Solving problems together

Knowledge Center
Thought Leadership

Learn more about the problems we are passionate about solving. Hopefully you find this content relevant and helpful. Please don't hesitate to reach out to learn more.

5 Principles for a Modern Data Strategy

Blog 5 Principles for a Modern Data Strategy

Taylor Culver

Taylor Culver

Apr 2020

Having worked with a handful of organizations trying to launch or improve upon their data strategies it’s evident that a pattern is emerging. Organizations have made tremendous investments in data but are having less than desired outcomes than originally intended. The reality is that many outcomes were solutions led, that the business has inherited over time, and has not been resourced appropriately to maintain or improve upon the original investment.

Having worked with a handful of organizations trying to launch or improve upon their data strategies it’s evident that a pattern is emerging. Organizations have made tremendous investments in data but are having less than desired outcomes than originally intended. The reality is that many outcomes were solutions led, that the business has inherited over time, and has not been resourced appropriately to maintain or improve upon the original investment. 

 

Furthermore, data teams are being spun up to fix the problem; but absent a clear objective and sponsorship they become an added layer of complexity. Today, now more than ever, it is critical that we revisit our data strategies more than ever to achieve the benefits of machine learning, predictive analytics, and robotic process automation. Business intelligence & reporting is not the stopping point, but rather the beginning.

Purposeful:

  1. Data is not a project, solution or team but rather an ongoing service to the business similar to operations, HR or finance.
  2. Doing nothing with data is an acceptable strategy if there is no business benefit to doing something.

Empowering:

  1. An individual should be held accountable to an organization’s data strategy
    1. This individual is entitled to executive sponsorship and adequate resourcing to do their job to the best service of the organization
      1. If this individual is not resourced appropriately, they are put in a position to fail at which point that individual should consider resigning from the position
      2. If this individual does not resign, they should be reassigned by their manager as data projects can lead to more distraction and negative impact if not properly staffed or managed

 

  1. Having a departmental data strategy is acceptable assuming the organization is willing to carry redundant costs across the organization.
    1. It is common for data to emerge out of a department as different departments carry greater sophistication in their analytical needs than others
    2. When two departments have data teams, a single individual should be put in place above both for oversight.
    3. In the long term, having parallel teams for data will create greater organizational complexity and productivity loss than the efficiency it affords each departmental need

Impactful:

  1. In turn this individual is accountable for demonstrating tangible business impact from managing data.
    1. Offsetting risk, new insights, being innovative, implementing a solution are not avenues for positive & tangible business impact
    2. The only way to demonstrate tangible business impact is through data initiatives that drive financial benefit to the organization.
    3. By eliminating data driven inefficiencies, productivity can be improved
    4. If productivity is improved and no action is taken and there is no economic benefit to the initiative
      1. If the cost of the initiative is greater than the productivity gains identified the project should be stopped until the benefits are better understood
        1. Time spent discussing what data means, manipulating data manually to produce content or improve data quality are all sources of lost productivity.
        2. By identifying redundant software spend or removing poorly adopted technology solutions relative to their costs, data teams can drive cost out of organizations
      2. Licensing data is a means to data monetization
        1. Licensing is difficult for many organizations to achieve and is a less probable outcome for immature data team

Engaging:

  1. Critical to achieving this business impact is driving engagement with the data community.
    1. The data community consists of any employee that uses data in their day to day whether it be accessing data from a tool, spreadsheet, or receiving information from another person to make organizational decisions.
    2. Data consumers are going to be engaged to various degrees from detractor to enthusiast
    3. Rather than try to change the personas you are working with, understand what they want and need to achieve their data related needs
  1. The data lead must source common business needs across data consumers to clearly defined use cases
    1. A use case must have a sponsor, a business impact, and a clear goal
    2. A use case is more compelling with a greater business impact, number of users with the lowest risk and complexity
  1. The data lead must work with the business to document data requirements at the metric level before solutioning for the problem
    1. Key business terms, definitions and owners must be established and mapped to a use case
  1. The data lead must map business terms to what is in physical systems, as opposed to mapping physical terms to business terms
    1. Giving the business all the tables and columns in the database and asking them to define them is a waste of time and extremely frustrating
  1. Identify solutions that solve the problem within the scope of the overall use case
    1. Use cases with common solutions can allocate the cost of those solutions across those use cases
    2. Buying a solution before identifying a use case is a recipe for failure
      1. This is a known fact, but it doesn’t stop people from making the same mistake – be courageous

Transformational:

  1. Effective Data Governance is the operationalization of the principles above
  2. Data Governance meetings become conversations about use cases and timeline for delivery, followed by driving consensus on metrics with overlapping use cases
  3. New business needs can be added into incremental use cases
  4. Omitting any of the above steps will lead to likely lead failure of a data initiative

 

XenoDATA™ is a boutique management consulting firm partnering with you on making your data strategy successful. We strive to understand your core data issues through a business lens, and partner with leading technology providers to achieve these goals both in an implementation and advisory capacity. Companies cannot afford investments in technologies that do not produce results. XenoDATA™ work to empower teams with strategies and technologies that drive financial results.

 

Our services include data strategy advisory, data product development, and analytics implementations