Welcome to our opinion series. This month’s writer is Karen
English, Director of International Procurement and Trade at NZGP.
Data matters – part 2
(To read Karen’s previous article please follow this link)
I want to share reflections on the challenges with
collecting and using data and how we can use data to gain insights to inform
better decision making.
The Procurement for the Future strategy acknowledged we need
to lift our data game. It sounds easy to create a data plan and implement it, yes?
I’m not convinced it is straightforward.
Let me start by telling you about my experiences in discussions
in the World Trade Organisation Agreement on Government Procurement (GPA).
Let’s focus on statistical data parties share to show what
procurement is open to bidding and how much it is worth. This information is intended to show a substantial amount of
procurement is open to international bidders meaning being a member of the GPA
is a worthwhile undertaking. However, when we discussed how GPA parties collect
and report information, we found no common method. Some record estimated spend
(like us) and some record actual, like the United States. We found no common
taxonomy for categories or how categories are assigned to a particular
procurement. Add to this difficulty in separating out the value of procurement
of goods and the value of services, where many contracts are a combination of
goods and services. Or even defining when something is a good or a service. Any
suggestion we could easily create a common method was kiboshed. It would
involve considerable system change for each party. In the GPA context, we
continue to report statistics, but frankly, the best that can be said is there
is some indicative information about the level of procurement activity open to
international competition. This experience highlights challenges with using
data.
Let’s talk generally about collecting data. This is
challenging because there are so many competing demands for data. For example,
Cabinet Ministers are interested in demonstrating their policies are effective
so the information they want is quite specific, but can change as government
policies and priorities change. It often involves setting up bespoke collection
systems, typically at considerable effort for agencies, plus the central effort
needed to consolidate, interpret, and report.
This data is important, but how much does it tell us about
what is occurring in procurement across government? We have global numbers on
how much is spent on procurement across all government agencies, derived from
government accounts, but this does not give a picture of what individual
agencies spend, on what, and with whom.
There are different aspects to data collection and the stories
data can tell us, depending on the audience and what their interest is. So how
much of this do we touch and what is the direction of travel?
The data and transparency workstream established under
Procurement for the Future is working hard to improve our data. We already have
multiple sources of information – for example, the All-of-Government (AoG) data
– but these are stored in different places making it almost impossible to have
a consolidated view. So the priority is to work towards this consolidation of
government procurement spend data held by NZGP. We will then get a consolidated
view of spend. This is only one step on the road to a data nirvana. Other
agencies also hold information without which we won’t have a complete picture. Those
agencies need to bring their data sets into the frame.
Initiatives, particularly technology based, go some way to
filling data gaps, but as with all technology, the base information needs to
get into a collection pond before it can come out – there will always need to be
some effort around this. The rewards can be huge. Decisions based on reliable
data are usually sound and hard to argue against. I am optimistic artificial intelligence can help. It could streamline
processes in collecting, aggregating and analysing data. Human effort could
then focus on delivering better public services.
The philosopher Lao Tzu said a journey of a thousand miles
begins with a single step. In the same vein, every step on the data journey
takes us closer to data nirvana. In the meantime, live, laugh, record, and
report!
