4 steps to building your data management foundations

From housing associations to healthcare trusts, community-focused organisations need high-quality data to drive decision-making. Amplifi looks at the four core steps you need to build a solid data foundation. 

Data can be a force for good in community-focused organisations, giving you better insight into every aspect of your operation: from the people that you support to the finances that make your services possible.

Yet as we cover in our guide, Data for Good: A guide to enhancing your organisation’s social impact with data, there can also be a dangerous side to data for the likes of healthcare trusts, housing associations, charities and local councils. Holding so much personal, confidential and critical data carries certain risks. If that data isn’t accurate, it has the potential to do more harm than good – creating distrust, influencing poor decision making and damaging reputations.

The difference lies in how your data is managed. With robust data management, the reward for getting data right will far outweigh the risk of getting it wrong – but there’s so much more to it than investing in a data management platform or carrying out sporadic data quality exercises.

Data for Good Blog 2 Image 2

To get the accurate, reliable data you need to make informed decisions, create better experiences, tighten up your operations and make your organisation as effective and financially efficient as possible, you need to build a data management foundation that doesn’t just tackle technology but changes the way you manage data for good – from infrastructure to cultural acceptance.

Every organisation will need a different approach to data management, depending on their goals, their structure, the data they have and the purpose of their organisation, but there are some steps that every community-focussed organisation will need to take to build a strong data management foundation.

Data strategy: what do you want to achieve with data? 

You’d be surprised how many organisations launch into a data management project without a strategy. It rarely ends well, at best resulting in outcomes that aren’t fully relevant to organisational goals, at worst wasting significant investments on the wrong processes and technologies.

Before you do anything, ask: what do we want to achieve with our data? Think short and long term – are you looking to improve customer insights, to open self-service portals? Do you want more detailed financial data, for reporting and compliance but also to improve operational efficiency? It’s important that this isn’t left for one person to decide – your data strategy needs to be a collaborative effort amongst all key stakeholders, not isolated to an IT or technology team.

While the details of your strategy will depend on what you want to achieve both urgently and in the future, many of the foundations will remain the same. You’ll always need good quality data, data governance and a strong data culture, but how you apply them and the technology you use will change. Once you’ve asked that initial question, you’ll be able to outline exactly what the rest of your strategy will look like – from addressing the relevance of your existing data to changing organisational attitudes to data.

Data Quality: is your data up to the task?

Poor data quality has been blamed for rent non-compliance in Housing Associations, financial strain and overwork in the NHS, mistrust in government departments and local councils and countless other mistakes, miscalculations and misestimates across community-based organisations. As organisations become more reliant on data, poor data quality opens up more opportunities for errors in every area.

At the risk of stating the obvious, the quality of your data determines the quality of its output – or, as we have said before, “rubbish in, rubbish out”. Data quality is determined by how complete, unique, consistent, valid, accurate and timely your data is. If your data is riddled with duplicates, missing fields, out-of-date information and invalid data formats, you won’t be able to trust the reporting, decision-making, automation and other actions that come from it.

Say your goal is to establish a self-service portal for your users (tenants, patients, donors, etc). You want to improve customer experience, but you also want to use the data you collect to guide decision-making on services. To do so, you need to make sure that your data is:

  • Complete – the same critical data is available on every customer, with no missing information.
  • Unique – the customer is not duplicated, either as a whole data set or a partial data match.
  • Consistent – data meets the same criteria and is consistently available, e.g., every customer has the same key details available.
  • Valid – the data meets your organisation’s data rules and is in the right format for processing.
  • Accurate – the data reflects reality. For example, data on disabilities in housing association tenants accurately reflects the person’s real-world experience.
  • Timely – the information has been updated or checked recently, such as new contact details or name changes.

If your data doesn’t meet those attributes, then that self-service portal won’t work – whether it’s by providing users with the wrong options and wrong information, or by prohibiting internal access to that information where it’s needed most.

A data quality initiative is the first step to tackling data quality, but it won’t maintain it over time. For that, you need Data Governance.

Data Governance: can you maintain data quality over time?

Data Governance is part people, part process. Successful data governance does more than present a list of rules that people need to follow to meet data quality definitions – it’s the “hearts and minds” aspect of a good data strategy, that promotes buy-in and changes organisational attitudes to data.

You can, of course, just lay down the data law and threaten consequences for those who don’t follow it, but this approach rarely works. The best data governance establishes the reasons and goals behind the project and demonstrates why they are needed, as well as how to achieve them.

Research in 2021 found 33,645 data breaches caused by human error in local councils alone. It’s important to remember that people can be the biggest problem when it comes to maintaining data quality: use your data governance to get everyone on side and working towards shared data goals, as well as tighten up your data processes.

Data for Good Blog 2 Image 1

Data management technology: what platform is right for your needs? 

Technology alone won’t fix your data management issues, but it will help, and getting the right platform for your organisation depends on what you want to achieve and what will work best for your organisation. Centralised data repositories like MDM are good for making data accessible and consistent across various departments in an organisation, but they will only work if you have the right processes, education and buy-in across the organisation to drive them, as well as the infrastructure needed to support them.

You’ll also need to select the technology that best delivers against your objectives. A common mistake that organisations make when selecting a platform or vendor is to choose the most popular, or those that tick the most boxes in the Gartner Magic Quadrant. Yet even Gartner recognise that this isn’t the best method of selecting a vendor, which is why they recommend partner-led selection to make sure you get the technology that is best suited to your needs.

Download the guide, Data for Good: A guide to enhancing your organisation’s social impact with data, to find out what you need to do next to make data a force for good in your organisation.

Guide: Data for Good

Amplifi Data For Good Guide Mockup