How to Create a Data Quality Policy
Introducing Data Quality Policies
What is a data quality policy and what are the key sections?
Who should create it and how is it executed?
There are many questions posed when creating a data quality policy for your organisation. Use this article as a guide to getting started and creating a data quality policy that is right for your organisation.
The problem with data quality policies
First, a warning. The aim of a data quality policy is not to tick some administrative box that constitutes a milestone on your data quality roadmap. A data quality policy should act as an actionable statement of intent that galvanises your organisation into further action.
A data quality policy is also never an inanimate object. It can (and must) change as your organisation matures and adapts to the increasing dependence on high quality information assets to drive your operations.
Wikipidia describes a policy as…
“...a principle or rule to guide decisions and achieve rational outcomes. The term is not normally used to denote what is actually done, this is normally referred to as either procedure or protocol”
This sums it up neatly, your data quality policy should not get into the detail. This must be specified by procedures and rules defined elsewhere.
Who should create your data quality policy?
It doesn’t really matter how the policy is created but what is important is that it is ultimately signed off by the most senior figure in your organisation (we’re talking CEO, Managing Director, Chairman etc.). You may also request sign off from your respective heads of information, security, IT etc.
A data quality policy must be sanctioned at the highest levels of the organisation because it will invariably be used to negotiate and expedite resolutions when conflicts about data arise. Your data quality policy must be the pillar against which all future data quality decisions are built upon.
What sections should be included in your data quality policy?
This is open to personal taste and corporate standards but a common format includes:
Purpose Statement: Why do we need the data quality policy? Why should people pay attention to the directives enclosed? Your purpose statement should be clear and direct, no essays please.
Background: Useful for explaining the history of data quality in the organisation. Why is it important now? How does it align with other policies and strategic objectives in the organisation? This section is important for providing context to the policy and making it clear who will benefit and why.
Scope: In what circumstances should the policy be enabled? For example, does the policy apply to all data in the organisation? What about 3rd party data? Having sections like scope forces the company to continuously update the data quality policy.
Roles and Responsibilities: Which groups (or individual roles) will be responsible for ensuring the policy is executed? Will the policy be governed centrally or federated out to lines of business or operational heads? What will be expected of each role?
Policy Statement: This is the nuts and bolts of the data quality policy and serves to guide future rule and decision making about data quality strategy and execution across the organisation. Considerations here may include questions like - How will specific situations and conflicts be dealt with?See the resources section below to give you ideas on what type of sub-sections to add to your data quality policy statement section.
Definitions: Don’t let jargon cloud the aim of the policy which is to enable clear communication and action. Include a definition section so that everyone can understand any acronyms or uncommon terms that only a data quality practitioner can rattle off.
Legislation: Which acts and compliance directive are the company bound by? It’s a good idea to include them here and provide further depth in the individual procedures and frameworks dictated by those acts but having executive oversight of exactly how critical data is from a legal standpoint is always beneficial.
Reference Documents: Which other policies and standards are linked to this policy? Were any other documents used to formulate this policy?
IMPORTANT TIP: Don’t store your data quality policy away in a drawer marked "Corporate Policies” or a file directory that no-one ever visits.
Why not post your data quality policy on the corporate Wiki so that the entire organisation has access. You could even go further and make the policy visible on your website as a statement of your commitment to the cause. By having the policy online it’s also a lot easier to link to various other procedures and resources that are impacted by the policy.
Remember that your data quality policy is not static. It must be constantly updated and expanded as your organisation matures its attitude to data quality.