Hi everyone, thank you for joining us for today’s webinar “lead your data revolution: how to build a foundation of trust and data governance.” My name is Erin Haselkorn and I am the head of our market research here at Experian’s data quality business. On the line with me, I have my colleague Kevin McCarthy. Kevin is a on our product marketing team, is a frequent blogger on DataVersity and has a lot of great insight into the space.
Thanks for joining me today Kevin!
A couple times a year, Experian does research to look at the latest data management trends. For our most recent study, we wanted to take a further look at how successful companies have been at actually becoming data-driven. Unfortunately, we found that many companies are still struggling to make this a reality.
Data is confusing. It is dirty, complicated, and spread out all over the place. We see a lot of data management initiatives happening to try and fix these issues, wrangle data into place, but we still aren’t seeing wide-spread success. However, we are starting to see a shift in the form of data enablement, which we will talk more about as we get into the webinar. If companies can empower a wider data usage then they are better able to achieve some of the insight they are looking for.
With that in mind, here is what we are going to cover off today. Because much of today’s presentation is based off research, I am going to start off with a brief run-down of the survey methodology followed by some of the key findings that will frame our discussion.
Next we will dive a bit deeper, talking about the challenges we face in becoming data-driven, the trends we are seeing from the data, especially the rise of data enablement and the wrap-up with what we see as the profile of a mature organization. At the end of today’s webinar, we will have time to answer some questions. Feel free to ask those throughout and we’ll try to get to as many as we can at the end.
First, let’s talk about the background of the data we are going to discuss today. In July of this year, Experian worked with Insights Avenue to conduct a study on how organizations are enabling the use of data across the business. We spoke to over 500 people in the US who had knowledge or visibility into their company’s data management practices.
Given that background, we spoke to a number of folks across different departments and with varying levels of seniority. However, as you can see, about a quarter of the respondents were actually from the c-level. These individuals also came from companies representing a wide range of industries, such as financial services, public sector, retail, healthcare, manufacturing, utilities, etc.
As always, we like to keep our population set fairly diverse because as we know, the individuals working with and benefiting from data insights is also diverse.
Now with a bit of background, let’s cover off some of the high-level themes that stood out from the data.
We see a lot of companies working hard to use data. That is a very positive sign. Business leaders understand that without strong and accurate insight, they wont be able to engage with customers in the digital economy, or make efficient decisions, or keep up with the competition. However, we don’t see businesses succeeding in this effort. 69% say that despite many ongoing data initiatives, their organization still struggles to be data-driven. It also takes too long to get actionable insight for over half of businesses.
There are some clear reasons we see this happening, besides the fact that data is just challenging to begin with. It is our belief that business leaders often underestimate the level of data debt within their company and do not even realize it is dragging down the benefit of new technologies. They also have traditionally underinvested in the problem.
That shows up in some of these stats: 64% say they do not have enough talented data professionals, 66% say those improving data quality often do not fully understand the needs of the business. In addition, a lot of the data management initiatives that do occur are happening within individual departments. Now while there is nothing wrong with experimentation and frankly some of the individual department needs will certainly vary from the business, it is important that we approach data management with some degree of scale so we can have some common framework to work from as we continue to improve. Especially when we try to get data into the hands of more people across the business.
But we do see the tide turning. We are personally very excited for a new trend in data enablement. Data enablement is the practice of empowering individuals in a business with the support and tools they need to responsibly leverage trusted data to achieve real business outcomes. This is a great initiative, but we see 57% of respondents saying it is a priority over the next 12 months.
With that in mind, here are some of the major take-aways from this data.
First, while most companies have a lot of big data projects going on, most are struggling to be data driven.
Next, there are a variety of different approaches to data management, from departmental to full enterprise and project-based vs ongoing disciplines. However, most organizations are missing a key cornerstone to data management success, which is a foundation built on data quality and trust.
Finally, if we are going to improve data usage, stakeholders need to enable the business through the right data talent, the right technology, and also promote organizational changes in the culture related to data.
We are going to get into each one of these areas in more detail.
For now, I am going to turn things over to Kevin to have him walk us through the types of data investments we see companies making.
As Erin mentioned, we see a lot of investment happening in data. However, companies aren’t necessarily seeing the results that they set out to achieve.
And when we mean a lot of projects, we mean a lot. 80% of the folks we talked to are actively pursing multiple big data projects. The top ones include areas like data quality, big data analytics, and data governance.
These projects sit in a wide variety of areas across the business. The bulk still sit with IT, which isn’t surprising, this is where data has traditionally been managed. However, we see 42% sitting directly with business users. This probably means they aren’t getting what they need from the existing data management practices and are trying to take matters into their own hands.
But again, even with all these projects, we aren’t seeing the desired results.
Let’s dive deeper into the projects folks have going on. You can see the dark blue are projects currently happening, purple are projects they plan to start in the next 12 months, light blue are projects on the radar, but they don’t have a fixed timeline. Finally the pink color is a project that is not planned yet.
The top three projects are data quality, big data analytics and data governance. However, we also see areas coming up like data literacy, which is becoming more popular, and of course ML and AI.
But let’s first take a look at data quality.
Data quality has been around for a long time, but companies have also struggled with building trusted data for longer. This is clearly a foundational discipline when it comes to an area of investment.
Part of the problem with data quality is it isn’t necessarily the most exciting way to think about your data. It is traditionally something that companies have underinvested in and an area that I personally know, we still need a lot of work around. It is exciting to see that a lot of people think they have made progress when it comes to data quality. The pie chart on the screen is the response when we asked people if they felt they had made progress in improving data quality in the last 12 months. 93% said yes definitely or yes possibly, which is great.
However, we still see a lot of roadblocks in this area, which Erin will discuss shortly.
So we are investing in data quality, we are also investing in big data analytics.
Analytics is one of those more exciting aspects of data management. We all want analytics, especially when we can get insight across all the massive piles of data we have been accumulating.
79% say they are focused on analytics and how to gain more insight with data. That is great, but getting insight from this data is hard. 88% report challenges.
We see some of the common challenges in this bar chart. A lot of folks may not have access to the information they need. Now that could because of various regulations, but probably, more likely, the data is just set up in a way where it is complicated and maybe the systems that are in place for analytics aren’t built for the data structures that we have today.
There is often also too much data to analyze. The volume of data issue certainly isn’t going away. But people are also struggling with the different data types they are going to analyze. Are we talking about structured data, or maybe social media data that isn’t as neatly stored. Those things all create issues.
It also takes too long to prepare the data. I was able to review a report recently from Stewart Bond at IDC and he was showing from his new research the large amount of time data professional spend just preparing the data. It is a huge amount of time and goes back to needing some of those foundational elements of governance and quality. The last one I will talk about is not having the right technology. Now, from my experience, I know there are a lot of different technologies out there. Some are very good and some are designed specifically for different roles. A lot of the technology that is available today is designed for more technical users. What I mean by that is that it may require coding or some level of sophistication to be able to use it. The challenge is that now everyone wants data, from marketing and finance to the sales team. Tools need to be procured that can allow more people to access data, not fewer. Especially when you think about gaining insight for analytics. The last data management project we will touch on is data governance.
Many companies are investing in data governance and different organizations are at different levels on this journey. This is similar to what we see across other data management disciplines.
We see 29% that are relatively mature, taking a holistic approach that involves a governance board, technology and data governance related roles. Then we see 27% who are just starting on a program in the next 12 months. 26% say the process depends on individual departments and then we have 16% who have a governance board, but have not decided how they want to move forward with technology or enforcing policies.
This isn’t surprising, again, it is like a lot of other data management disciplines. But when it comes to governance, we see some clear directives as to why people want to invest heavily in these programs.
Companies want to understand how data is used. They also want to comply with regulations. Now this is what governance was traditionally thought of for, compliance. However, in recent years, data governance has started to encompassed a much broader definition that takes into account areas like data privacy, data usage, data context, etc. However, compliance is still a central component.
We also see that companies want to improve the quality of data for decision-making. They want to be more data-driven, all great reasons to invest in governance.
While there is reason to invest, there are also a lot of challenges to be aware of. It is hard to get people to follow data rules, as we are all aware of. They don’t have the right technology or enough knowledgeable resources. There may also be a lack of executive buy-in. While these are all challenges that can cause problems when implementing a governance program, they can be overcome with the right approach. A governance program is certainly worth doing to get the benefits of accurate and trusted data.
Thanks Kevin. As Kevin said, we are making a lot of good investment in data management programs, but we still aren’t achieving the results that we want to see. And part of that is because of the way we are approaching certain aspects of data management.
In some ways, we are stuck in a bit of a data rut. Again we have a lot we are investing in, but we still struggle to be data-driven. We also still have a lot of inaccurate data that is undermining initiatives. Not just data management specific initiatives, but everything from customer insight and customer experience, to operational efficiency, compliance. Think about all the ways you use data now and when it is inaccurate, it can cause a lot of problems for the business. We also only see 29% of companies taking a holistic approach to data governance, which is indicative of a lot of data management efforts.
We are stuck in this rut because many of us are not approaching data management initiatives in the right way. Sometimes we get carried away and excited about the latest thing that we forget about the foundational elements that take a lot of time and effort to get right.
Right now we see different approaches to data management. There is certainly no one-size-fits-all approach to managing data assets. And each initiative should not and cannot be set up the same way for each company.
However, our research shows certain organizations are having more success than others when it comes to data management and becoming data-driven. There are four common areas that we looked at:
- The scope of the initiatives, if they were departmental or enterprise
- The maturity level of data quality
- The approach to the management processes, be it one-off or an ongoing discipline
- Finally, we looked at data debt
But lets tackle each of these variations, starting with the scope of the initiative.
Must of the investment in data management happens within individual departments or pockets of the business. Again 69% are reporting data management initiatives occur in individual departments and only a few on an enterprise level.
You can see on the screen the departments that are most advanced when it comes to data management. These are logistics and operations, finance, marketing segmentation, senior management decision-making, sales, etc.
Now this siloed approach happens across all data management disciplines. While different department have different needs for leveraging data, there is a risk that companies may not leverage best practices around the various departments or across the business in general. This can lead to a lack of insight and certainly a lack of trusted data.
Now let’s move on to data quality.
You may be thinking, well this is just a general data management initiative. However, it really is a foundational discipline that a lot of others are built on. Without quality and trusted data, how can you expect to drive analytics? If you don’t understand the quality of your information, how can you build a data governance program? Without quality data, what are you going to use to train your new machine learning or AI algorithms?
We have found that the more mature an organization is in terms of data quality, the more mature they typically are in other areas of data management. However, we see that most people have typically underinvested in this area. And despite a general sense that companies have improved data quality in the last year, we only see 11% of businesses saying they are mature when it comes to their data management.
When looking at maturity, we looked at three areas: data talent, technology, and the processes in place around data quality.
You can also see from this chart that over half of companies are still on the bottom two levels of maturity, limited and emerging. That means companies still have a long way to go when it comes to data quality.
Next we have the approach to data management. There are two different types of approaches we typically see. Some look at projects as a technical project, and they will scope out that initiative, buy a piece of technology and then solve the issue in a set amount of time. These organizations will do an initial clean-up, and then the information should be all set. There is a technical scope and budget allocated for a specific period of time.
Other companies take a different approach and look at data management as an ongoing business initiative that ties to specific outcomes. For example, a company may look at data quality as an ongoing initiative, looking at how to prevent bad data from entering the system, and cleansing data over time. This may be used for customer experience and revenue generation, but it never actually stops.
We see half of companies taking one approach and the other half taking the other.
Finally we have the concept of data debt.
Gartner defines data debt as: The accumulated cost that is associated with the suboptimal governance of data assets in an enterprise. All (non-theoretical) organizations are suboptimal in data and analytics governance. Since no enterprise can operate at 100% effectiveness in data and analytics governance, time, effort and cost are associated with meeting the deficit between the ideal condition of data that is required for business needs and what is actually available.
Data debt is a lot like technical debt. If you are operating with a large degree of inaccurate or poor data, it doesn’t matter how much you invest in projects like machine learning, analytics or AI. Without that foundation of good quality data, these initiatives will not help the organization to the same degree that the business was hoping. Therefore, they wont’ achieve the same positive outcomes.
Organizations need to understand the quality of their data so that they can tackle other data type initiatives with confidence.
Now I will turn things over to Kevin to get us started in talking about data enablement.
Thanks Erin. With all of that backdrop on the types of investment people are making and the way they are approaching these projects, let’s dive into data enablement, which I personally think is a very cool new topic in the world of data management.
We aren’t the wide-reaching success we want to see in leveraging data, so organizations are starting to adopt new approaches. Data isn’t just for IT anymore. Ideally, it moves throughout the business. Organizations are turning to data enablement to empower more people to leverage data insights.
To level-set, let’s start with a definition. We defined data enablement in our survey as the practice of empowering individuals in a business with the support and tools they need to responsibly leverage trusted data to achieve real business outcomes.
While this is a relatively new concept in terms of terminology, we see a lot of businesses starting to gravitate towards it. Data enablement is a key focus over the next 12 months for 57% of businesses and a further 41% report that it is a focus to come extent.
However, we also still see challenges related to those projects. Then we also looked at some ways that people can enable the better use of data, looking at data talent, the right technology, and also creating the necessary cultural.
Why do companies want to think about data enablement? Well here are the outcomes we saw from our survey from improved data usage.
Not surprisingly, we see at the top of the list complying with regulations. With new regulations around data privacy and industry-specific regulations, compliance is top of mind for many companies and better data usage can certainly help with identifying issues and risk, as well as with compliance reporting. But compliance is more of a defensive play for data management. There are also a lot of ‘offensive’ outcomes that help achieve business growth that companies are also seeing benefit from better data usage.
If you look at the next few, we have enabling better decision-making, improving the customer experience, and better understanding the customer. All of these areas specifically enhance the bottom line.
Here are some of the common ways companies are looking to improve that data usage.
First, they are looking to provide standardized data across the business. Then we see things like, better utilizing data governance to ensure the right data usage, putting data professionals in specific departments, and consolidating certain sources of information. For those that have undertaken these initiatives, more than 95% report these actions have resulted in better business outcomes.
But there are also a lot of challenges companies face when looking at these practices. 89% of the individuals we spoke with reported having challenges when enabling the use of data. They often lack skilled human resources or data professionals. The communication can be challenging between siloed departments, they may not have the funding they need, or they may lack the data literacy within their staff.
All of these are serious challenges that need to be addressed. Companies cannot frame their data enablement program purely around the data, they need to think about the talent, the technology and the culture within the organization.
I am going to turn things over to Erin to go into each of these areas in a bit more detail.
Thanks Kevin. Let’s start with data talent. With any new initiative, there is often a rush to hire. Data professionals right now are in very high demand, so a lot of people are competing for data talent. That might be part of the reason why 64% say they do not have enough data professionals in their organization. This also relates to the top challenge of enabling data, which Kevin just mentioned. There is a big lack of skilled human resources.
Without the right talent, it can be very difficult to move initiatives forward in the right way. However, remember with data enablement it is about empowering the masses. That means you need to focus on a core group of data professionals, but then you need to think about how you can empower other folks with technology or processes or other resources to leverage data more effectively.
For that core group of data professionals, the bar chart shows some of the common roles being hired. You have your data analyst, data engineer, CDO, data governance manager, etc. What you need all depends on your focus areas and where you need more strength within the business.
I will say that Forrester recently came out with some very interesting data on the CDO. They see that 58% of companies say they now have a CDO, which is certainly a larger number than they have reported on in the past. And these CDOs don’t just exist within big enterprises, they are also showing up heavily across the SMB space so really any size company. I think while the CDO is third on the list here, they often form the cornerstone of the data professional group within a business so certainly one to watch here.
Next we have the technology. As part of data enablement, you want to expand the use of technology to ensure it is business-friendly so more people can leverage tools for insight.
The problem is that most traditional data management technology is built for IT professionals. However, more vendors are starting to produce software specifically designed for non-technical users.
On the screen you can see some of the tech that is being used for data enablement. The tools most likely to be in place are data preparation, excel and data quality. While some in the IT space will roll their eyes at Excel, many business users haven’t been given easy-to-use technology to replace this software. With Excel, they generally know how to use it, so they are going to leverage it.
But we do see some concerns with the existing tools and tech. 87% of companies report concerns with the tools and technology around data enablement. The biggest concerns are that they don’t have enough training on the tools, there are too many different technologies, and they don’t have the right people in place. Again, shows the need to look at this from multiple angles.
Finally we have the cultural shift that needs to take place. An organization has to be properly aligned to achieve data insights from a broader group of stakeholders.
When we look at the biggest challenges that Kevin went over earlier, and there are some in there that specifically point to culture, like communication between departments and data literacy.
A lot of data management initiatives are happening in siloes. That means as these initiatives move forward, there is very little coordination. While experimentation is certainly a good thing, there can be a lack of efficiency and scale if there isn’t communication.
Then we have data literacy. This is something again, relatively new to market. Data literacy is defined as the ability to read, work with, analyze, and argue with data. The goal is to make more people within the business data literate to empower insights. We see from the study 45% of organizations are working on this right now. Now data literacy is something we see a lot of different people working on. Sometimes the CDO takes the lead, but it could be human resources or teams that are focused on empowering and training up staff. Data literacy is something everyone needs at a basic level so it can take a lot of different forms.
You cannot advance in leveraging data if people don’t understand what they are looking at. You need to advance data literacy if you are going to improve you ability to be data-driven, it really is another one of those foundational elements.
Now before we jump into the questions, I want to quickly walk you through what we see as the profile of a mature business, someone who is probably doing data enablement well.
It is our belief that maturity in data quality is a good indication that people are doing something right around data management. We see from our research that companies of all industries and sizes can achieve this level of maturity.
There are a few characteristics these companies share. They are more likely to undertake more data management projects. They are also more likely to see data enablement as a focus (83% vs 57%). They are more likely to have data roles in place, especially a CDO. We see 64% of data quality mature organizations with a CDO vs 38% of others. Data management is more likely to be seen as a set of continuous processes, although these projects are just as likely to sit with IT as they are business users. Finally, they are less likely to see data quality undermining key initiatives and they are less likely to experience data management project delays.
With that, let me put up these final key takeaways one more time.
Again, we see a lot of investment happening, but not everyone is achieving their data-driven outcomes. We see companies are missing some of the key elements they need for data management success, specifically data quality. And finally, organizations need to focus on the right talent, technology and data culture to enable the use of data.
With that, I am going to turn things back over for questions.