One of the things I’ve been keen to explore – and haven’t yet really got round to – is the way that we can use web analytics on WriteToReply and JISCPress to monitor site usage (easy), improve the user experience (harder), and provide feedback to consultation publishers about the extent to which people have engaged with material on the platform (important for the business model…!).
If we were a commercial site, then a straightforward metric based on sales conversions would be one of the most useful things to report about how effective the site was. (That is, we could measure how many visits to the site were actually converted into sales – people handing over their credit card and closing a sales transaction.)
Working out equivalent metrics for non-sales sites is a little bit harder, and something I’ve thought about before on the OUseful.info blog in the context of course websites, open educational websites and Library websites.
So to get my thinking started about appropriate conversion metrics on WriteToReply, here’s my starter for 10.
A conversion is some positive action that a user must take on the site, such as:
1) posting a comment;
2) obtaining a reference to a section or a paragraph on WTR and linking to it from another document;
3) embedding a paragraph published on WTR on a third party site using some form of content transclusion e.g. via a quotation embed code (which we don’t actually do yet);
4) subscribing to a daily feed from a document and reading it that way;
5) subscribing to a comment feed.
Note that these are not equally weighted/valuable, nor are they are all contributions to the bottom line of getting discussion going around an issue. For example, posting a comment or embedding a quote in a post that discusses the matter at hand generating content that might feed into a consultation exercise, whereas subscribing to a feed is just a random act of consumption (although if a consultation publisher subscribes to a comment feed, then they are at least engaging with this mode of consultation).
We might also identify pure ‘amplification’ or ‘publicity/PR’ metrics, such as people tweeting or retweeting a link to a document, or posting a link to a consultation document in a blog post or news story without actually engaging in any other form of commentary.
What other measures am I missing?