Performance measurement

Where do I start?

Start by identifying a performance measurement framework and establishing what terminology you will use:

• Select a set of terminology (such as outputs, outcomes and indicators).
• Define what each component of that terminology will mean (to minimise debate and dissention later). Document these definitions.
• Apply that terminology consistently.

What's next?

Develop a reporting framework

A reporting framework is the basis for completing the performance measurement process. You need to include:

• who has responsibility for capturing and analysing performance data (both grantmaker and grantseeker can play a role in capturing data)
• a measurement and reporting timeframe (formal performance reports may be produced annually, for example, but performance monitoring may be more frequent)
• the processes for reporting (including how information on individual grants will be aggregated to create program-level reports)
• the management sign-off process
• details of feedback arrangements, such as how management feedback will be provided to individual grant managers.

Obtain agreement from stakeholders

It is important that all of the stakeholders mentioned in your frameworks are given the opportunity to comment on them and then later to agree to them. Obtaining their agreement and participation early in the process can make it smoother overall.

Consider including grant recipients in the process of formulating measurement and reporting frameworks.

Create a measure profile

A measure profile is a detailed description of key information that underlies a measure. For example:

• description
• method of calculation
• target
• responsibility for reporting
• data source
• frequency of reporting
• details of lead indicators (early indicators of things otherwise measured only annually).

Where you are using only a small number of measures, a table containing the information above is usually sufficient. At the program level, or where there are large numbers of measures, a database may be more useful. A database makes it easier to generate lists of measures for which particular parties are responsible, and performance reports.

Using key performance indicators (KPIs)

KPIs can indicate whether program outcomes and outputs are being delivered. KPIs can consist of:

• indicators relating to an individual grant; or
• an aggregation of performance outcomes to provide a program level outcome.

Transforming data into information

Tools that can be used to transform data into information include pareto charts, control charts and scatter diagrams. These tools can structure data in a way that allows, for example, for the identification of:

• key drivers of poor or good performance (pareto chart);
• when variations of performance are significant enough to warrant corrective action (control chart); and
• relationships or dependencies between aspects of performance (scatter diagrams).

Analytical techniques

A number of analytical tools can help draw meaning from data. They include:

• Comparative (gap) analysis. This approach uses a contract or other specifications as the benchmark from which to compare actual performance. It relies on a clear understanding of requirements and comparing this to what has been achieved. This approach requires 'apples with apples' comparisons, so performance information needs to be comparable between the contract/grant service specifications and the grant recipient's measured performance.
• Cause and effect analysis. Also known as root cause analysis, this approach seeks to determine why something was different to what was expected (or why it was the same). It is often used in conjunction with gap analysis to determine where the gap is, then to determine why it occurred. A cause and effect or fishbone diagram is a tool often used to assist with this type of analysis.
• Trend analysis. This approach seeks to determine performance trend over time. It can also be used with the two techniques above. Trend analysis allows grant managers to gain an understanding of systemic performance and identify those one-off or unusual issues that crop up from time-to-time. This approach assists in avoiding 'knee jerk' reactions and allows for informed management.