According to the latest Experian Data Quality Global Research, almost 63% of organisations lack a coherent, centralised approach to data quality, and more than half (51%) say individual departments still adopt their own strategy. With 92% of businesses citing they still find some element of data quality challenging, this clearly points to a lack of a consistent and centralised approach as one of the problems.
Firstly, let’s start with the definition of data ownership. According to the Data Governance Institute, “Enterprise data doesn’t ‘belong’ to individuals. It is an asset that belongs to the enterprise. However it still needs to be managed”. Hence the concept of ownership arises, with organisations assigning an “owner“ to make the ultimate decision about data. However this doesn’t have to be one operational owner for all data, sometimes it makes sense to federate the ownership in more complex organisational structures, by having different owners for customer, product and financial data, or even regional owners for global organisations. However, ensuring there is someone at the top of the chain, such as a Chief Data Officer, ensures that even federated ownership can be kept in check.
Lack of data ownership can be a barrier to increasing your data quality sophistication.
Not having any data owners can mean different departments make their own decisions when it comes to data. Take the example of the organisation operating in data silos, where the marketing department extracts customer data, cleans it to use before a campaign and then reports on the success of that campaign, taking a PROACTIVE approach to data quality. However a point to note here, none of the clean data actually makes it back to the operational systems. So when the successful customers signed on from the campaign then contact the support centre, their data is still out of data, and this leads to a poor customer service experience. Now the service department embark on their own REACTIVE and expensive data cleanse process, duplicating the effort of the marketing departing and leaking money for the organisation.
Now consider the same organisation with a central data owner for customer data. This data owner, with the support of data governance function, is able to assess when and where customer data is cleaned in isolation, thus securing a more consistent customer experience and also saving the organisation money. Taking a step further the data owner commissions a data audit of where poor quality data enters the organisation and creates the case for preventative data quality technology supported by a training programme, reducing the cost of on-going data cleans. This organisation is already showing a more mature approach to data quality through the concept of central data ownership.
It is not just about having a data quality strategy in place; it’s about doing it in a streamlined, consistent and efficient manner that creates the more mature data quality organisation.
Key tips on how to overcome this barrier:
Who owns the data?
The first action should be to assign the data owner, and identifying best candidates may actually mean understanding how the business works with different types of data. Assigning people whose organisational down line is responsible for that data often makes them best candidates, as they not only sit with the position of authority, but also can fight the corner of that data as a strategic asset. For example, the Head of Product is probably the best person to own product data. However it is important to have a C level executive owner, like the Chief Data Officer to ensure that the data ownership model is fit for purpose.
Collaboration is Key
When data may have a single owner, it is critical for that owner to create a support network of peers and people with vested interest in data. Understanding how your peers in data ownership are fighting their corner can be very useful. These peers may also sit outside your organisations and networking with the wider data industry is a useful resource, especially when there are relatively few roles focussing on data. On the other hand, understanding who benefits from data in your organisation is also critical, and having them on board can help your case in elevating data as a strategic asset.
Put data to work
Facts and figures speak a thousand words, and actually using data to collate evidence on why it is a strategic asset is one of the most underutilised techniques in the industry. I have been shown many a business case for data improvement, with no factual evidence of the state of data and what quantifiable value will improving it bring to the organisation. It is important to link concepts such as data quality to tangible facts and figures. For example evidence statements such as the following can make the impact difference in getting a data improvement business case
Poor data quality takes my team two weeks each month of reactive cleanses to meet our regulatory data submissions. Over a year this costs us £ X00,000. The manual cleanse at the last minute can also mean poor quality control, and there is a risk of regulatory fine that could be up to £Y,000,000.
20% of the poor quality data is linked to our top 50 strategic accounts worth $Z,000,000 projected growth, this data is a risk to that growth plan.
Celebrate your successes
All of the above effort would be futile if there is no success story to tell early on. As a data owner, you are responsible for elevating your data as a strategic asset within your organisation. So it’s not only reporting what is bad and needs improving, but also the value it brings to the organisation when the quality is good enough. At one of our recent roundtables, one of the attendees talked about using the marketing department skills to make data more business-friendly. This person was able to use their communications and PR knowledge to help spread the success of their data improvement programme internally.
How to use technology to your advantage?
- Use profiling technology to uncover the state of data and use analysis techniques to link business value to the data issue, and also uncover patters of root cause. This may mean going outside the boundary of the data to other information such as historical transactions but prioritise and justify action based on business value associated to the data.
- Use data quality and business rules monitoring to track and measure the problem areas through from red to green, it is important to monitor over time and show trends of change, and the value that brings. This means linking business value to the monitoring reports. Use visual reporting to display this progress and success story, and try and piggy back on other platforms. Nothing is more effective than having the reports on the same platform as other critical management information and business KPIs!
To get a full view of the research behind this post, download the full Global Data Quality Research paper.
To understand more about technology that can support data owners in executing their data quality strategies, please visit our Experian Pandora page.