Data-driven government requires real action

Data-driven government requires real action

Data-driven government is not new, or innovative – but it is essential for policy and operational decision-making. Despite this being central to the government’s digital strategy for years, we are still struggling to see results, particularly in the outer pockets of the pandemic response.

We know the problems – legacy technology, a skills gap and cultural blockers – but what practical steps can we take today that can really move the needle so that all public services are designed to truly benefit citizens?

We talk a lot about user value in digital transformation, and working with data is no different. It’s somewhat surprising how much money is spent building data platforms without applying the techniques we use when building digital systems. For example, if you’re building a website, you’ll use user research to identify problems, test ideas, and validate solutions, and only then deliver on those requirements.

Across the public sector, many expensive data platforms are built with a “build it and they will come” mentality. This ignores what people actually need, so systems are not adopted and, as a result, those platforms are considered failures.

Instead, we should build our data platform with the same techniques we use when building something digital. If you solve people’s problems, you make things easier for them, and a data platform becomes sticky because there’s a reason to use it.

Finding balance is important to both in creating a creation Data platform But use it. The data space moves fast, and it’s worth remembering that what you’re building will only last so long. This means you have to balance this new-vs-old mentality. But, equally, you don’t just want to add new tools to the toolkit. Iterate until you really need to deliver value to the people you’re designing for. Creating space for innovation is crucial – but don’t underestimate new technology fatigue.

Create a shared language

A significant blocker to the adoption of data-driven practices is the lack of a common language. It is very easy for one word to have multiple meanings within an organization. Moving to a domain-driven, commodity view of data can help.

We’ve found domain-driven development to be a great starting point. It’s a concept that allows you to see your organization as a set of bounded domains and identify the roots of terms and their meanings This allows you to create an organizational data model to clarify meaning and create better conversations between teams.

Once you really understand your organization in this way, you can begin to develop a common vocabulary, where the words “person” and “asset” have the same (or at least an agreed-upon!) meaning to everyone.

The reason data platforms fail is rarely because of technology – it often is Culture behind its use. Even something as simple as ownership can cause problems. It is usually clear to the parties that they are responsible for the data in the database that they will view. What is less clear to them is that once the data is copied to the platform, the data still belongs to them.

Helping teams feel connected to the platform, as they use it to solve a problem, will give them a reason to care once their data is in there. This also extends to administrative and legal issues – just because data is copied doesn’t stop the party being responsible.

There is a cultural aspect to making sure you train all your people to be data literate as well.

Data literacy It can take many forms, but you’ll never become a data-mature organization if you haven’t undergone a cultural shift.

Your technology will replace the architect

Creating replacement technologies is hard to do in the digital and data space. There are some foundational pieces that you should keep in place that won’t change. But, equally, try to use open technologies as much as possible.

There is a cost with open source frameworks. They are free to use, but they can be expensive to maintain. But using open technology, you can take your data and move it from one system to another without rewriting everything.

The days of having a governance committee that reviews all things data feels antithetical to what we’ve done in the digital sector.

We, as a community, should agree on what policies we want to apply to our data Agreeing on a definition of “adequate testing” and finding ways to share/contract data schemas is much more effective than taking control of a remote panel. Shifting responsibility to the teams building the products is a huge step toward true data maturity.

You never create an API [application programming interface] for your customers and then change the interface without alerting them. With APIs, we use techniques such as version numbers or upgrade paths to ensure continuity or service. Unfortunately, this is not always the case with data.

Often, we find that data is collected from points within a system that are not actually intended for use, so the schema of the data can be considered bad or even worse, changed without notice. By creating data as a product, where it is intended and designed for use by others, we can prevent this problem.

This comes back to ensuring that the team that creates the data owns the data They need to maintain and take care of it, otherwise people won’t use it and it will cost the organization time, money and effort.

Focus on the ethics of personal data

Having access to an individual’s name or address often seems critical to completing a portion of the analysis. In most cases it is not. Only people want to see names and postcodes that seem familiar to use instead of a column of random numbers, but from a mathematical perspective, it hardly makes any difference. In fact, we usually convert them to numbers to use values.

Wherever possible, we should ask questions when we see personal details, and more secure features. By default, we should not have access to them and should alias our values.

As data platforms become more mature and people start using machine learning, ethics become more important. One of the only exceptions to the pseudonym rule is to ensure that any selected training data is representative of the population and contains no bias. Even in this case, we would not be able to identify an individual, but only know enough to evaluate the data for bias.

Data continues to be a hot topic in both the private and public sectors. And while all the aforementioned foundations come from a technical perspective, they are particularly applicable to pockets of legacy-oriented portals within our public services. If we want to realize the benefits of data-driven government, we need to lay the groundwork – and there’s no better time to do it.

Jim Stamp is a data principal at Med Tech

Source link

Leave a Comment

Rams vs. 49ers Black Panther: Wakanda Forever Trailer Eatman: Believable or Not, the Turnaround is Real Monica Bellucci, happy birthday! New MW2 Grand Prix Map Images