into slices, bringing in different vendors for each Tower and appointing a Service Integration and Management (SIaM) contractor to co-ordinate and manage the services provided by the Tower contractors. The SIaM role traditionally includes the following:
· transition management;
· technical architecture leadership;
· service integration;
· service management and delivery, dealing with problems;
· change management; and/or
· inter-contractor liabilities.
My colleague, Roger Bickerstaff, discussed many of the most challenging aspects of Tower Model procurement in a previous article in this magazine. Those challenges include the allocation of liabilities, the treatment of delays caused by one vendor impacting on another, methods of encouraging effective co-operation and the incentives and scope of authority and autonomy given to the SIaM.
Knocking Down the Towers
Despite the recognised challenges of the Tower Model, there was considerable surprise in some quarters following the publication on 18 February 2015 of an article entitled ‘Knocking down the Towers of SIAM‘ by Alex Holmes, the deputy director of the GDS. Mr Holmes’ post described the Tower Model as unhelpful because it ‘combines outsourcing with multi-sourcing but loses the benefits of either‘ and concludes ‘the Tower Model is not condoned and not in line with Government policy‘ and ‘doesn’t work‘.
Instead, Mr Holmes expressed support for an approach based on the following principles:
· It is more effective for government to design and own the overall solution so that it knows it works and can put it together and run it. This role should be undertaken by a team largely made up of civil servants, who own the accountability and architecture.
· Effective multi-sourcing should avoid a situation whereby the customer buys individual components that might be incompatible and then asks the SIaM provider to put them together.
· Generally speaking, the preferred methodology is for government to do the following before procuring and commissioning what’s needed:
– understand user needs;
– bring in the right capability and skills;
– analyse existing applications;
– architect a disaggregated desktop using cloud infrastructure; and
– consider platform options.
· The preferred multi-sourcing approach is not ‘one-size fits all’, but should consider both outsourcing and in-house methodologies, as appropriate.
‘This announcement has caused our members some alarm, given in many procurement exercises, supplier briefings and techUK events in the last two years the applicability of the “tower” disaggregation model has been advocated by public sector bodies, including Cabinet Office and GDS representatives. The tech industry responded by investing in methodologies and proposals to help them deliver against the government requirements.‘
Whatever the extent to which government representatives have previously advocated the Tower Model, it is worth noting that within the public sector it has been largely a central government phenomenon and little used by local authorities.
Seeing Past the Towers
Alex Holmes posted a subsequent blog in March addressing some of the commentary his initial post generated. Of particular interest is his observation that looking forward, ‘The debate shouldn’t be about different fixed models (and certainly not what is the Government model), but how do we, like any other service-orientated organisation, build and buy services to meet the needs of our users? IT based around services, not departmental or contract silos.‘
The Tower Model, it seems clear, was only a transitional approach in moving towards this objective, which achieved some degree of success in breaking down government services into the smaller constituent parts of the Towers and SIaM roles, but which does not go far enough in creating a genuinely multi-sourced public sector environment.
The GDS’ approach appears to envisage the civil service increasingly taking responsibility for more of the SIaM’s role. To some extent this might result in a formalisation of what is already happening. In some recent projects there has been a tendency to regard the issue of the allocation of SIaM responsibilities as ‘all too difficult’, with the result that the customer de facto takes on the SIaM role itself in any event, often without fully acknowledging or understanding the risks.
A Low-rise Future?
The approach to central government procurements over the next decade is clearly still a work in progress, but the likely implications of recent developments are:
· Reduced role for the private sector SIaM: The ‘classic’ quite wide-ranging role of the SIaM would seem to be particularly transitional, with the Government aiming to take on more of that role in-house.
· Increased interest in G-Cloud and similar solutions: The GDS’ recent announcements and other efforts by central government to continue to educate public bodies about G-Cloud and other parts of the Digital Marketplace are likely to encourage greater uptake of those and other ‘off-the-shelf’ and ‘pay as you go’ services. However, the very flexibility of ‘pay as you go’ services with short contract terms and the generic nature of ‘off-the-shelf’ services will present barriers to their suitability in many contexts.
· Variations of the Tower Model: In recent years the Tower Model has been applied in hybrid ways, including the emergence of the so-called ‘guardian vendor’ model in which a leading contractor combines the responsibilities for SIaM and delivery of one or more key Towers. Some commentators, from industry in particular, have suggested that criticisms of the Tower Model can be addressed by adapting the Tower Model in various ways rather than rejecting the entire approach. Implicit in the GDS’ comments is the concept that the key driver should be user needs in a particular situation. Consequently, where a Tower-like approach is appropriate, that would not seem to be completely prohibited. However, following the GDS’ intervention, public bodies are in practice likely to be wary of proposing Tower-like approaches unless there is a clear justification, having examined other options.
· Development of other hybrid models, with the customer taking a greater role in SIaM: At a recent Bird & Bird event on the future of multi-sourcing in the public sector, the Director of Complex Transactions at Crown Commercial Service, Chris Hall, confirmed that the Government anticipates increasing its pool of appropriately skilled staff to make a new approach more feasible. Of course, it will take time for central government to build up in-house skills and it seems unlikely that many public bodies will feel confident in taking on all the SIaM roles in the near future. At the same event, Mr Hall also suggested it may be that in-house resources will be focused on key procurements – for example those supporting the GDS’ ‘Government as a Platform’ vision.
· Greater diversity and smaller public sector IT procurements: The Government’s overall approach in many ways brings the focus back to the guiding principles published on the Cabinet Office technology blog back in November 2013, with their focus on building a new eco-system based on shorter-term and more flexible arrangements and bringing digital skills back into government.
Implications for Practitioners
At one level, a move away from classic Tower Models may simplify contractual discussions relating to matters such as liabilities, SIaM and co-operation since, if the civil service is effectively taking on more of the role of the SIaM and greater responsibility for ensuring the pieces fit together, the risk of civil servants getting this wrong will be borne in-house by the public sector rather than allocated through contract. The preference for lower value and more flexible arrangements would also suggest a trend away from big ticket procurements supported by extensive private practice legal teams. On the other hand, an increased role for the public sector in managing multiple vendors and the development of a more diverse multi-sourcing environment is also likely to result in a greater need for legal support in advising on more complex and inventive co-operation and cross-liability regimes.
One thing that is clear: public sector multi-sourcing looks set to evolve significantly in the near future. Those of us involved closely with these projects will await these developments with keen interest.
Adam Gillert is a senior associate at Bird & Bird LLP, where he is a member of the Technology & Communications team.
