12 CCS delivers an important service to all departments and the wider public sector, and saved organisations £521 million in 2015-16. CCS measures savings against what would have been an equivalent spend in 2014-15. Our review confirmed that the savings reported by CCS were generally supported by an appropriate method and documentation. Departments endorsed the new calculation of central government savings. However, these savings were calculated on a different basis and are not directly comparable to the planned net benefits of £3.3 billion over four years and we could not tell whether these savings would have been achieved if the central buying functions had not been transferred to CCS (paragraphs 2.4 and 2.6 and Figure 11).
13 However, some customers complain that CCS's services can be poor quality. Although CCS has helped its customers to save money, customers have consistently reported that they are not satisfied with CCS's performance. Customers have reported issues including poor communication, unreliable services and the way CCS has managed procurement frameworks. CCS itself reports that service delivery has not always met service agreements. When we explored these issues with central government customers, most of them told us that CCS adds value and praised the knowledge and skills of individual staff. However, they think that CCS needs to improve the way it manages and communicates with customers. Departments also said that the quality of CCS's services was highly variable (paragraph 2.11 and Figure 12).
14 Our work in summer 2016 found that CCS's management of its services has not supported consistent value for money and quality. In particular:
• CCS's services were not integrated or standardised. CCS was formed by merging the Government Procurement Service and parts of departmental procurement teams. As departments had different procurement models, this approach resulted in CCS inheriting a diverse set of procurement services. CCS then agreed individual service levels and charges with departments. It was only in June 2016 that CCS began to standardise and commence the integration of services. This variability has made it difficult for CCS to consistently achieve high performance and constrained its ability to grow in line with its original forecasts (paragraphs 2.19 and 2.20 and Figure 16);
• CCS cannot demonstrate to its customers that its deals are always the best available. CCS benchmarks its deals against historical prices but has limited current market benchmarks to demonstrate that its central arrangements offer best value for money. Furthermore, CCS does not manage the lifecycle of procurement frameworks well: it has extended 21 of the 39 frameworks we identified as due to expire in 2015-16 without market testing or competition. It is also extending frameworks after having already exercised all options for extension, and CCS and departments are sometimes using expired frameworks to let contracts. Purchases made under extended frameworks are a risk to value for money as the prices paid may not be best value in the market. Contracts issued under expired procurement frameworks contravene public contracting regulations (paragraph 2.14 and Figure 13);
• CCS's control environment has been weak but shows signs of improvement. CCS inherited a limited control environment from its predecessor, the Government Procurement Service, and attempted to merge teams from seven departments into it. In CCS's first two years of operation, its internal audit team repeatedly found weaknesses in CCS's internal control environment. More recently, our external audit engagement with CCS has shown clear signs of improvement in governance, risk and internal control (paragraph 2.18 and Figure 15); and
• CCS's has not fully developed the way it manages processes. Our review of CCS's process management found that CCS has some characteristics of an organisation that manages its service delivery effectively. For example, CCS has documented some of its processes and seeks customer feedback to improve performance. However, we found weaknesses. For example, until 2016 CCS lacked internal control mechanisms and appropriate technology such as workflow management tools to prevent staff from using expired frameworks to issue contracts. CCS also does not assess new work requests consistently to confirm that the requested services correspond to what it has agreed to deliver (paragraphs 2.14 and 2.17).