Answering the Alpha Question?

Every profession has a secret language all of its own that frequently baffles newcomers. Working at the Office for National Statistics after a career mainly based in higher education and research institutions I am consistently tripped up by terminology that seems familiar but is inevitably is not on closer inspection (doubly so as the Civil Service has a language all of its own as well.)

I’ve spent my entire working life in a variety of web related roles and realise I am as guilty as anybody for peppering conversations and blogposts with my own shibboleths. In recent months I have spoken incessantly about the idea of an ‘alpha’ and while I certainly had every intention of explaining myself I’m not sure I have been wholly successful.

So I am going to spend the next couple of hundred words presenting an Alpha 101 that outlines the what and the why of this project.

The idea of an ‘alpha’ stage is well established in software development. Depending on what you believe it can be traced back to IBM in the 1950s. In recent years the ‘alpha/beta/live’ product cycle concept has been particularly popular with teams building internet tools. Recently, and most relevant for us, it is an approach wholly embraced by the Government Digital Service [GDS].

GDS define an Alpha as;

A short phase in which you prototype solutions for your users needs. You’ll be testing with a small group of users or stakeholders, and getting early feedback about the design of the service.”

Notably before the formation of GDS there was an embryonic team that worked on Alphagov [ and if you look closely you’ll see that yes I did create that Wikipedia page back in 2011.] Alphagov not only proved to be the prototype proving the user needs for the Government single domain it also proved the need for a team to build it. Thus the formation of GDS.

A useful guide to working on these prototypes (a term that seems more familiar to people) is available as a part of the GOV.UK Service Manual . This includes this short video outlining a bit more about how GDS think about it;

One of the key elements of this approach is the hardest for people to grasp when coming from outside of a digital focused environment.

As GDS put it;

“Another output of the alpha can be the early decision not to continue into Beta. This represents a successful alpha since it reduces wasted time and money.”

This essentially means an ‘alpha’ can be abandoned or ‘thrown away’. This is not quite as cut throat as it might sound. It is unlikely that any ‘alpha’ will completely fail to deliver any benefits. Even in failure it is providing intelligence that will ensure better decisions are made in the future.

Building something complete enough that you can test it with users and receive genuine feedback early and often is the key. This prevents the project blindly continuing down an expensive and time consuming route before discovering that it is not a match for user needs too late in the process to make real change.

For our project here we are closely following the GDS roadmap for these things – I’ve written a bit before about our ‘discovery’ phase  – which we have actually continued to build on since that point – and now we are about to embark on a short, sharp development phase as recommended in the ‘ideal alpha’ outline.

We aren’t trying to do everything.Our alpha is very much aimed at demonstrating what a new site could look like and work like for our users and as such we will perhaps be more open with it than is usual at this stage of a project. This kind of ‘public alpha’ is rarer but as I mentioned earlier it worked out OK for GOV.UK. We will be trying out new approaches to navigation, search, layout and design as well as trying out something more mobile friendly.

What we are not looking to do at this point is delve too deeply in to the behind the scenes technical decisions we will need to deal with our security and requirements under the ‘Code of Practice’. There will be a parallel piece of work investigating and experimenting around these issues and making recommendations for an approach for the Beta build (assuming the alpha is successful.)

So hopefully that at least explains a little bit about what an Alpha is – really it is just a quick(ish) prototype that is good enough that we can get some decent feedback that in turn allows us to work out if we are on the right lines.

2 comments on “Answering the Alpha Question?”

Comments are closed.