What is Data Ownership? Data ownership broadly refers to designated business stakeholders who define how the business data will be used to achieve the desired business outcomes.

Is data ownership necessary? With data ownership comes accountability and responsibility which may not be desirable for teams to take on however, it is a necessity for teams to achieve their objectives. Often times stakeholders make decisions which are not backed by data but based on instinct – imagine the surprise when their data gives a complete 180 degrees decision to the instinctive one. The business decision maker might feel their knowledge and authority is being challenged and therefore immediately refuse to accept the data output.

I am of the school of thought that Data owners are essential to every organisation and I hope you agree – if only for the number of times I have worked for clients who did not have data owners and felt how painful it was to navigate the data solutions, then yes, data owners are absolutely necessary. Otherwise, it’s like sailing a ship without a captain – it’ll end up going nowhere!

Setting aside my personal anecdote, very often data is not properly defined in an organisation leading to user dissatisfaction with the data and lack of confidence in data.

This seems a logical outcome because data challenges existing beliefs and understanding. If the business expectation of what is good for their business area is not what the data is representing, the instinctive reaction from the business may be that the data/information must be incorrect because it doesn’t align with the business world view. This lack of trust is often based on instinct only.

The data team or IT is then approached with questions inquiring into why the data isn’t “correct”. An example would be the number of customers a sales representative is able to successfully register or sign on to a business. This number reported by the sales representative could turn out to be significantly inflated by the end of the month or quarter when Finance or Management teams are interested in looking at the metrics. The difference in numbers may be due to customers canceling registration after an initial trial period expires. The data can be correct at any such checkpoint however, this will be based on the user definition. Finance is a data owner, Sales is a data owner and so forth. Each business area therefore needs to own and define what a registered customer is.

Who owns the data?

The use of data requires both business and IT input with the business owning the data which IT curates.

I recently read an article on LinkedIn which piqued my interest and led me to the question who owns the data and why anyone should own the data ? The LinkedIn article referred to a recent case of the firing of the Chief of BLS (Bureau of Labor Statistics) in the United States due to labour data reporting. This sent waves across the data world. Making a data practitioner take the heat for perceived failure is a real threat to data practitioners.

Data teams are being held accountable ..but are they being listened to? This begs the question; “Would the next bureau chief give a different data or statistic?”

Data cannot be manipulated although the context in which data is interpreted can be changed or chosen. If data owners choose to view the business data on a narrow spectrum of carefully selected data points and limited dataset, the output of the data can be controlled. This however still requires business input in the form of ownership of the data.

Why should there be a data owner? There is the need to understand the business, it’s data and how the data is to be used. The same data can be used to arrive at different outcomes depending on the use or intended use. Another example of same data being used in different ways is in Pharmaceuticals, in order to create a medication, pharmacists will look at the same human cell composition and how they react to the various components of their medicine. One pharmacist may choose to use 10% higher ingredient composition to achieve a cure while the other pharmacist may choose to vary the amount of dosage prescribed. Two pharmacists presented with the same data but the data is being used differently. A pharmacist here is the data owner who prescribes what the data needs are. A data practitioner facilitates the availability of the data in the required format. This can include implementing the metadata, ontology or ensuring the data quality.

If a data practitioner is not a data owner then what do they do?

The business users need to have an understanding of their data – this is helped by documenting the metadata, documenting the context as a data model, developing and documenting the taxonomy and ontology for the organisation – performing data management activities. It is also key to have data governance in place which connects the various data management capabilities. All of these are functions of a data practitioner who work closely with or within the Technology team to ensure the right data is made available to data owners in accordance with their specification.

Often business users do not acknowledge the value of data until there is an hindrance to the normal operations of their siloed business domain. When problems arise, it is essential to know what the problem is and where to look? Sometimes problems arise due to data input procedures.

Improved business data literacy would also be advantageous to making business users understand the reason why proactive support of the data team by provision of the business context and definitions of their data is beneficial to their goals and objectives.

The ultimate objective of the data should be communicated to the data team to ensure the output is aligned to the business strategy and goals.

Leaving the data team to provide answers without the proper stewardship of the data will provide outputs for sure however without context, it might be outputs which will be undesirable to the business and lead the business to make unfavorable decisions .

Is only one data owner required or owners in different domains?

Data is very powerful and therefore needs to be handled with care. Organisations rarely have one domain handling the entire operations of the organisation. Due to the various requirements of the various domains within an organisation, it is beneficial to have data owners for the respective domains. Owners of data need to be established across business domains in an organisation such as data stewards.

Nobody wants to lose their job especially in an established career where decades have been spent building the knowledge and skills required.

However, the context and definitions of data should not be left to the data and technical teams alone because it is the business who use the data to make decisions and inform strategic use. It is important that there is a handshake between business users and the Information Technology teams to understand data. Because data can be interpreted in different ways and can be stored in different ways, it is important for the business to provide their own specification for their data and how the data is to be used for their needs. To make the data fit for purpose, this cannot be done by technical teams alone. The various business domains therefore need to take ownership of the data. Otherwise the data is orphaned and can be claimed to be anything or nothing at all.

Are you having doubts about your data? Data literacy can be what you need. We cover data management capabilities which can help establish a forward looking data landscape for your organisation. Reach out today to discuss your requirements.