How much do your agency websites cost - and are they cost-effective?

I have long struggled with techniques for costing websites in Government. Due to how resources and budgets are allocated - with program areas funding and conducting some content work, corporate areas other and infrastructure and network costs often rolled into a central budget in IT teams (which provides excellent economies of scale, but makes costing individual web properties harder) - it can be very hard to come to a complete and accurate figure on what any government website costs to launch or maintain.

Regardless, we are all driven by budgets and must determine ways to estimate costs for planning new websites and set management, improvement and maintenance budgets for existing ones.

A step further than costs is value, a necessary part of any cost-benefit equation. In order to assess whether a website is cost-effective - or at least more cost-effective than alternative tools - it is vital to be able to demonstrate how websites add value to an agency's operations.

Unfortunately value is an even more nebulous figure than cost as it often has to measure qualitative rather than quantitative benefits.

Sure you can count the number of website visits, visitors or pageviews, or in social media terms, fans and followers, however this is much like judging a meeting's success by the number of people who show up - the more people, the more successful the meeting.

This metric works when you can place a commercial value on a visit - so this may work effectively for ecommerce sites, but not for most government sites.

Another approach is to look at the cost per visit, with a presumption that a lower cost is better. However this relies on fully understanding the cost of websites in the first place, and also assumes that a cost/value ratio has meaning. For some websites a high cost might be appropriate (such as a suicide prevention site), whereas for other sites a lower ratio might be appropriate (such as a corporate informational site).

Perhaps the key is related to that ecommerce site example, where the sales of goods is an outcome of a visit, therefore the commercial value of a visit is effectively a site outcome measure.


The next challenge is assessing the outcomes agencies desire from their websites and giving them some form of quantitative value. Completing an online form, rather than an offline form might be worth $5 to an agency, reading an FAQ and therefore not calling or emailing an agency might be worth $30, reading FOI information online rather than making an FOI request might be worth $500, whereas reading emergency news, versus having to rescue someone might be worth $5,000.

Of course this quantitative measure of values for outcomes is relative and has very large assumptions - however it does provide a model that can be tweaked and adjusted to provide a fair value of a site.

It also has a far more valuable purpose - it forces agencies to consider the primary objectives of their website and how well their most important outcomes are satisfied by site design, content and navigation.

If the main purpose of a site is to provide information on a program such that program staff aren't responding to calls from media and public all day, then the appropriate information needs to be front and centre, not hidden three levels deep in a menu. If the main purpose is to have people complete a process online, then the forms must be fillable online and back-end systems support the entire process without having gaps that force people to phone.


Are there other more effective ways of measuring cost and value of websites? I'd love to hear from you.

And for further reading, the posts from Diane Railton at drcc about UK government website costs are excellent reading, How much does your website cost?