The centralization of a staffing model is defined as the extent to which management has direct control over the staffing function. In a highly centralized staffing function, management may exercise control over strategy, resources, budget, and process. In a decentralized staffing function, responsibility for the recruiting process is managed locally, with strategies and processes that are tailored to local circumstances. The corporate world heavily favors centralization: 59% of companies structure the staffing function on the centralized model, according to “2003 Recruiting Metrics and Performance Benchmark Report, by Staffing.org. Only nine percent of companies deploy a fully decentralized structure. The remaining companies (32%) place the staffing function somewhere between fully centralized and fully decentralized model, wherein some aspects of the recruiting process may be managed centrally and others locally. For instance, 65% of companies characterize the sourcing function as centralized. A centralized staffing model allows a company to pursue a coherent strategy, set out a consistent process, and take advantage of certain economies of scale. On the other hand, a decentralized staffing model may be more focused on the hiring manager’s needs and is more flexible in nature. The challenge for any company is to maintain a level of service for the local hiring manager while setting corporate strategies and standards, and measuring progress towards those goals. It is important for corporate executives to know how staffing is performing, regardless of the structure of the function. What then are the issues pertaining to metrics that are common to, and unique about, a centralized and a decentralized staffing function? Good Metrics Regardless of the structure of the staffing function, we know that to be effective and reliable, the metrics we choose to use need to have these key characteristics:
- Aligned with business goals: Corporate business targets and HR strategies need to be synchronized and then translated into the tactics HR implements.
- Actionable: Too often, HR measures for the sake of measuring. A good metric must provide information that can be acted on, and must have implications for a clear plan of action.
- Consistent: A good metric must be consistent in what it measures and how the data is gathered. The data and methodology must be defined at the outset, and remain consistent; otherwise the value of its comparison is useless.
- Tracked over time: A good metric must be tracked over time, to provide information on the trend in the metric, not simply its state at one moment in time.
Metrics in a Centralized Model Adherence to corporate metrics is commonly mandated in a centralized staffing environment, often to track compliance to a standardized process. Therefore, metrics can focus on the steps of the process in great detail, with the aim of driving incremental improvements in performance. A shared, centralized database can easily be the source of data. A perennial issue, afflicting centralized and decentralized staffing models equally, is that reporting is often an afterthought. If recruiters have to stop what they are doing and fill in data after the fact, then there will be issues with the quality and integrity of the data. Reliable metrics are possible only if the reporting infrastructure is integrated with the recruiters’ daily work. The danger in a centralized environment is to measure and report on too many metrics, simply because the data is available. Be careful not to be too focused on the data. With too much data, it is difficult to know what is wheat and what is chaff. Metrics in a Decentralized Environment Much of the discussion of centralization and decentralization of staffing functions has a subtle bias against decentralization, implying that it is less rigorous and a second-class business practice. But a decentralized model is sometimes a necessity, depending on the business model. For example, a retail company with numerous small outlets typically employs a decentralized structure, suited to the lightning-fast hiring process typical of such environments. Introducing metrics is one way of increasing rigor and raising visibility of a decentralized staffing function. If a centralized staffing function is faced with an embarrassment of riches for metrics, a decentralized function typically is faced with a desert of data. HR generalists and hiring managers commonly are carrying out the staffing work; collecting and reporting data takes them away from the real work of the day. Again, the reporting platform has to be integrated with their daily work, or else the data will be missing, inaccurate, or worse ó made up. By definition, there are few shared processes in a decentralized staffing function. Each division may devise its own procedures. Without shared processes, you cannot craft metrics that measure specific steps of a process. What you have to focus on instead is the output of the process. Find outputs that are common across all divisions, and define consistent and actionable metrics for those milestones. The challenge then becomes working backwards from the outputs into what little data you have to identify ways to improve. Imposing a limited number of standard metrics on the branches of an otherwise decentralized staffing function is a move towards centralized control. Be careful, however, not to impinge on the very reasons why decentralization makes sense for your organization, such as freedom to react to local market conditions. The Role of Technology An enterprise staffing management solution provides the ability to capture and report enterprise-wide metrics. The efficiencies of a shared candidate database and the powerful reporting capabilities of an enterprise solution make a compelling case for centralization. The same technology, though, also enables more effective decentralization as it provides visibility across an enterprise into the demand for and supply of candidates, as well as real-time collaboration among HR generalists and hiring managers. Metrics are the key for HR to become more visible in the organization. In either a centralized or decentralized staffing structure, the requirements for metrics are the same: do not measure for the sake of measurement; keep metrics to a minimum; and measure output and what is in alignment with the strategic goals of the organization.