In my line of work, keeping track of the threads is half the battle. At work, we have (for now) three corporate sites, a sandbox, a development environment, and more. We have social media channels – some corporately-managed, many managed by external agencies in support of our campaigns – and an active stakeholder and media community who like to talk to us and about us, along with ten busy ministers.
We’re also expected to respond quickly to news stories which break in the media on the issues we cover, as well as be responsive to our colleagues in the Press Office, including helping them to monitor and evaluate the reach of their material online.
So ever since some nice chaps from the Foreign & Commonwealth Office blew me away with an internal dashboard they had developed for this purpose, I’ve been keen to set up something similar. Something which I can have open all day and which lets me see quickly if our sites are up, what’s hot on them right now, who’s sending us traffic, and what we’re putting out there in terms of news releases, tweets and multimedia.
I’ve stolen their idea pretty much wholesale, tweaked it slightly towards social media, and come up with this (click to open a larger version):
1. Site availability: we have Pingdom monitoring set up watching our various domains to measure their uptime, and this box uses its API to tell us what’s up and what’s down. Green is good.
2. Popular content: Google Analytics has a little-known API and the excellent GAPI PHP library to help you access it. In more or less real time, this box lists the top 30 pages on the site today. There’s a lot more to the API, which I might write about another time.
3. Top referers: if there’s a spike in traffic, chances are somebody important has linked to us – this shows a list of the top 20 referers today, again powered by Google Analytics.
4. Search engine keywords: More Google Analytics goodness, this shows the top 20 keywords people entered into Google recently which sent them to our site.
5. Custom Site Search keywords: Slightly squiffy this, as the Great Google haven’t quite sorted out their own technology, but in principle this shows the popular search terms people have used within our own site search (which is powered by a Google Custom Search, covering all our key domains).
6. News Releases we’ve issued: using the RSS feed of our news releases which we retrieve via COI’s News Distribution Service, this lets me keep track of what press releases have gone out recently, to help cross check against popular pages on the site and to help us know when to press the button on digital activity in support of them.
7. Social media output: powered by the RSS feed of our FriendFeed account plus some PHP jiggery-pokery, this is maybe the box I find most useful. At a glance I can see new YouTube videos we’ve posted (in red), Flickr sets (navy), and corporate tweets (gold). The aqua boxes show me what agencies are putting out there as part of our marketing campaigns.
8. Replies and mentions: it’s useful to see what people find re-tweetable and how they respond to tweets from @bisgovuk – this box runs off the RSS feed from a Twitter search.
9. News coverage: Not enough for full social media monitoring of course, but for those reports which do mention the Department by name, this RSS feed from Google News Search provides a helpful list, right next to the news releases which they often refer to.
10. Blog coverage: Often a surprisingly different focus from the mainstream media mentions, this box runs off an RSS feed of Google Blog Search results.
11. Our issues in the news: Believe it or not there’s a world beyond our doors, and this aggregated feed (a bundle of RSS feeds from sections of BBC News online relevant to our policy areas, gathered together and shared out again via Google Reader) helps me keep track of the big stories.
So there you are. I’ve been refining and tweaking it while I road test it over the last few weeks. It’s surprisingly simple (around 500 lines of PHP all told) but helps me get on with more interesting things while keeping half an eye on the shop I’m supposed to be minding. And there’s a hint of geek cool in there too. Whatever gets you through the day, eh?
n.b. This code was developed in my own time, using my own resources and information, and is not Crown Copyright. I’m happy to offer anyone who wants one (including my employer) a royalty-free, non-exclusive licence to use it, bearing in mind it’s early code and I can’t provide much in the way of support – for now, just leave a comment or drop me a line if you’d like a copy.