Bulletins – where do we go next?
Statistical bulletins are a critical part of what we do at ONS. They are read by more people than any other part of our website, and have the broadest range of users.
And while they have been at the heart of what we do for many years, so much about them is based on what we’ve done and what has gone before, and not what we need to do now or in the future.
Digital Publishing, together with statistical colleagues, have been taking a deeper look at bulletins, their role and how they are being consumed by readers.
We’ve been helped by the brilliant work done by Awen Jones, our analyst, to build out a better view of the ONS website through Google Analytics, and by our senior content designer Kieran Forde, who has taken that data to produce real insight into bulletins’ performance.
The plan is now to work with that insight and with partners in output areas to rethink the role of the bulletin. I’ll follow up in a new blog post on what we are planning to do but here’s a rundown of some of the headlines so far.
There is no definition of what a bulletin is and its role inside ONS
What is a bulletin? Who is it for? When should a release be a bulletin or an article? You might have your own view but it’s possible it’s not shared by all your colleagues.
Externally, the word “bulletin” means little to users – user research has consistently highlighted a preference for this to change. Does it even matter that we have products called bulletin, article, compendium?
Statistical bulletins are read by the widest group of users but not written for the widest group of users
Analytics show 40% of all visits to a bulletin are on a home broadband connection and 80% of all visits to a bulletin come via a Google Search. Experts are a small proportion of our overall users but have to be sure that the bulletin works for them – they often have specific needs relating to aspects of our data and analysis so we need a product that routes them to where they need to go quickly.
The inquiring citizen often wants a simple question answered so how do we successfully compartmentalise and signpost so many different needs in one product?
The bulletin has become a multipurpose product attempting to be all things to all users
Some of our most important bulletins take 9.5 times longer to read than people actually spend on the page. 80% of users to those bulletins never read further than the top 25% of the page.
The bulletin as a digital product is still a mirror of print products discontinued many years ago
One in five visits to the ONS website is on a mobile phone. If you haven’t done it already, go and look at one of our bulletins on a mobile phone and put yourself in the shoes of a reader – how easy was it to read? Was it a satisfying experience?
Bulletins are not always meeting the obligations for “statistical reports” as described in the Code of Practice and in Better Statistics, Better Decisions
The Code of Practice doesn’t provide any definition of a bulletin or statistical report but it does say that “statistics producers should use appropriate ways to increase awareness of the statistics and data, communicate effectively with the widest possible audience, and support users and potential users in identifying relevant statistics to meet their needs.”
There are structural challenges that put real constraints on time and resource spent on bulletin creation in statistical output areas
We’ve spent a lot of time speaking with output areas and there are very real challenges around how much time can be spent crafting bulletins from scratch. We are under real pressure to deliver statistics quickly and with insight, and that means we need to have a bulletin production process which is efficient and gives real impact…which neatly leads to the next point.
The bulletin takes sizeable resource to create but the value is unclear with no impact measures in place
A lot of time and effort goes into bulletins but how we measure their success is somewhat fragmented and unclear. We need some joined-up definitions of success.
Over the next few months Digital Publishing will be working with output areas to better understand what users need from bulletins, and to make recommendations for how we ensure we are meeting those needs.
Nothing will be done without consultation and collaboration internally and externally. And it can start with your comments on this article.
What do you think? What should a bulletin be?