Channelling Sherlock Holmes

It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts.

A Scandal in Bohemia

You’ll be glad to hear that we, like Sherlock Holmes, are not going to pick our content for the Beta newzealand.govt.nz website with the help of gut instinct or divination. In line with our commitment to being user-centred, we are using evidence of what people want or what needs fixing to give us a first cut of priority services for launch.

Evidence-Based Decision Making

There are pots of rich data around government and freely available online. This might not be Big Data, but aggregated together, it’s definitely middle-sized.

To help us prioritise our content I checked a variety of sources as they all provide evidence of different areas of experience online:

  • To find out what topics are in demand, I categorised 500 search queries from the current newzealand.govt.nz’s Google webmaster tool into the Beta website topic areas (e.g. ‘Education and training’, ‘Driving and transport’). The analytics and customer query data from the current newzealand.govt.nz supported the in-demand topic areas.
  • To get a feel for what a user experiences when searching for government stuff, I searched google.co.nz using the sub-topics (e.g. ‘Low income financial hardship’) as keywords. This showed up the poor quality of the SEO (search engine optimisation) as I often had to limit the search by the .govt.nz domain to find any government services.
  • To make sure people won’t have to work so hard to find the Beta services, I used the sub-topics as keywords to search the Google keyword tracker tool.
  • To ensure we incorporate services that are in demand, we’re also taking full advantage of the service data from the Better Public Services Result 10 team who are creating and maintaining an overall catalogue of government services that can be used by New Zealanders.

I also got super helpful responses and data from:

  • Ministry of Justice
  • Ministry of Business, Innovation and Employment
  • Ministry of Health
  • Inland Revenue Department
  • Ministry of Social Development
  • Department of Conservation
  • Ministry of Education
  • Parliamentary Counsel Office
  • Department of Internal Affairs

Thank you, helpful peeps at agencies—you know who you are.

I am also immensely grateful for the New Zealand Citizens Advice Bureau for providing their data on high demand services. They provide such a great service with limited resources, so taking time out to help is much appreciated.

What Does the Evidence Show?

It’s no surprise, post global financial crisis, that topics like redundancy, financial help and budgeting are in demand. There’s also a lot of interest in speeding fines and divorce (no correlation there, though??)

Analysing this wonderful data has given us both a list of prioritised topics and high demand services. This means we are putting our immediate effort into what matters most to the most people.

But It’s a Bit Broken or Difficult?

As I mention earlier, the data also show us where people are struggling with government on the web. This helps us prioritise where we need to target our efforts. We also have to take into account how users find information, as our recent testing found, 100% of users ‘Google it’.

Silo Mentality

We found that only 30% of the services under our topic hierarchy have good search engine ranking, but they unfortunately also contain duplicate and overlapping sites and agencies. Users are faced with a maze of potential government paths and overlapping services that they could take on their user journey. The hard work is all on their end.

Hard to Find

19% of the services we found have poor SEO. This meant that I had to filter the results by site: .govt.nz to find the government services. Given the evidence about the high percentage of people who don’t look past the first search results page, this is a major problem.

Normal Person Speak

You say potato and I say po-ta-to? Actually, it’s more like government says ‘vehicle safety’ and the public says ‘car accident’, or ‘unemployment benefit’ vs. ‘the dole’. If ‘brand names’ are used, government services won’t be found in the search. Government’s language has to be the language of the people.

We want more

I’m really aware that there are a whole lot more agencies, call centres, etc., who I haven’t contacted. This round of services for the Beta site is our best effort based on the evidence we have. In keeping with our agile way of working, the topics and headings will be tweaked and shaped in an iterative manner.

So please feel free to contact me with any research you have that can add to our collective understanding and help deliver what people want. Oh, and questions are always welcome.

Comments are closed.

Navigate Posts