Digital Twin Webinar: Understanding the Retail Analytics Glow-Up
Register Here →The Enterprise Leader's Quick-start Guide to
What every leader should be thinking about right now in order to get the most out of their data and make more well-informed decisions
A report published in August 2019 found that more than 70 percent of organizational leaders—from Chief Data Officers (CDOs) to Data Analysts—said that suboptimal data quality has negatively impacted their business decisions.
This resource is intended to get to the crux of how to improve the quality of your data sets so that you can learn to trust your data and make more well-informed decisions; it’s a “quick-start guide” because it outlines the most important activities every enterprise leader should be prioritizing right now in order to reduce risk, improve data efficacy, and make better decisions.
How Treating Data Quality as a Discipline Improves Decision Making:
Before your organization puts any data quality plan into action, it’s crucial to have the support of C-level leaders who understand how data fits into key organizational goals. While individual teams recognize their own siloed needs for data efficacy, top-level executives have the broadest perspective on how data quality, or lack thereof, impacts overall business performance. This big-picture view should be the driving force behind data quality efforts, guiding the necessary people and process transformation.
Keep in mind, improving data quality is not a quick fix or one-time undertaking. It’s a long-term investment of people and resources that demands the support of leaders at every level. As organizations overall become more data-driven, there has been an increase in C-suite involvement in data quality. Yet, buy-in can still be difficult to secure and often requires further education to overcome executives’ lack of experience with data and analytics. In the end, the C-suite should lead the charge toward data literacy for the entire organization.
Appointing a dedicated data quality team is the most effective approach organizations can take to improve overall data quality. Yet, 70% of organizations say they do not have a dedicated data quality team, according to O’Reilly’s State of Data Quality 2020 report. When it comes down to it, improving the reliability and efficacy of data should be a team effort spearheaded by experts who appreciate the need for standardized data processes.
Your team of data experts should take ownership of all elements of your data strategy and oversee adherence to organizational standards, such as data literacy, governance, lineage, and conditioning. Interestingly enough, organizations with dedicated data quality teams use technology like AI and machine learning to manage and condition data at a much higher rate: 59% of respondents with dedicated data quality teams versus 42% or organizations without dedicated teams, according to the same O’Reilly report. Though these technologies do not ensure data quality on their own (AI/ML is only as useful as the data fed into it), they introduce more efficiency and accuracy into the data conditioning process.
Many organizations appoint a Chief Data Officer to lead the charge and garner executive support from other high-level leaders, as delegating these responsibilities to a new team with this single focus may require a significant investment of resources. Look for opportunities to leverage existing teams and talent in order to round out your data quality team’s expertise and save money on organizational investment.
How Focusing On Data Literacy Improves Decision Making:
Treating data as a discipline means, among other things, committing to the ongoing education necessary to ensure base-level literacy for all teams and individuals.
Still, most teams today lack even the most basic foundational competencies. In fact, Gartner reports that half of all organizations today lack sufficient AI and data literacy skills to achieve business value, while 80% will initiate deliberate competency development in the field of data literacy to overcome “extreme deficiencies.” Thus, it’s critical that leaders take the steps necessary to quickly reduce such a critical skills gap; the first of which is gauging your team’s current ability to “speak data."
Data literacy is now an essential, fundamental skill that all individuals must learn. Yet, in its most recent Chief Data Officer Survey, Gartner also reports that ‘poor data literacy’ is the second-biggest roadblock standing in the way of creating a more data-driven organizational culture. Poor data literacy can undermine your efforts to improve the quality and usefulness of your data; it’s the difference between organizations who are able to consistently and accurately make data-driven decisions and those who struggle to find meaningful insights in their data.
Thus, at a minimum, you first need to establish a benchmark for data literacy within your own team or organization. There are a number of tools you can use to quickly assess data literacy among your team (some will be better than others). But, if you're looking to move quickly, this data literacy assessment created by the Data Literacy Project is a good place to start, especially if your goal is simply to assess individual and team literacy while creating conversation in rapid fashion. Doing so will give you a starting point to better understand where critical gaps exist. Oftentimes, it’s helpful to assume that you’re starting from the ground up, especially because you’ll likely need to create a common, consistent data vocabulary that’s then easily understood by the rest of the organization.
Developing your own team’s data literacy is a starting point, but in order to make a lasting impact, aim to champion data literacy across your organization or business unit as a whole.
It’s the responsibility of everyone in your business to do their part in working toward data proficiency and to raise concerns or issues along the way. Getting to this point, however, requires a strong data-minded leader, a role you may need to assume yourself if no one else in the organization has assumed the responsibility. This role comes with its own challenges, but the impact on your team and organization can be felt almost immediately.
Even taking incremental steps toward data literacy can improve data-driven decision making for individuals and teams, not to mention the increased confidence and comfortability in working with data that comes with their enhanced understanding. Data literate teams typically have a better sense of how the business operates on a high level, which also impacts the ability to make well-informed decisions. If your organization doesn’t have a literacy plan in place, find out why. If there is a plan in place, aim to uncover how you can make actionable improvements. Establish your baseline for literacy and a hierarchy for how information about data processes and standards will be disseminated throughout the organization.
How Standardizing Data Governance Practices Improves Decision Making
Designating a team to carry foundational data governance methods forward does not necessarily require that you make a substantial investment in both new personnel and tooling right off the bat. Instead, it can often be just as effective—especially toward the beginning of such an undertaking—to redirect existing personnel or resources in order to establish critical momentum early on. While it isn’t a perfect long-term solution, it’s typically one that’s more palatable and actionable for those who may already be limited by resource-constraints or haven’t yet made a full commitment to data as a key discipline (though the latter is becoming increasingly less common in 2020).
In fact, Gartner has reported that nearly 75% of large organizations will have appointed a Chief Data Officer by 2023. So, if there isn’t a CDO in place at your organization already, the process to fill that role has likely already begun in earnest. Typically, the job of appointing a cross-functional team who will champion data governance falls on this individual. Thus, there may already be such a team in place. If not, however, this critical task usually falls to another leader who has both recognized the importance of such an initiative and actively taken steps to make it a reality.
Identifying such a team is critical for both short and long-term success (even if that team is only temporary, perhaps to be replaced as a CDO has had time to implement his or her own strategic plan). In short, this team is typically responsible for securing resources (budget, especially), setting governance goals and objectives, designing the data governance model, selecting the appropriate tools to manage the governance practice, and, in many ways, they’re the internal “face” of data governance internally. Simply put, the members of this group are the stewards who’s primary task is to maintain the integrity of all data inside of the organization.
It’s wise to establish a “charter” for the team, or, at the very least, to crystalize a set of goals that will help direct the initial governance efforts.
The aim isn’t to outline all the things the team could do. Instead, the objective is to identify the most important mission for the team. You’ll want to make key decisions that help set the scope of what the team will tackle, including potential blockers, dependencies, and milestones. A decent place to start is to determine where the team should ultimately focus the majority of its effort out of the gate.
For instance, is the team’s primary responsibility to catalog all existing data and associated elements? Is the goal to manage all data elements necessary to achieve a specific business outcome? Or, for example, is the primary aim of the team to manage the key data elements that power the organization’s top-level objectives and strategic priorities?
Remember that the goal here is to identify a starting point, then give that team (and any others involved) the autonomy to evolve their charter as data governance builds momentum within your organization. Make sure that the team and other key stakeholders are on the same page to ensure continuous improvement and long-term commitment to your governance strategy.
This step is necessary both at an organizational and functional level. In many cases, this step is either lacking entirely, or it’s been done rather haphazardly.
Thus, it’s worth spending the time and energy necessary to map existing data flows within your specific function or team. When doing so, identify any “dead ends” or silos you encounter along the way. Even if there is a governance team who’s working on a similar initiative, it’s still worth a second look. You want to intimately understand where critical data is housed, how it’s collected, and where it moves over time (both upstream and downstream).
There are a variety of tools that can assist you in this process, many of which will allow you to simplify and automate some of the critical tasks involved in discovering, profiling, and indexing data. These tools can also be used to identify and resolve any existing data quality issues. In fact, if your organization already has a dedicated data governance team (or a group or business function acting in a similar capacity), it’s likely that an enterprise data intelligence platform is already in place. If that’s indeed the case, check with that team to better understand how the tool and technology stack can be used to assist in this process. Any sound data governance practice will be actively using these kinds of tools at a global level.
At a minimum, there are typically three, critical roles specifically designated within any cross-functional data governance team:
1. Data Owners
2. Data Stewards
3. Data Users
Each plays a specific role in both enforcing key governance practices and improving data quality. Owners and stewards typically work closely together to ensure that data quality does not degrade as it moves from system to system, while users not only lead by example, but identify inconsistencies in usage, policy, and procedure when appropriate.
At present, only 20% of organizations say that they publish data provenance and data lineage information internally, and most of those who don’t publish such information say they have no plans to start.
Don’t make this mistake. Making this type of information readily accessible to all teams is one of the most important steps toward creating a more data-driven culture, which, if maintained and governed properly, are critical to improving data quality directly at the source.
How On-Going Data Conditioning Improves Decision Making
Data “conditioning” simply refers to the on-going process of organizing, structuring, and cleansing data so that it can be consumed efficiently and securely by the business.
Clearly, maintaining the accuracy and quality of your data will require on-going effort, energy, and investment. Where you choose to prioritize that effort and energy is highly dependent on your own strategic objectives (or those of the company, as a whole).
Still, there are typically two, common challenges that typically require consistent, dedicated attention: managing multiple data sources (or reducing the number altogether) and simply making sense of the sheer volume of available data. For instance, 70% of organizations say that their top data challenge is simply the ongoing management of multiple data sources, while almost 50% say that the sheer volume of available data is a tremendous efficiency killer and burden in and of itself. These are both critical problem areas that plague most teams and organizations and should be prioritized as such.
The requirements gathering process here is not unlike that of other initiatives, though it’s important to note that it will take coordinated effort not just from the data governance team, but other organizational leaders as well. So, time is of the essence, otherwise there’s a risk that resources and attention may be unknowingly diverted to other, higher-priority projects. Thus, you should have a solid grasp of which data initiatives are most deserving of immediate, continued investment. Doing so will allow you to make decisions more quickly (if you are the ultimate decision maker, which is sometimes the case). Still, doing the due diligence will prepare you to make the best decision as to where to invest additional resources.
Your approach to allocating and prioritizing resources may vary considerably depending on a variety of factors, namely time, personnel, and budget.
This is the point, however, at which the rubber meets the road. You may be tempted to delay this part of the process, but you typically will have more to gain from targeted experimentation than you do from waiting for permission. Prioritizing effectively (with an ability to secure the needed resources) requires someone who can see the bigger picture, identify the critical priorities, get others to recognize those priorities, and make progress toward identifying what will be needed to maintain ongoing data quality.
Even still, when it comes down to brass tax, be sure to factor in lost opportunity cost if you’ll be dependent upon personnel that are juggling multiple, strategic priorities (which is often the case). Ongoing data conditioning may initially stretch current staff in some parts of your business more than others, so plan for an uptick in time and support for the teams and individuals who will be carrying out such initiatives.
Ideally, this type of prioritization becomes a regular part of your data management practice (if it isn’t already). Resource availability, budget, and needs are likely to evolve over time, exposing new gaps along the way. These kinds of assessments, then, should be performed on a rolling basis, should have clear owners, and should be specific enough to drive recommended action to close said gaps, especially where additional dollars will be needed in order to fund continuing data management initiatives.
Regularly review existing data conditioning processes and their resulting outcomes with other leaders, particularly the appointed data governance team.
Identify both the critical aspects that produced the intended outcomes as well as the roadblocks that may be limiting your success or degrading your data quality. This sort of on-going data “health check” will better allow you to monitor data quality, both from a quantitative (data quality scores) and qualitative perspective (gathering relational information to better identify where bottlenecks exist). Your goal should be to monitor what you’re observing, compare against what you had expected, then use that information as inputs to determine the efficacy of your processes, policies, practices, and quality standards.
Over time, the expected result is that you, your team, and your organization will become much more aware of any kind of real-time degradations in overall data quality.
Poor data quality isn’t something to ignore; it must be identified, corrected, and improved. Many organizations today—particularly in 2020—have admitted to making poor or ineffective decisions because of poor data quality. Nearly 70% of enterprise leaders have said that suboptimal data quality has had a negative impact on their decision-making processes.
Improving data quality starts at the top, requiring strong, C-level support that then permeates through the organization. No matter your role, however, you’re a critical cog in the data machine. You have a responsibility to play your part.
Consistently making better, more well-informed data decisions requires (in most cases) a near systemic transformation toward more modern data management, governance, and conditioning methods. Improving data quality is a task with no end; it requires regular maintenance and upkeep in order to mitigate risk exposure, maintain security and regulatory compliance, and ultimately, turn your data into a core, transformative asset.
