Multi-vendor or ‘tower’ outsourcing arrangements have become an established part of the public sector IT contracting environment over the last three or four years. The multi-vendor approach was introduced as part of 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 SME IT contractors.
Since 2011 most significant central government IT infrastructure projects have been conducted on some form of multi-vendor basis. Experience has shown that the transition from a single supplier to a multi-vendor arrangement is a less than straightforward transition. However, there is a now a good degree of experience in carrying out procurement processes for multi-vendor arrangements. This article presents some of the lessons learnt.
Single or Sequential Procurement Process
There are considerable benefits associated with the conduct of a single procurement process that seeks to award all of the contracts in a multi-vendor outsourcing at the same or roughly the same time. Cross-lot issues are more likely to be identified and dealt with during a single negotiating period. For example, it is highly likely that there will be issues associated with the integration of the services to be provided in the different towers and these can be discussed and resolved between the bidders during a preferred bidder process in a single procurement process.
It is highly likely that the service boundaries that are bid for by one supplier will not fully match the service boundaries of the other tower contractors. During the preferred bidder stage, these service boundary issues should be identified by the putative SIaM Contractor and dealt with through negotiation prior to the award of contracts. If these issues are not resolved prior to award, any changes to the scope of the contracts after award will need to be resolved through change control. This may have financial implications for the customer and can be problematic for suppliers.
Of course, this level of input is a significant task for the putative SiaM contractor to carry out at the preferred bidder stage. It may be preferable to make an early appointment of the SIaM contractor in order to free up resources and capabilities to work on behalf of the customer during the preferred bidder stage.
Cross-tower liability issues can also be resolved through negotiation between the relevant preferred bidders and the contracting authority at the preferred bidder stage if a single procurement process is carried out. There is an inescapable liability problem associated with multi-vendor contracting where the financial values of the towers are significantly different. In these circumstances, where the outsourcing arrangement has the possibility for inter-contractor claims, there is a risk that a ‘larger’ tower contractor could be precluded from making a claim against a ‘smaller’ tower contractor due to the relatively low financial limit of liability in the ‘smaller’ contractor tower services contract. From the ‘smaller’ contractors’ perspective, there is a risk that it will be exposed to relatively large claims as a result of the scale of operation of the ‘larger’ tower contractor. The larger tower contractor is likely to incur significant costs and expenses through the scale of its operations in a relatively short time.
In a prime contract situation, the prime contractor generally accepts these risks as being part of the prime contract risk profile, on the basis that it will receive a prime contract ‘mark-up’ as part of its payments. In a multi-vendor outsourcing, this does not occur. The issue has to be resolved either during the negotiation process or it will remain as an unresolved issue during the lifetime of the project.
Despite these advantages, it must be recognized that conducting a single procurement process for the award of a range of tower contracts, the SIaM service agreement and the integration agreement requires considerable resources and expertise from the contracting authority. From the bidders’ perspective the award process will almost inevitably become cumbersome and protracted, with associated costs and corporate concerns. The process can only proceed as quickly as the timescales of the slowest tower negotiations. Bidders undoubtedly prefer a shorter procurement process. The overall process often becomes delayed when problems arise in one tower which do not have a direct impact on the other tower procurement processes but which nevertheless delay the overall process towards the award of the tower contracts in a coordinated manner.
For the customer, obtaining sufficient resources and capabilities to be able to carry out a multi-vendor procurement process effectively is a significant challenge. The award of each contract may well be a significant procurement process in its own right. Each tower procurement process needs an allocation of technical, operational, commercial, financial and legal resource. It will be rare that even a large public sector contracting authority will have these resources available internally at the right time. Consultancy expertise will need to be bought in at additional expense, with the risks that the consultants do not understand the customer’s operations and may be have insufficient expertise to carry out the work.
Use of Framework Agreements
The use of public sector Framework Agreements for the award of the tower service contracts is now frequently being suggested or mandated. In particular, the G-Cloud Frameworks have a wide variety of service solutions catering for a wide variety of customer situations. For many smaller public sector customers, a call-off under an existing Framework Agreement appears to be an attractive option, compared to the complexities of running a competitive OJEU procurement process.
Complexities emerge, however, where Frameworks are used in a multi-vendor context. Currently, none of the central government Framework Agreement call-off terms are drafted for use in a multi-vendor environment, although G-Cloud does now include a short-form Collaboration Agreement. More fundamentally, the ethos of G-Cloud is at odds with the multi-vendor approach. G-Cloud is designed for the provision of stand-alone commodity services. The customer should be able to contract quickly for the G-Cloud services, with the G-cloud contractor providing the services on the basis of their standard services description with no (or relatively little) modification for the needs of the particular customer.
Multi-vendor contracting inevitably means that changes and customisations are required to a vendor’s service offering both to meet the needs of the customer and to make the modifications that are needed in order for an integrated service to be provided to the customer by the ‘bundle’ of tower services. G-cloud service providers may be ill-equipped to make these modifications as they are not consistent with the general G-Cloud service philosophy. It can be something of a shock for a G-Cloud contractor that has been appointed to a tower outsourcing arrangement to find out that it needs to get involved in significant integration activities in order to be part of an overall delivery solution for its customer.
Role of the SIaM Contractor
The role of the SIaM Contractor is an essential component of the public sector multi-vendor environment. The SIaM contractor should take the lead in sorting out the technical, operational and commercial integration issues that arise in the implementation and operation of the multi-vendor arrangement. In the private sector, these activities are often carried out by the customer itself but in the public sector the necessary skill-sets were generally outsourced to the private sector some years ago as part of the first wave of outsourcing processes.
Many of the skills and expertise required for the SIaM role are reasonably commonly available in the private sector. They are similar to the skill-sets needed for the prime services integration of a services supply chain. However, there is a distinct lack of interest in the service provider community in acting as SIaM contractors in public sector projects.
For larger service providers the role is not particularly financially attractive. In most instances the SIaM Contractor is not allowed to bid to be a tower contractor and the tower contracts are always of a much higher value than the value of the SIaM contract. A specialist SIaM contractor market-place may emerge over time, but there are few signs of this at this stage.
One reason for this may be that, although the SIaM role is not particularly lucrative, it can carry a significant amount of financial risk. The customer is reliant on the SIaM contractor for the overall integration of the tower services. As such, there is a tendency for the customer to look to the SIaM contractor to guarantee the service delivery arrangements. However, this is beyond the control of the SIaM. The SIaM can only take the steps that are provided for in the integration agreement for the management of the services. The liability of the SIaM contractor should be assessed on the extent to which it has carried out these steps, rather than whether the tower services have been properly provided to the customer. The SIaM contractor is not in the position of a prime services integrator. Its liability should be assessed on that basis. Its limit of liability should be assessed on the basis of the commercial value of the SIaM services contract, rather than the value of the services to the customer.
Full or ‘Lite’ Integration Agreement
An integration agreement is a key component of the public sector multi-vendor operating model. It should be entered into between the customer, the tower contractors and the SIaM contractor. At the very least, the integration agreement should establish commitments between the tower contractors that they will cooperate between themselves, with the SIaM contractor and with the customer in order to facilitate the provision of integrated services to the customer.
Quite frequently the cooperation obligations are reinforced by a Cooperation Code of Conduct. The Cooperation Code of Conduct is generally not legally binding on the parties but it sets out the parties’ expectations on the manner of their conduct in their dealings with each other. It is particularly important to establish these ‘ground-rules’ where the tower contractors are business competitors. If they are competitors in other circumstances, there is a risk that there will be a deal of ‘finger-pointing’ and ‘point-scoring’ in the conduct of the inter-contractor relationships. The Cooperation Code of Conduct helps to provide a baseline to the parties’ expected conduct.
The integration agreement must also establish confidentiality obligations between the tower contractors. It is inevitable that the tower contractors will come into contact with the confidential information of the other tower contractors. The integration agreement needs to give assurance to the tower contractors that their confidential information will be both properly protected by the other tower contractors and that the other tower contractors will not be entitled to use such confidential information for any purpose other than the provision of the services to the customer.
Beyond these issues, the integration agreement can also deal with intellectual property right matters and inter-contractor liability matters. Not all integration agreements deal with these issues and these ‘full’ integration agreements are more complex and difficult to conclude. However, by dealing with these issues in the integration agreement a road map is established on how to deal with the issues as and when they arise in practice. Integration agreements ought to deal with these issues but if the customer’s negotiating leverage is relatively low (such as where the tower service contracts are awarded by means of call-offs under Framework Agreements on a sequential basis) it may not be possible to negotiate and enter into a fuller form of integration agreement.
If the IP rights issues are dealt with in the integration agreement then the key issue is for each tower contractor to grant a cross-licence to the other tower contractors granting a right of use to the other tower contractors over their IP rights to the extent necessary for the provision of the services to the customer. This ought to be regarded as being the corollary of the confidential obligations. It should not give any rights of usage outside those which are necessary for the provision of the services to the customer. However, as IP rights issue discussions often become emotionally charged, these provisions can be viewed with suspicion.
The integration agreement ought to deal with the treatment of inter-contractor liabilities. If the issue is not dealt with in the integration agreement there is a significant risk that the customer will become liable for inter-contractor liabilities on a de facto basis, without realising that it has incurred these liabilities and without making any provision for the liabilities. My previous article on ‘Multi-Vendor Outsourcing: Approaches To Liability Allocation‘ deals with this issue in some detail and gives some background to the alternative approaches of dealing with inter-contractor liabilities.
Options and Variants
The extent to which options and variants can be allowed or encouraged in multi-vendor procurement processes needs careful consideration. The evaluation of options and variants can become extremely complicated, with many alternative solutions having to be considered in parallel.
For example, a bidder may offer a more advantageous price if it is awarded two tower contracts rather than one. However, if that contractor was not the winner for both towers in the procurement process there is a risk that the bidder that offered the most economic tender for one of the towers may have a right to challenge the contracting authority. As a result, it is very important that the contracting authority makes very clear in advance, in the tender documentation and the evaluation methodology documentation, the various parameters that will apply to the evaluation of options and variants.
It may appear to a bidder that the approach to options and variants is rather heavy handed and allows very little flexibility for innovation and creativity. However, the complexities that can emerge in evaluation if many alternative approaches are allowable needs to be carefully considered in advance and an approach needs to be identified which is workable for the contracting authority.
Overall Approach
The approach to multi-vendor contracting by the UK central government has significantly increased the complexity of the IT procurement process for these customers and has increased the costs and complexities of bidding into the processes. This is directly contrary to the Government’s desire to simplify IT procurement processes, to speed up the processes and to enable SMEs to have a greater involvement in public sector IT service provision.
These are all laudable ambitions but the approach to multi-vendor contracting is not helping to deliver on them. Central to the simplification of these processes is the in-house development of the SIaM role and the acceptance of commercial integration risks by the public sector. The current approach can probably be seen as being transitional in the sense that it will take time for central government to build up its own SIaM capabilities. Some departments are actively seeking to do this. The acceptance of commercial integration risk is more difficult for central government, although departments are accepting these risks in many instances on a de facto basis without being fully aware that they have accepted them.
It will be interesting to see how these issues are dealt with as problems emerge in implementation and operation of these multi-vendor arrangements.
Roger Bickerstaff is Chair of SCL and a Partner at Bird & Bird LLP, London
