Cabinet Office
User testing will show you if your services are useful and usable. However, once the website is up and running, it is vital that meaningful access statistics are logged and turned into meaningful information that can be used to improve your website.
Meaningful access statistics will confirm what services are used and to what level. They can identify the parts of your website do not attract an audience. They can show you how users find their way to your site, and where they go when they leave your site. They can confirm that your site is in good technical order, and can warn you when users are having difficulty downloading certain pages.
Meaningful access statistics will confirm what services are used and to what level. They can identify the parts of your website do not attract an audience. They can show you how users find their way to your site, and where they go when they leave your site. They can confirm that your site is in good technical order, and can warn you when users are having difficulty downloading certain pages.
The Guidelines for UK government websites: Illustrated handbook for web management teams establishes the minimum requirements for access statistics. These are:
It is not acceptable or useful to refer to hits. By counting hits, one is simply totting up all the individual files—including graphic files—that the visitor requests.
Additional useful metrics can include:
For search engines, it is important to know how many searches users have to run before finding the page that they want.
Statistics about uncompleted transactional services—‘dropped baskets’—may also prove informative.
Evaluation should not be confined to computer-generated statistics, but should involve user groups that are representative of the authority’s population. Note that it has been shown repeatedly that what users say they need or prefer is not always borne out by what they actually do. For this reason, it is very helpful to observe actual user behaviour, and to measure objectively a user’s level of comprehension and ability to complete a task successfully.
Conventional market research can be used to reveal the level of local awareness of your website, and people’s attitudes towards it.
Comparing the take-up of web services to other service delivery channels can help establish the relative value of the website to both customers and stakeholders.
The objective measurement of results is another useful way to evaluate your website. If you are recruiting staff online, is your website achieving its recruitment targets? Is advertising online more cost effective than advertising in newspapers? What sort of staff does online advertising attract?does it produce more technically aware staff if that is its goal? Questions like these can be answered through evaluation.
Local authorities should also consider periodic reviews of their website to ensure the website meets the current needs of citizens and business of the authority. This review should cover dimensions such as:
The Guidelines for UK government websites: Illustrated handbook for web management teams Section 1.4 on Evaluation provides definitions of terms used in access statistics, caveats and common problems, and a checklist for auditing and statistical analysis:
http://www.e-envoy.gov.uk/Resources/WebHandbookIndex1Article/fs/en?
The Quality Framework for UK government website design reviews the steps in standard user testing and suggests some cost-effective ways to test the usability of your web services:
Quality Framework for UK government website design