This week we started the work to “piggy back” on the work already undertaken by Daro in the Data and Insight Team with regard to joining up staff data. In theory it was going to be quite straight forward. In practice (as is often the case), it was a little less so – but not insurmountable.
We started the week with what we called our “quick & dirty” data analysis. As we got in to the exercise, we discovered that it was going to be somewhat less quick, but a whole load more dirty!
In everyday language, what I mean by this was that the data was more inconsistent and unreliable than we had first anticipated. The goal of the exercise was to think about what new insights we might want to try and glean using Qlick, the existing tool for analysis previously configured by Daro. (Both the insights we could get now and the insights we might be able to get by building on the work already done).
The new challenges and questions that started to pose themselves however, were along the lines of how useful these new insights might be and what we might actually need to do to create something meaningful.
It started to create a bit of a chicken and egg question for us… What comes first… cleaning up the data or creating a mechanism to join data up.
There are pro’s and con’s to each position… joining up data gives a more overall coherent picture of what’s there. But equally, if joining up one poor data source to another poor data source, we are left with a set of doubly poor data that is potentially produces an insight that is twice as unreliable as the first.
Ultimately, we took this as far as we could on our own and decided to pose some questions to our wider audience of stakeholders at our show and tell. We specifically asked what insights might be most useful.
This information is most useful to us, because by understanding this, we can get an idea of the level of accuracy that we might be able to achieve based on the correctness of the data sets that such insights belong to and then make a call on whether we think that by joining them up we might get be able to get data that would answer such questions to a level of reliability that will actually help anyone.
If we can’t do that, then the question is answered for us by default.
Talking of the show and tell, we felt much better about this one.
We worked harder at preparing it, assigning responsibility for articulating a ‘section’ rather than a slide and rehearsing a couple of times to make sure that we were all settled and ready for the review.
Collectively we all felt that it went better than the first and we treated ourselves to a box of Quality Street to celebrate!
In terms of how we are working, I’m pleased that at our retro (where above mentioned Quality Street were devoured), the issues that we collectively discussed were all different to those that we had worked on at our previous retro, which suggests that the iterations that we made around trying to balance work had been a success.
In terms of going forward, we are going to try making ourselves more visible to our regular teams of practice and also we are going to have a look at further flattening the team structure by finding ways to more evenly distribute work between us. This means that some of us may try taking on the generic admin tasks of team members who are peaking on technical aspects of the projects.
We’ve also decided to take a weeks hiatus this coming week, as Lisa headed off to another continent to celebrate Thanksgiving and Ian took out some training time to get Scrum Master Certified and Tim is carrying out a swift piece of discovery on another arm of the Managing People Data suite of works. (Tom I am sure is being kept very busy on the Data and Insight Team in Lisa’s absence!)
We’ll be back next week to explore how we might build on the work undertaken by Qlik so far and to think about how we might make Qlik more widely accessible. We’ll also be thinking about what recommendations we’ll be making at the end of the alpha phase on this project.
As always, thanks for taking an interest in this project and if anyone has questions – feel free to reach out and ask any of the team!