How the Service Innovation Lab used Kumu to share learnings from their design research.
The Service Innovation Lab often undertakes design research into topics that impact multiple agencies and the teams within them. That means it’s important to share our learnings in a way that supports a number of work streams and potentially reduces the need for repeat user research.
Testing ways to share our findings
While many other people are interested in the specific topic/s associated with our work, such as accessing services, identity and privacy concerns, our project team thought the Becoming an Adult project would be a good scenario to test a different way to share our findings.
The usual outcome of our research would be a report and/or presentation; this provides a highly curated view of the data relevant to our project objectives. Furthermore, in the process of doing field research you often learn a bunch of things not specifically related to the project objectives — but this data might be useful to others.
This prototype looked at how to share the collected data with others to assess its connection and usefulness to their work. This involved digitising all the data we found compelling from our transcripts, even where it didn’t have a direct link to our key insights.
Visual presentation
We met potential consumers of this data and realised it would be useful to present our findings visually. The goal was to provide an at-a-glance view of any data in our transcripts and enabling potential consumers to easily and quickly search or identify topics of interest.
This would enable other project teams to not only decide if there was relevant data but understand our logic — from insight to verbatim quotes that could then be referenced in transcripts for full context.
Combining Kumu with Google Sheets
The tool we chose to test this idea was Kumu. However, we didn’t want to be bound to a particular tool, so we chose to create a database using Google Sheets that could be tested with other tools.
Combining Kumu with the Google Sheets app dramatically sped up the process of data entry and validation, as both apps could be accessed and updated by multiple users simultaneously. This enabled quick updating and version history of the data and visualisation.
Digitising the data
We estimate it took two team members a day and a half to enter all the data from post-it notes into spreadsheet format, ready to be imported into Kumu. At this point the data in Kumu reflected what we had analysed as a team and posted on the wall.
From here, time was spent familiarising ourselves with Kumu and how we could further synthesise and arrange the data beyond what’s viable with post-its. This is when we really started to notice the power of Kumu to highlight areas in our data that needed attention and when data entry errors were corrected.
We re-drew how we wanted the data to be organised, and revised hierarchies on a whiteboard to help inform data cleaning and create a less confusing visual.
Presenting data in a shareable and reusable format
Overall, this process felt faster to get to a shareable and reusable format for our target users (researchers and designers). The process of digitising and recreating the synthesis also added extra rigor to our analysis as it was an opportunity to question how we categorised and interpreted the raw data.
Kumu enabled the data to be more intricately connected using tags and connections, reflecting a realistic complexity to what we learnt and made the often abstract process of analysis and synthesis more tangible.
When we shared this, there was excitement at seeing this type of data presented in such a way and initial reviews are that it feels intuitive, and “certainly a lot easier to go through than other data sets I’ve been given.”
Video transcript
Kia ora, welcome to the demo of the Becoming an Adult data in Kumu.
00:05
You're currently looking at the home view for the project here.
00:09
You can read an overview of the project and a guide to how we've structured the data.
00:14
The project objectives and some context around how the data was collected.
00:19
The data structure here helps to align with this legend which color codes throughout the Kumu.
00:28
You'll see that there's some objects floating around the outside. These have not been connected to help simplify the structure.
00:42
Before we started defined what types of data we had and how we wanted to arrange that hierarchy. So this structure has been somewhat controlled by us.
00:52
But also the Kumu itself making its own interpretations what you can see is when I try and move any of these nodes, it'll try and put them back where it thinks they're most relevant.
01:09
This menu at the bottom allows you to focus in specific data types within the context of the whole Kumu structure.
01:18
So here I've selected just the themes to make it a little bit easier to understand how different types of data are connected.
01:31
You can select more than one at a time
01:38
At a glance you can see that there's particular areas of focus they have a lot of data around them.
01:49
We have clipped the text to aid in navigation
01:53
But at any point, you can click a note and display it in full by revealing a sidebar on the left hand side by following a branch.
02:08
You will see that we go from loose data on niche topics related to most people or categories to things including many individuals data.
02:25
As we get close to the center. We reveal more concrete items such as devices that people might own or ID's people might have.
02:40
At any time you can hover over a node to highlight relevant connections.
02:56
Now we'll take a look at a node in detail. Let's view an insight for an example.
03:06
When work is selected you can reveal the full information about that node. We've got a description to explain a little bit about what we learned, as well as tags that are relevant to this description and insight.
03:22
You can hover over the tags to reveal them within the data.
03:30
If you think click a tag that will show you all instances of it within the Kumu structure.
03:51
This can also be put together by using the focus menu.
03:59
This feature makes it a lot easier to read through the relevant results and under this understand the full context on a particular issue aAnd clear that to get back to the main menu.
04:19
You can use that same feature to highlight the results of a particular person so I've selected Hannah.
04:29
Selected her tag.
04:31
Taken all information from Hannah and then again that can be put together by the focus feature so you can look at a particular individuals data and understand differences between different people easier using that feature.
05:01
Another way you might like to search the data was by using the menu to search fans specific terms that you might be interested in.
05:12
For example, we'll try driver's license.
05:18
You can see as I'm typing it is already searching within all the data we have entered into the Kumu.
05:25
Let me go for this one looks very specific.
05:30
This is revealed the driver's license node as it is an ID.
05:36
I can now see who had that ID.
05:39
And then by revealing it in the sidebar.
05:44
Selecting it's tag.
05:48
Focusing on their data I can see all mentions of driver's license in one place.
05:55
At the same time Kumu has selected specific parts of the structure.
06:00
To bring into this view to make it easier to navigate all these points here are relevant to the process of using or applying for a driver's license.
06:13
You can also see that the driver's license ID node has been brought through this menu.
06:25
This is a very quick way to discover if topics you are interested in have been covered in our conversations and allows you to see how important they were and how often they were discussed based on the amount of data that's poured into your focus view.
06:43
That's a rough overview of how our data can be viewed in Kumu so that's all I wanted to show for today.
06:51
Please be in touch if you have any further questions.
If you have questions or would like to know more about Kumu or insights from the Becoming an Adult project, please contact the Service Innovation Lab
Service Innovation is part of the Department of Internal Affairs. It’s charged with working collaboratively across agencies to centre government services around the needs of people. You’ll find us either at ‘the Lab’ on Thorndon Quay or Level 10, 45 Pipitea St in Wellington.
If you’d like to stay across the work from the Service Innovation Lab, please join our mailing list
Published
Categories
Collaboration, Content management, Testing, User research, Service Innovation Lab,
