Monitoring Data Publishing

It is relatively simple to measure certain aspects of open data publishing, such as:

What data has been published. For example, a simple count of the number of datasets that have been published on a portal, sometimes classified by subject matter, or by technical criteria such as format;

- The quality of the data. For example, the proportion or number of datasets that are published with acceptable open licensing arrangements, or with an acceptable completeness of metadata;

- The number of visitors to your data platform and information about them that can be obtained from portal logs. For example, which browser or device are they using to access the portal and in which country are they located;

- Access to data. For example, which pages on the portal do visitors look at overall and the number of downloads for specific datasets.

These measurements are very useful. They help assess:

1. The rigour and efficiency of the publishing process, e.g. data quality reports will indicate if the process is publishing data with an appropriate license.

2. The design of the portal, e.g. an analysis of pages accessed may show how easily users are able to find datasets.

3. The choice of datasets, e.g. portal logs may show if there are many datasets that are rarely accessed, or those which are particularly popular.

However, they don't reveal much (if anything) about how the data is being used, or what impact it is having.