What are the most effective ways to develop and document data standards and policies?

Understanding your data needs and goals involves conducting a comprehensive data inventory and assessment. Identify data sources, types, formats, ownership, users, and quality issues. Align these insights with business objectives and strategies to define the scope, purpose, and value of data standards and policies. This foundational understanding ensures that data management efforts are aligned with organizational goals, optimizing data utilization and protection.

In order to understand your team's data needs and goals, you need to honestly assess your pain points with data storage, quality, formats, and analysis. Ask yourself the following questions: - Do we have proper version control? This is a must to track changes and maintain paper trails for reporting purposes. - Is our data storage system intuitive and easy to introduce to new team members? If one person leaves the team, will we know how to trace our data sources? - What are the most common pain points surrounding data management when prepping client deliverables, and how can we prevent those? - Can we automate any of our routine data management processes?

I always believe in the top-down approach from organization goals, distributed LoB goals, collective business objectives, and precise functional strategies to understanding your data needs. Once the approach is aligned with all the functional areas, the entire application and the underlying data have to be thoroughly analyzed to understand the system. Now comes the utilization (data analysis) and the valuation (business insights) as the desired outcomes.

Start at "why" to figure out the value data can bring to your business. You need to understand the data you have, but also what data you don't currently have but need to fulfill your business objectives or goals. How to go about acquiring that data and ingesting it into you data platform also needs to be part of you policies.

Developing and documenting effective data standards and policies is crucial for maintaining consistency, integrity, and security within an organization's data landscape. First and foremost, it's essential to involve key stakeholders from various departments to ensure a comprehensive understanding of organizational needs and compliance requirements. Collaborative discussions and workshops can facilitate the identification of critical data elements and the establishment of standard formats.

I agree.. In addition, there should be t he involvement of an ethico-legal expert who should be closely aligned with this working group to identify any ethical and legal implications of the data that will be shared or that will have to be preserved.

Creating a data governance team is like assembling a dream team for your organization's data health. In my book, effectiveness comes when this team is cross-functional—imagine having IT, Legal, and Operations all in one room, hashing out data standards. Regular, maybe monthly, check-ins among this group can keep everyone aligned and accountable. By clearly defining roles and updating them as needed, you avoid the "that's not my job" syndrome. It's all about teamwork to keep your data shipshape and audit-ready.

Regardless of the operating model (centralized, distributed, etc), ensuring that all areas within an organization have a clear understanding of roles and responsibilities within the Data space sets a critical foundation for effective discussion about data standards and policies.

Facilitating collaboration among data experts involved in the process can create investment and buy-in to shape and use these standards and policies effectively. Adherence is strengthened when experts are able to understand more clearly how their work affects others’. Strong relationships can enhance processes and reveal efficiencies unavailable if colleagues are siloed in their function. Beyond the required data governance oversight, the system benefits when participants have a more complete understanding of their contributions.

Roles and responsibilities are essential in any good governance. Google "roles and responsibilities template", take your pick, and get started. Your governance should be dynamic based on how your business evolves, and remember that the people you would like to assign roles and responsibilities to, might not understand exactly why, so you need to apply your great change management skills, so everybody understands what is in it for them and for the company.

Adopting and adapting industry standards is imperative. There should also be cognizance of the fact that multiple standards may exist in the organization. The challenge lies in arriving at a consensus among multiple stakeholders and legacy integration.

When it comes to metadata management, start small and build from there. In the initial phase, capture only the essential information that will make your data discoverable, easier to understand and therefore promote its effective use by prospective end-users

Again this cannot be generalised as every data domain has to be treated on a case by case basis. The best example here would be the microbiome data, where its extremely challenging to define standards and policies as the field is a constant moving target.

Following industry best practices has far reaching benefits, some of which are very real but which can be difficult to define quanitatively. It can help ease and speed up onboarding for new employees in the data space and can also help improve the ability to effectively communicate with vendors and collaborate with others in the industry.

Don't consider frameworks and best practices as restrictive molds but as tried-and-tested recipes. Just like in cooking, while the recipe offers a baseline, there's always room to innovate and tailor to your unique organizational taste. But remember, the key is to start with a solid foundation.

Here, its highly important to define the data with standardised ontologies and terminologies for provenance and make data sharing easy for data submitters/users.

Without effective communication of data standards and policies, the purpose of developing them fails and also leads to possibilities of more standards being developed.

Part of effective communication is striking the balance between reinforcing key messages while not overwhelming audiences. Repetition and consistency is key to "getting the message to stick". But it is important to monitor audience engagement and feedback. If an audience appears to be overwhelmed, it may be time to pause or slow down the flow of new information to allow them to digest what you've already shared.

Adopting proven data management frameworks is like getting a tried-and-true recipe for success. Why experiment when you have a roadmap? I'm a big fan of DAMA-DMBOK for its comprehensive approach to data governance. It gives you a strong backbone for your policies and ensures you're covering all the bases. Plugging these frameworks into your data governance team's agenda can make their monthly meetings way more productive. It's a strategic move for making sure your data practices stand up to any audit, while also being scalable and effective.

Defining and documenting the data standards and policies is essential. However, the crucial aspect lies in ensuring their adherence. Implementing mechanisms to monitor and enforce compliance with data standards and policies, including audits and automated checks is imperative.

Most companies can achieve this by conducting an annual review of their standard policies and procedures. Training and awareness is also key to ensure employees read the actual policies, at seem companies I have seen this gamified, whereas the company make its fun and spends a whole day doing activities around awareness, case studies, games with examples and prizes. This garners more engagement and ensures employees walk away with some key take aways from the day.

Policies and standards are a useless waste of time if they are not actively used. They need to be developed collaboratively and woven into the existing governance fabric of the organisation. For example: 1) make sure your Internal Audit Dept are on board from the beginning and reference the data policies in their reviews and monitoring processes, 2) ensure they are integrated with IT Security policies and processes

Implementing data standards and policies is akin to planting a tree. Monitoring and nurturing it ensures it grows strong and in the right direction. Regular check-ins, feedback loops, and adjustments keep the tree healthy, ensuring it provides shade (or in this case, actionable insights) for years to come.

Automated data checks in In Clinical Trial. Case Report Forms (CRF) are essential for maintaining data integrity and quality in clinical trials.. Suppose a clinical trial is investigating a new medication for adults aged 18 and above, A data validation rule can be set up to automatically check the entered age. 1. Range Check: The system can be programmed to only accept entries within a predefined range, for example, ages 18 and above. If a value outside this range is entered, the system could flag this as an error immediately. 2. Consistency Check: If there is another field where the date of birth is entered, the system can automatically calculate the age and cross-verify it with the entered age to ensure consistency. .

Setting the expectation up front with all stakeholders that data policies and standards are not static can help build trust and facilitate discussion. By expressly stating that you are open to feedback and suggestions, you can set a collaborative tone with your audience rather than risk them viewing the policies and standards as something being done to them.

Create SOPs and make it accessible. Make SOPs as a mandate to follow and keep SOPs upgraded from time to time out of past learnings.

Complacency is a silent killer in the realm of data governance. Even if your data standards seem perfect now, the digital landscape is ever-evolving. Regular evaluations aren't just about finding flaws—they're about uncovering new opportunities and staying ahead of the curve.

Develop and document are two very different things. When developing, I think it’s important to identify the right level of depth of standards and/or compliance. Overcommitting simply calls for violation. In most circumstances, data is a mean to a goal, so need to balance the effort between governance and use. Also need to cope with the fact the things keep changing…. Documenting is a very different thing. In an ideal world, we would write executable specs that can be directly embedded in processes. In a practical world, we write guidelines… maybe some attention to definitions would already help a lot.

In my experience, developing and documenting standards and policies should NOT be an exercise owned by IT departments or IT functions. It should be driven by the business side of the organisation. Simply because they are the ones who generate, know, understand and use data ultimately. IT functions should be part of the overall effort but the business must call the shots at all levels. On the other hand, the development, adoption, documentation and refinement of standards and policies should be iterative and aiming at small steps that bring measurable value and progress. Large organisations can typically afford such kind of exercises, in contrast to small ones, which have other priorities and need to have strong data strategy trade-offs.

In my time auditing various organizations, I've seen data governance projects really take off when there's strong leadership buy-in. One key lesson I've learned is the importance of ongoing communication across all departments, not just IT. Having a regular cadence of updates and check-ins fosters a culture of accountability and continuous improvement. It turns what could be a one-time project into a living, evolving practice that better prepares you for audits and enhances data quality.

Data policies enforce rigorous standards for data collection, processing, and analysis, ensuring the data’s reliability and accuracy. This is pivotal in clinical trials, where the quality of data directly impacts the validity of the trial outcomes.