Shape 01 Shape 02 Shape 03

Top 5 Tips of Good Data Management for 2020

Suzanne Coumbaros Suzanne Coumbaros on 8 January 2020
5 minute read

Clean out the garbage

The start of a new year is as good a time as any to remind ourselves of the immutable truth about managing data effectively: it starts with getting the basics right! With all the talk about new opportunities to use techniques like data science, big data, AI and ML it’s easy to become distracted by the latest trends, like the hypnotic temptation of a shiny new toy. However, my advice is adopting a reactive approach to data management won’t do you any favours. Data provides a big opportunity for companies, but it can only be leveraged as an asset if it is managed appropriately and the proper care taken to ensure quality.

Too often we see data silos created by teams and departments operating in isolation and without consideration of the downstream impact of their actions on their company’s core data. This is created by poorly managed processes and negative business practices, with the unfortunate consequence of compromising data quality. The net result is far from reliable, trusted data that can be put to work to deliver valuable insight and competitive advantage. Instead, we see bad data that offers little value, that no one can trust. If this sounds all too familiar, then time to take action is most definitely over-due.

Single source of truth

Strong leadership is needed to align people and processes across an organisation under a single vision. This is the foundation for success in all things business related, including data management. Fragmented IT systems, with their many specialist applications, operating in a mixed economy of cloud and on-premises, create a data headache for any business to manage. A single source of master data must be made available, that is naturally cleaned, de-duped and enriched. This needs to become the standard way of doing things and not the exception.

A single authoritative view of your data is a business imperative, one that will ensure a return of value, as well as help enforce legal compliance, promote better data standards and improve security. When you have core business processes that are reliant on people creating semi-automated processes using applications like Microsoft Excel, it creates a significant risk that must not be ignored. The data in these ‘systems’ is not held securely and will have little to no application support when things go wrong, and when the ‘system’ and data is not being backed up things tend to go wrong very quickly. This also usually means there is an unhealthy dependency upon a single person or a small number of people, which simply does not scale with the organisation. If you feel you are in this downward spiral, it’s time to stop and take stock.

Stop the madness

One thing that scares me most is when I see a company where everyone looks busy, but people are only focusing on the task in front of them and have no idea if what they are doing, even impacts what the rest of the company is trying to do. This is usually the case when the company’s goals have not been fully communicated and understood at an operational level or they are changing constantly. This then creates duplicated processes, shadow IT and poorly managed data, leading to packed board reports full of inconsistent information and charts that nobody is interested in or even looks at.

Make data quality a priority

Instead, all data processes must be aligned with business goals. This focused approach means providing quality information that adds true value and dispensing with anything else, to the benefit of everyone. This is not a reactive approach to data management but supports the strategic vision of the business. Once your company has decided what it must do, you can then focus on the core processes and data to support those goals.

I cannot stress enough the importance of having a single vision and culture for your organisation. The vision is the starting point, from which everything must hang and waterfall down to all teams and departments. I’d like to share one of my personal experiences, of a financial services organisation that reaped great rewards when everyone started working together to align data processes with the business goals.

Financial services use case example

We were fortunate because the leadership team, once inspired, were fully engaged and drove ownership of the data initiative. They were bought into the importance of high quality information and what it could do for them. They understood it meant doing things differently and properly, they just needed help in getting there. Having the backing of the company’s leadership team is the first and most important milestone to drive the change needed.

With the C-level executives on board, we were able to make swift progress. We started by discussing the key objectives for the business in the next 12 months and shortlisted the top three critical things. We then looked at which members of the executive team were accountable for delivering those objectives. Once we had identified that the focus would be on the operations function, I committed to helping the Operations Director with the data they needed for these key objectives. It’s important to start an initiative like this in the right place and not try to do too much too soon.

I tend to focus on the shortlist of critical goals the company must achieve and build from there. We started with the operations function in the first phase, and then I moved on from there by asking the same questions again. I established Executive Data Stewards and Business Data Stewards within the operations department, with specific KPIs linked to their bonus rewards, focused on data quality. It’s important to understand your people and use incentives that motivate them personally. In this instance, it was a monetary benefit but that’s not always the case as not everyone is the same.

We started small and began to see tentative improvements as the data delivered to the board gradually became more trustworthy. The cycle of success then began to accelerate, along with the quality standard of the data. As confidence in the data grew, confidence in decision making also improved. I did this by looking at the data lineage to understand where the problems were and how processes could be improved. All relevant staff members were engaged to support the project through a series of workshops and training sessions. They were then asked to implement the new processes and ways of working that they had helped to develop and that prioritised data quality.

By examining the data journey in detail, we were able to focus on the right things. I drilled down to specific tasks carried out by individual roles, refining them to ensure alignment with business needs, supported data. Once we had achieved success, I was able to hand over fully to the operations department, for them to continuously improve, while I moved on to the next shortlist of key objectives.


Where to focus in 2020 and beyond

So, before we all get carried away with the next big trend in data, my message is simple: getting the basics right in your company data will always be on-trend because doing anything meaningful with it depends on this. This should always be your priority and using initiatives like GDPR legislation as an opportunity to enable best practice, rather than as a tick box exercise, will help you in significant ways. In the meantime, here are my top 5 tips of good data management for 2020.

Top 5 tips of good data management:

  1. Single company vision and culture
  2. Stable, accountable and supportive leadership
  3. Core processes refined, defined and documented
  4. Managed data flows that focus on fulfilling those core processes
  5. People, process and technology fully aligned to deliver the above
Back to Blog
Suzanne Coumbaros
Suzanne Coumbaros

Data Governance Director at Sodexo and Executive Committee Member of DAMA UK. A data management professional with 20 years' experience in data governance, data architecture, data warehousing, business intelligence, data operations, data migration, data quality, data development and data strategy in a variety of organisations.