Effective Strategies for Building a Data-Driven Culture

November 6, 2017

Effective Strategies for Building a Data-Driven Culture

Slide: 1

It’s been a great day of content and I’m thankful for the opportunity to wrap it up here. Let me dive into what I’m going to talk about today. Effective strategies for building a data-driven culture.

Slide: 2

The first question I have: what is data-driven culture? That was something that kind of hit me. We talk a lot about being more data-driven in organizations. But what does it mean to get a data-driven culture? This is the definition I’ve come up with: An operating environment that seeks to leverage data whenever and wherever possible to enhance business efficiency and effectiveness.

I think that represents what I’ve seen at companies that have a data-driven culture. They’re trying to leverage or use data in all situations. Wherever they can, whenever they can. Obviously the goal is to improve the efficiency and effectiveness of their organizations. One of the questions is why do you want to be data-driven? The obvious one is that then you can make better and faster decisions. It’s going to give you the information that you need to make smarter decisions, and also iterate to them, and make them much faster.

Slide 3:

What does that mean to your business? Well MIT did a study, and they looked at data-driven companies, and compared them against their peers that weren’t as data-driven, and found that the data-driven companies had a 5 to 6 percent higher output in productivity. Nucleus research has come out with research showing companies that are data-driven and are able to take advantage of their analytics. For every 1 dollar they spend on analytics, they’re able to make 13 dollars in return. Also as companies, as they make data more accessible to people, every 10 percent increase in that data accessibility, research stated that you could anticipate a 65 million dollar net income increase for each 10 percent increase in data accessibility.

Slide 4:

So those are some reasons why we want to become data driven. Now the next question is how data-driven is your organization? As you think about how you use data within your organization, most companies have a lot of data.

Slide 5:

You might say, “Well you know, our company is data-driven because we collect a lot of data, we generate a lot of reports.” But I actually feel like that’s not really a good gauge of whether a company is data-driven or not. A lot of companies are going to be, today as you do business, you’re going to be collecting a lot of data on your marketing campaigns, sales performance, or what have you. But that doesn’t necessarily make you data-driven. So I was trying to think, “How can I explain to people the difference, or to really acknowledge how data-driven they are.”

Slide 6:

And I thought of how painful is it for you lose your smartphone? I think that everybody’s gone through this at one stage. Hopefully you haven’t spilled coffee on it, but there’s a time this year where my cellphone went down. All of the sudden, I realized, “Man, I really depend on this device for a lot of things in my life.” I couldn’t reach out to contacts because I didn’t have their phone numbers, I couldn’t use apps that I use on a daily basis. I felt really lost without my smartphone. Getting it back was great.

Slide 7:

So the question I would ask for you: what if you lost access to your data? How would that make you feel? As you individually, and also for your company, how much of an impact would that have on your company? For some people, looking at a spectrum, on the left, you might have some people who it doesn’t really have that much of an impact on a daily or weekly basis if I lost my data for a day or a week. It wouldn’t have that much of an impact because, it’s not really ingrained in how we operate as a company.

If you go to the far right, you start to see it starts to become destructive and painful. I believe those are the companies that are more data-driven, because it’s become such a part of how they operate. How they make decision. The data is so critical to them. So depending on where you are on this scale, we’re going to talk about how you can become more data-driven, and drive that data-driven culture that maybe you don’t currently have.

Slide 8:

There are a number of roadblocks that any company will face when they’re trying to become more data-driven. I thought I’d just run through those.

Slide 9:

I’ve got 10 that I’ve seen in my experience in working in analytics, working with different companies that were very data-driven, and others that were not data-driven. I was able to identify these 10 roadblocks.

I start with probably the most important one, which is obviously if you have weak executive sponsorship, it’s going to be very hard for you to make great progress in becoming a data-driven organization. That top-down leadership and guidance and support in terms prioritization, budgets, getting resources, all of that stuff really depends on having a strong executive sponsor.

An unclear strategy. You can collect a lot of metrics and data, but without that clear strategy to guide your organization around, “this is what’s important to our organization, this is what we’re trying to achieve”. Then it becomes, if you have a clear strategy, it becomes that much more easier for you to then say, “Ok, these are our KPIs”. Once we know what the business objectives, or the key goals are of the organization, the strategic parties for the company this year. All the sudden it then creates clarity on the data and what we need to focus on in terms of KPIs.

Obviously, if people question the quality of the data and they’re not sure if they can trust it because of bad quality, that’s going erode the confidence people have in using data. And they’re just going to go with their gut, or they’re going to go with not using data to make decisions. That can be a problem.

Sometimes you have conflicting versions of the truth. Especially in marketing. You have different systems that may be looking at different aspects of your marketing spend. How do you bring it all together? How do you know what is the one version of truth? You come to a meeting and people are bringing different spreadsheets with different versions of the truth. That can be paralyzing, because we really don’t know how we’re performing as an organization.

A lack of analytics talent. A lot of companies will spend on the technology, but then pull back, or not invest as well on the resources side. That can be a problem, because you want everybody to embrace data, but you’re going to need some people who are experts on it and that could be coaches and help people to get the answers that they need, and answer the business questions when they become more technical, or hard to understand.

Insufficient automation. Again, a lot of companies are spending a lot of time building manual reports. Doing a lot of manual collection and different things like that. It really comes down to, “how can we make this more automated?” We want to have people focusing on what’s strategically important. We want to focus on the high-value tasks, rather than these menial labor intensive things that really, if we took the time and made the investment, we could automate a lot of it. Then that would free up these people to focus on more important areas and tasks.

Weak data-literacy. You create these great reports and data visualizations, and if people are not able to interpret them correctly, or understand the information within them, that’s going to be a problem as well. Part of this could be: what is the data-literacy of your organization?

Say you have people who are very data-literate. But they’re not able to access the information they need? It’s all in different data silos and nobody can get at what they need to make decisions. Again, having that limited access can be a real problem for companies that are trying to move forward more data.

A lack of accountability. I’ve seen this time and time again where organizations build out dashboards, they have all these great KPIs and everything. And when it comes down to holding people accountable, it kind of gets forgotten and overlooked. Really, you’re not going to change your behavior if you’re not being held accountable to hitting a target, or to accomplishing certain goals with your metrics and the data. It’s really about judging the performance. If you want to become a performance-driven organization, accountability is going to be a key part of that. Make sure that people understand that, “We’re going to look at the numbers, we’re going to learn from the numbers, and we’re going to improve.”

Lastly, sometimes you see a fear of failure. A fear of transparency. Nobody wants to looks bad, nobody wants to be seen as making mistakes. I think part of that is cultural. If you shift that culture to where, “we’re going to test, and we’re going to learn, we’re going to be experimenting, we’re going to be trying new things, and sometimes we’re going to make mistakes, and we’re going to learn from those mistakes. And we’re going to apply that back to how we operate, and take the learnings and get smarter over time.” Failing fast. Taking those learnings and applying them back to the business will be really critical.

Slide 10:

When I look at how we address the problems that I just went through, these roadblocks, it distills it down to these four key colors of data culture. You have mindset and skillset, which is more on the people side. Then you have the toolset and the dataset, which is more on the technology side.

From a best practices perspective, looking at mindset, what we want to do is shift the mindset. There’s some key areas that we can do that. One is the executive sponsorship. That can be a huge factor in shifting the mindset. Identifying and leveraging quick wins is another great way. Then, as I mentioned on the last slide, I talked about experimentation, or that test-and-learn mentality to overcome the fear of failure. Shifting the mindset that that’s ok. Those are on the mindset area.

We want to strengthen the skillset. If we can look at how we can make people more data-literate, how we can help them with communicating stories through data storytelling, and also that we make sure that we have the analyst resources that we need for our organization. Do we have dedicated analyst resources that can help make sure that they can support and build up people throughout your organization.

The next area is toolset. You want to sharpen the toolset. That means having a single source of truth. We want to enable people to get the answers, or as many of the answers as they can on their own through self-service models. We want to minimize that labor intensive report generation process, we want to automate as much of that as possible. Also, we want to embed analytics, we want to embed data within our processes, and make sure that it’s starting to see that throughout our processes as we run our business.

The last area is solidified dataset. We want to make sure that we’re aligned with the strategy, that we have good quality data through data governance, and that people understand and can interpret the data by having good context around the data through a data clickstream.

Slide 11:

So today, I’m going to dig into three areas, I’m going to look at quick wins, I’m going to look at leading-by-example, which would fall into the executive sponsorship bucket, and then data-driven meetings, which would fall into process integration. Let me get into the first one: quick wins.

Slide 12:

Quick wins are really great because, at the end of the day, what we’re trying to do here, is we’re trying to get some momentum within the organization. A lot of times with analytics products or projects, they start with a long—it takes a long time to implement, or build out. And that can be a real problem, because we really want to show some success to the organization.

If we were to identify a quick win and what the elements of the quick win is, first is: can we target something that an influential audience will appreciate? This is like a business outcome that’s important to the executive sponsor, or to specific key stakeholder within the organization. Maybe it’s of interest in general to most of the stakeholders or executive team. But we have somebody who’s got an intent interest in something.

The next thing is we want it to be visible. If we do it where it doesn’t get a lot of wide exposure across your organization, it’s going to limit its impact. The more visible the project is, the better. We want to make sure that it’s easily accessible to users and that it’s used on a frequent basis. Then people could realize, “Wow, this really is valuable to us.”

The obvious last one is short time-to-value. This is something that we can complete in 30 to 45 days. It’s got a clear focus scope. It’s going to be low-risk and easy to implement. And also we’ve got a high confidence going into this that’s going to generate a positive impact. If we can get all three of these things, we’re going to have a solid, quick win that’s going to help us and propel our culture forward, because people are going to see that quick win and realize that it’s, “Hey, wow, this stuff works.”

Slide 13:

The key benefits of quick wins, and I think this is something that’s really important, is that first of all, it provides a payoff. So as executives invest in this, the wins help to justify the short-term cost that have been involved. It shows that this is actually going to generate meaning and value to us. It also rewards the early adopters. So as you work with teams, or a team to do a quick win, all of the sudden, now you can show successful they were and promote them as winners and get the momentum behind early adopters, and build morale through their positive feedback that this stuff works.

Also, you get to fine-tune your strategy. Part of the challenge when you’re building out a bigger analytics project, you may not have any earnings for 12 months, or beyond. That can be a real problem. Whereas if you have smaller, more finite quick wins approach, you can take those learnings and apply them to the next project, and the next project, and get smarter over time and learn what works, and what didn’t work so they can be more successful more quickly.

Be having a more quick win approach, you’re able to maintain executive buy-ins. They’re going to see, “Hey, we have some tangible results here. The analytics platform you’re promoting or leveraging is delivering business value.” And it warrants their continued support. That’s really important to make sure they stay onboard.

Internally, with each new quick win, you’re taking people who may have been neutral about data and using it, and converting them over, and turning them into supporters and active advocates for using data in their role or their team. Each new project represents a way to convert people to this and build internal momentum. So quick ones are really powerful, and really time-and-time again I’ve seen them been really successful regardless of whether you’re just starting. Or for a year or two, you’ve kind of struggled, maybe it’s about time generating a quick win and building momentum that way.

Slide 14:

The next tip is leading by example.

Slide 15:

My question would be: how much do your executives care about data? And you’ll probably say, “Brent, they care about data. They bought us X, such and such analytics tools, and yeah, I think they care about it.” The question, though, would be: how do they lead by example in using data? I think this is really where the rubber meets the road. Where this can really amplify your success in creating a data-driven culture. Or can undermine it. Because if executives are all about, “Yeah we need to start using data.”

But employees start seeing them as, “Do as a say, not as I do.” Then the urgency and adoption can fall apart, because if our own executives aren’t willing to use data in their decision-making or in how they operate, is that really important or critical for us in the organization below them to really start relying on data?

Slide 16:

One of the most powerful things that executives can do in terms of executive sponsorship of data and driving a data-driven culture, is by leading-by-example. There’s a couple of areas that I’ve identified. The personal actions that they take, and also in their public activities. One is simply to just using data on a daily basis. That sends a really powerful message to the organization that it matters to executives, if they can see those leaders using data.

It doesn’t mean they have to log into events analytics tools and start using it, but if they are looking at their mobile device, or looking at their dash, or while they’re traveling, or in meetings, that says the data is important to them. It also might probe the reports on questions around the data, making sure that people realize, “Wow, they’re looking at this stuff. They’re using it to make decisions.” It sends a message to the organization that data is important.

We’ve talked a lot about, in web analytics and digital analytics, about HiPPO. That’s the highest paid person’s opinion. Where maybe they make a decision that doesn’t really rely on data. That can completely derail the data-driven culture that you’re trying to set up. Whereas if executives are relying on data, they’re asking for data to make decisions, that reinforces data’s role as an important strategic asset. That is an integral part of decision making. Also holding their reports accountable. So if they’re looking their managers who they manage, and how they’re using data to make decisions at their level, that can be really important.

Lastly, on the personal action side, communication. Every email, every presentation, every meeting discussion, is an opportunity to share insights on the business performance, to promote data-driven wins, and emphasize data’s importance to your organization. When they include metrics and data points in their communications, it shows the top align with them, and it’s something the employees should also embrace.

Moving over to public activities, I’m going to talk about this a little bit later, but meetings are really a great way to leverage data, and make it a part of the decision making process. As we know, executives spend 40 to 50 percent of their time in meetings. Well, we can make those meetings much more productive and focused if we include data in them. I’ll show you in the next section how we do that.

Training. Training, I’ve talked about data-literacy, data storytelling. A lot of these skills are things that need to be taught and learned. A busy executive that says, “You know what, I’m going take time out of my busy schedule, and I’m going to the training.” It sends a message to people also in that training, “Wow, this VP or this SVP is at this meeting where we’re getting trained.” That shows this is important. That we’ve learned these skills and improve and become more data-driven.

Lastly, this is something that we’ve leveraged at our company, having displays, having digital displays throughout your organization in prominent locations. We’re displaying key metrics. It shows that when you have these metrics and targets displayed for a team or the organization, it shows that they’re collectively owned by everybody. They’re a part of how we operate, how we think about our business, how we manage and run the company.

One telecommunication company took this a step further, and added touchscreen displays for senior managers so they could interact with the data and ad hoc meeting places throughout the building and throughout the company.

Slide 17:

The last area that I want to talk about is data-driven meetings.

Slide 18:

As I mentioned earlier, meetings can be really ineffective, they can waste a lot of time and money. A lot of the executives spend a lot of their time in these meetings. One Fortune 50 company estimated it lost 75 million a year due to unproductive meetings. We think if we just add data to these meetings, we can make them more productive.

Slide 19:

Actually, that’s not always the case, if we look at a typical meeting. The first thing, if we’re preparing data for a meeting, we’re going to be spending time analyzing the data, preparing the data, creating visualizations, that’s going to take a lot of time sometimes. Also, the gap between when we start doing that and when we deliver the content, it’s no longer real-time. It could take us days or weeks to prepare and pull and analyze the data, and prepare it for presentation.

Then when we get to the meeting, all of the sudden I might be presenting some data, someone else might have their spreadsheets and different things. And they’re coming together, and we have multiple sources. So we have inaccuracies, we have conflicting data. This adds to the complexity of the meeting, and it decreases trust in the data. Even if we begin to understand, “Ok, we’re set on the data sources that we’re presenting here.” But then it’s like, “What do you mean by that metric? How are you calculating that metric?”

So there’s education or explanation of the data and how it’s calculated and what does it mean. When we finally understand what it means, we might be shocked by unexpected results that come up in that meeting. We might question if somebody got all of the data that they needed to. “Did you go talk to Dave and did you get his data? Oh, you didn’t, well this is incomplete. We’ll have to schedule another meeting.” As you can see, it can be quite ineffective.

Slide 20:

We have this paradigm where a typical meeting, we’re going to be preparing our data, a lot of our time and effort will be about preparing and delivering information. Very little of it will be focused on discussion or action. Then we have a very low effectiveness out of this meeting.

What I’d like to do is change this paradigm. I’d like to move it more of a data-driven meeting. I’ll elaborate what I mean by that, but basically, we’re going to take the amount of effort and shrink it overall. We’re not going to waste as much time having to prepare and meeting, but also we’re going to invert it as well. We’re going to reduce the amount of time we spend on preparation and delivery, and we’re going to change that to discussion and action. I have no empirical data or reports or studies to back this up, but I feel confident that I can at least double the effectiveness of your meetings by taking a data-driven meeting.

Slide 21:

Let’s get into what defines a data-driven meeting. There’s certain criteria that is ideal for a data-driven meeting. It doesn’t mean you can have a data-driven meeting always, but I think there’s certain criteria that makes it really conducive for data-driven meetings. First of all, driving to a specific goal or target is really important. If there’s a clear, “We have a marketing target this quarter, and this is what we need to achieve.” Then there is this desire by the team to review progress towards that goal on a recurring basis. That might be on a daily basis, or weekly basis, or monthly basis is that’s sufficient to understand how much progress you’re making towards that goal.

Then we need data. So there’s got to be sufficient data. It doesn’t mean we have to have complete data, but we have sufficient data or ample data to start monitoring our progress towards the target, and we can start conversations around what’s happening and what we can do about it.

Lastly, that there’s clear actions or leverage we can take to influence the outcome. Rather than just going along and watching us hit or miss a target, there’s actually steps we can take along that journey to influence and hit the target, or exceed the target based on the actions or outcomes that we can influence.

Slide 22:

When you design your data-driven meeting dashboard, your agenda can be set up like this. You have a business goal and you may identify different success factors that contribute to it, and there’s a number of KPIs and supporting metrics. You can then translate that into a meeting dashboard. So you have a collection of cards—that’s what we call it at Domo—there are different charts or data visualizations that are a part of each collection that is tied to one of the success factors.

They way that can look is say, for example, from a marketing perspective, we’re trying to generate 2500 leads in this quarter. So we have different marketing teams that are working towards this goal. Our digital marketing, event marketing, TV marketing, and print marketing, and how the different metrics they’re looking at to gauge from a leading and lagging perspective whether we’re making progress towards that goal.

Slide 23:

I thought it might be helpful to show what this may look like, to make it a little more tangible. This is based on a real customer example where a president of retail wanted to understand how his business was performing. So we had a meeting agenda, right now these are claps, let me elaborate what they are. Basically, the first section was focused on retail financials. So what was the performance up to this point? What were the retail forecasts? What are we expecting the revenue to be? Then there’s a number of key drivers that they identified and drilled down into each of those KPIs.

Slide 24:

After looking at how the financials and the forecasts were, he could drive into the key drivers overview and look at a summary of how those key metrics, then he might start to ask questions about one of the metrics, say, metric number one.

Slide 25:

Then he could go into a drill down of what’s influencing that KPI, what are the different component or supporting metrics behind that and start questions that maybe his people don’t want him to ask. But basically this is the format of their agenda. On a weekly basis, they’re going to this same dashboard and answering questions they may have and determining how they’re going to proceed forward.

Slide 26:

Some of the benefits of a data-driven meeting are one, it’s goal-driven and action-oriented. You have a target in mind. You’re driving towards that. Everything that you’re looking at is now about achieving that goal. A streamline a focus. You no longer have these different spreadsheets, and different metrics, and different views of how effective things are working. Everybody’s looking at the same reports, everybody’s looking at the same data, and have the same understanding of where they are and what needs to be accomplished.

There’s clarity and confidence in the numbers. It’s no longer Jeff’s spreadsheet or Jill’s spreadsheets, we’re looking at a dataset that’s been cleansed, that we understand. We understand the metrics, we know what they mean. We have confidence in the data.

And we have greater transparency. You’re not showing up to this meeting and for the first time you’re seeing the new numbers. So anybody before that meeting can actually see that data in the dashboard. They can go in and prepare for the meeting. Maybe there’s something that they need to talk about, they need to be prepared to cover in that meeting to explain how things are trending up or down in their respective area. There’s this greater transparency so people are coming to that meeting more prepared to talk about taking action and driving things forward.

Lastly, there’s more accountability. Everybody knows what the numbers are. Everybody knows what the tasks are and the jobs they need to do, and there’s going to be follow through built into this. As people, as they see their numbers and they see where they need to work, they can work on those areas and drive the business forward.

Slide 27:

Going back to the four pillars, I covered three areas here. I covered on the executive sponsorship with the leading by example, the quick wins. And also from a data-driven meeting perspective and that process integration. There’s lots of work here. There’s lots of different ways you can create a data-driven culture.

Slide 28:

Cultural change is not easy. It’s something that can be quite challenging. Here is this quote from Gabie Boko from HP, formally at Sage, she basically says, “we can get stuck in our ways. The ways that we’ve always done things.” If you’d like to see your culture to become more data-driven, I believe by starting on those three areas I just covered, it’s a great starting point for shaping that data-driven culture that you’d like to see at your organization.

Slide 29:

I’d like to end with that. Thank you for sticking around and watching the last presentation of the day. Thank you very much and look forward to your questions.

Previous Video
Actionable Uses of Google Analytics
Actionable Uses of Google Analytics

Explore ways to incorporate AI & ML into your marketing—or risk being left behind.

Next Video
Customer Intelligence Rockstar Tips and Tricks
Customer Intelligence Rockstar Tips and Tricks

Rapid-fire tips and tricks to help you get more value out of your customer data in Adobe Analytics.