Gaining competitive advantage from data analytics: businesses in 2018 are very focused on leveraging this emerging technology. Yet where there’s a lot of power in analytics, there’s a lot of confusion in actually using it. Gaining consensus within a company to using analytics to its greatest potential is a true challenge.
To discuss data analytics challenges like this, I spoke with two top experts:
Andi Mann, Chief Technology Advocate, Splunk
Bill Schmarzo, CTO, IoT and Analytics
Under the video, please see the edited transcript.
Gaining Competitive Advantage from Data Analytics
J. Maguire:
The question is, and Bill I’ll ask you this first, how do you gain alignment and consensus around analytics within a company, to make the most of it?
B. Schmarzo:
One of the things we’re doing, James, is we’re actually embracing concepts called “design thinking”, which may seem very bizarre, that I’m talking design thinking when I’m talking analytics. There are a lot of things that design thinking does.
One of the things I find very powerful about design thinking is that it has techniques for bringing everybody into the process. The challenge for analytics isn’t doing the analytics in many cases, though dealing with data is always a tough problem–Andi probably knows that as well as anybody–in many cases, your data scientists will come up with great algorithms for problems that no one cares about. The way to ensure that you are actually driving the operation and the adoption of the analytics, is to bring people early into the process.
We really embrace these concepts around design thinking as a way to bring everybody into the process. We know that analytics can be built, and really great insights can be uncovered, but if you don’t have the key stake holders brought into the process early so that they have a chance to contribute and understand what you’re building, your chances of operationalizing those and driving adoption, are very slim.
A lot of what we do early on in the process has almost nothing to do with technology, nothing to do with analytics and data, and has everything to do with organizational change and adoption.
J. Maguire:
How does the concept of design thinking deal with issues with power plays within the organization. Does design thinking address those issues?
B. Schmarzo:
Design thinking will certainly surface those issues. But if you have organizational power plays, move on. It’s just not going to happen. We can’t work magic. We can work magic with data, but we can’t work magic with people. If you’ve got organizational power plays going on, design thinking will raise the awareness and make people aware of what’s happening, and maybe even make them understand the cost of it–but it won’t solve it.
Let me give you a really good example. Data silos. There was a point in time when data silos were a technology issue. Not anymore. It’s a political organizational issue. Where organizations don’t want to give up their data. Design thinking can make you aware, can make you understand the cost of that and the pain, but it can’t make people change.
J. Maguire:
Andi, what would you say in terms of building consensus and alignment around leveraging big data within a company? Difficult topic, to be sure.
A. Mann:
Bill is absolutely right, for a number of areas that technology is easy, and people are hard. We can tell technology what to do, but people have a mind of their own, for better or worse.
Really, it’s about showing benefits. It’s about working with business. It’s one thing that I believe strongly in terms of innovation, is that you can’t drive it just out of one person or one business. I.T. can’t say, “We’re going to do analytics.” Because, if you haven’t got a consumer that wants to pull that from you, then you’re just never going to get the traction in the organization. You’re never going to overcome those politics.
For I.T. leadership, for analytics leadership, working with business stakeholders to try and get that real alignment–we’ve talked about I.T. business alignment for years, and I think working with the business areas to get that alignment, what is analytics going to do for you? What are you actually going to get out of it? We can provide data leaks until we drown, but we’re not going to give useful outcomes to our business peers, then we’re not really doing the right job for them.
B. Schmarzo:
Please let me add one other really key point and build on what Andi is saying. What we find with organizations that are trying to digitally transform, how they’re trying to become more effective alignment leveraging data and analytics to power their business models, is this little phrase. “You have to inspire, you can’t mandate change.”
The senior people in an organization can’t mandate and say, “This is what’s going to happen.” You have to inspire people so that they learn and feel like they can internalize it. Mandates don’t work. Inspiration is everything in this space.
J. Maguire:
However, inspiration when it comes to the data analytics may be challenging. The numbers and the metrics are coming out, they need to be interpreted in a certain way, there may be disagreement on how they’re interpreted and what they might mean and what direction we should go in based in those metrics. Where does inspiration come in there, in some of those authority issues?
B. Schmarzo:
I think inspiration is the whole basis for why you’re even doing it. Why are you undertaking this process? Why are you taking an organization that’s even keel, and probably operating just fine, and now you’re going to toss a snake in the kitty pool. It’s got to be inspired by something.
Maybe it’s Amazon moving into your industry, or maybe it’s a senior leader who has a broader vision and uses this ability to inspire people to help make this change happen. If you can’t get the rank and file to embrace this, and start using analytics to make better decisions, then you’re just not going to get the value from it that you really can.
Overcoming Data Analytics Challenges
J. Maguire:
Let’s drill down a little bit and look at some of the biggest challenges to making analytics work for your organization. Andi, what might be a challenge and what would you do to overcome that challenge?
A. Mann:
The inspiration challenge, I think Bill is really spot on, is that there is a push-me-pull-you aspect to this as well. Where, say if a sales rep comes to you and says, “I really want to figure out, how can I engage better with my constituents?” And then, as a developer, as a product leader, even someone in I.T. ops, you can start to work with that person and go, “Well let me see what I can show you, let’s try some things, let’s do some AB testing, some multi-variant testing, some blue-green, let’s see if we can’t tweak some levers to get that engagement up. You can have a really positive two-way conversation there.
The other alternative, and I say this occasionally as well, is when a product owner, a developer, or even an operations person can come to that business leader or sales leader or marketing leader, and say, “I was looking alignment the data of the mobile app,” or if you’re a website off the order entry application, maybe out of cash flow. I was looking alignment this data and I found these interesting patterns, I found these outcomes. Would it be possible to drill more into that?
Going to them and saying, “Look, I found that, when people are working around the store they use their mobile app quite a lot to check prices on a different website. Would you like to drill down into that?” I think that we can, as I.T. leaders, inspire out business peers, because we’ve been able to look at data.
If they’re not inspired at all, then Bill’s right, you may as well turn your energies to something different. But, I think that if you can find that common ground, whether it’s the business peers coming to the I.T. people with something they need, or I.T. people going to the business peers with something they’ve found out, if you can find that common ground I think you’ve really got somewhere to go.
J. Maguire:
Common ground between I.T. people and business people, Andi, such a thing does happen?
B. Schmarzo:
Cats and dogs, playing together in the streets! What’s going on here!
A. Mann:
Lions and lambs, lying down together. I know it happens James, it can be possible.
J. Maguire:
Bill, what’s your take? If there’s a challenge to using data analytics, if there’s an obstacle, how might you overcome at least one of them?
B. Schmarzo:
The biggest challenge that I find, is that people–especially older people–who are entrenched which really means senior executives, they do a crappy job of unlearning. They don’t know how to unlearn what they’ve learned. Things that we’ve held as gospel; how you build a data warehouse, how you build reports and dashboards, what the heck even analytics is, the definitions have changed.
And they will continue to change. You’ve got a lot of people in senior leadership who got where they are because they rode an E.R.P. implementation, they rode that to success. A lot of bodies lay on the side of the rode from E.R.P. projects, but this one person survived and now they’re the C.I.O., and their whole world looks like E.R.P. and they don’t understand.
They can’t grasp this whole new heavy-data, big data, science world that’s in front of them. I think that senior executives struggle with unlearning, and that is the biggest challenge.
J. Maguire:
Well of course some of the senior executives have had enough time in the field to have a gut instinct. Gut instinct vs. the numbers, and what do you do with that? Andi, what would you say?
A. Mann:
There’s a lot you can do with gut instinct. I am actually a very pragmatic kind of operator, myself. There are some decisions you don’t have enough data to make. There are some decisions that, even if you’ve got a lot of data, you still need to bring intuition and experience to the party.
But, I am very much with Bill, and I’m always up for a good Star Wars quote, so yes you “must unlearn what you have learned.” I can’t help it. It’s a very important point. I think, bringing data to the party can also be deceptive. You can bring data to a decision point, where if you don’t have the right data, you don’t have enough data, you don’t have convincing data, it can even be counterintuitive at times. I think it’s about striking the balance, but I think you’ve got to challenge traditional expectation.
It’s something I wrote about in a book on innovation a long time ago. Your best innovation is not going to come from your senior leadership. It’s going to come from junior people. One of the C.I.O.s I interviewed at that time from my book came out and said that they always sat down with their interns once a month, because they didn’t know what they weren’t allowed to do. They could break convention. Getting data in front of senior executives and challenging their preconceptions with that data.
This comes back to things like the challenger sell, the whole challenge around how do you sell ideas to people? One of those ways is to challenge their preconception. Show them data that says, “Look, this is what I found when I went to my data. Our application is not being used the way you think it’s being used. Our website is not driving the level of engagement, the level of business. Our stickiness is not there. Click through rates are not showing up like you think they are. “
Coming with that data, when you’ve got two opinions–someone said this the other day, if it’s just a matter of opinion then mine’s right.
Are Real Businesses Really Succeeding with Data Analytics?
J. Maguire:
Where businesses are really at with using analytics: are businesses really on board and truly guiding themselves with analytics? Or is this still more of a floundering nascent field, and we’re guiding our way through the darkness. Bill what’s your take?
B. Schmarzo:
I think we’re floundering. I think you’ll see pockets in organizations who have embraced analytics but as a whole, we won’t see universal change until we change the way people get paid.
When I was at Proctor & Gamble there was this great quote, “You are what you measure, and you measure what you reward.” Which is, you are how you pay people. If we pay people the way we’ve always paid them in the past, they’ll continue to do the things they’ve always done. How do you use compensation and recognition to change behaviors? How about paying people, incent people, based on sharing?
Sharing data, sharing analytics, sharing best practices. Create a collaborative environment. Reward people for collaboration. We don’t do that. We reward on outcomes, in some cases, or reward on effort. You only reward on outcomes if you’re in sales, you show up and put in your eight or nine hours and that’s good enough. There’s a lot that we need to do at an organizational level in order to harvest the real benefits that are out there in analytics.
A. Mann:
Looking at the sharing thing is really critical. I can’t tell you how many times I’ve worked with application developers who are seeing patterns, they’re getting their data, they’re seeing patterns in things like customer engagement, revenue generation, click through raids, all sorts of stuff. But they have no avenue to work with their business peers in sales or marketing and say, “Hey, look what I’m seeing, this is really important.” If you don’t share those platforms, I think that’s a really powerful barrier to any of this.
J. Maguire:
Andi, what’s your take on where businesses are on really using analytics: are they floundering? Or are they pretty well on board?
A. Mann:
The answer to that question is an absolute and 100% yes. [Laughs] There are organizations that are floundering. I see this very deeply, of course. Splunk is an analytics company and data company so I’m working all the time with people who are neck deep in all of this, so I see people who are floundering a lot.
Part of the big challenge is being unable to get access to the right levels of data to do the analytics, you know, the web team has the website locked down, but I’m a database person so I just see database data. And neither ne’er the twain shall meet. The sharing action, getting the data in, sharing it, having the people understand how to freeze it, so analytics people, having a data scientist to be able to understand “What can I do, what is out of the possible?”
Learning the right machine, the right algorithms, the right training data is also really critical. So, there are a lot of barriers in the way and I don’t blame people for floundering. But, by the same token, I’m seeing some organizations who are getting incredible results.
You look at Domino’s. Domino’s applies analytics to their pizza making, to their marketing, to their coupon business, to their intern operations and development teams, and they’re doing amazing things because of it. They’re able to balance the workload between different pizza shops, based on real time feedback on how well their coupons are working.
You look at airports, for example. Dubai, Gatwick, able to reduce the wait time in security from 25-30 minutes down to five minutes on average, simply by looking at the data on where are their customers, where are their security personnel, how long is it taking to get through the line, where is baggage? They’re able to reduce the time it takes for them to get their bags off a plane, they’re able to improve the Wi-Fi service through the airport, they’re able to improve the experience at security.
Technology benefits, business benefits, customer benefits, all through the use of analytics. It’s a mixed bag, James. Some companies are doing amazing things, and a lot of companies are, to Bill’s point, still trying to figure out what even can I do with analytics?
Looking to the Future: Who Will Own Data Analytics?
J. Maguire:
Where we will be with analytics a couple years down the road? Specifically, who will “own” analytics? Will the C-suite own it, will the brave, hardworking folks in the cubicles own it? Who will have access to it, and who will be making decisions based on it? Going forward, Bill, who will own analytics and what will that mean for the future of analytics?
B. Schmarzo:
Everyone.
J. Maguire:
Everyone! Yes, I would hope so. But, if everyone owns it, then who makes the decisions based on it?
B. Schmarzo:
Everyone! I teach at the University of San Francisco, I teach a class called ‘The Big Data NBA’. What we’re doing as part of that class, is we’re teaching every business student to embrace analytics as a business discipline. Like you would do it in marketing or accounting or anything else, right? It’s a business discipline.
What we teach is, we’re not trying to turn business people into data scientists, we’re trying to turn them into citizens of data science. We have a whole curriculum around, “How do I think like a data scientist? How do I start embracing what data science can do?” It’s an approach and framework problem versus a tools problem.
When we’re successful, we’ll have everyone in the organization looking at data and analytics to help them make better decisions. It won’t replace intuition, it won’t replace the human empathy factors, or things that are really hard to codify in a model. But it will help people to make better decisions with respect to what it is their organization is trying to accomplish.
I think in the end, when it’s successful, it’s just part of what we do. Let’s go back to Amazon for a second, I think Amazon is very interesting. What Amazon has done, and what they are progressing to do, they’re not looking to optimize key business processes, they’re looking to eliminate them. And they’re focusing their entire transformation process–I just wrote a blog on this called ‘Monetize the Pain’–they look at their customers’ pain points, and they are focused on, “How do I monetize pain?”
Think about the pain of me having to go to the store and getting a new box of Cap’n Crunch? My box is empty, right? Instead of having to drive to the store and find a coupon and fight traffic. What do I do now? I go, “Hey, Alexa!” Order me two boxes of Cap’n Crunch!” Boom, and it shows up.
What they’re masters of, is understanding the customer journey–which is a design thinking technique–and identifying the pain points and eliminating them. Monetizing them.
J. Maguire:
And that requires serious number crunching to do so.
B. Schmarzo:
It requires not only serious number crunching, but a deep understanding of your customers and what they’re trying to accomplish. It requires you to understand their personas and customer journey maps, it requires you to understand economics, it requires you to go old school Michael Porter value chain analysis, it brings everything together around how you’re going to drive and derive new sources of digital value.
J. Maguire:
Andi, if we’re talking about analytics a couple of years from now, what we’ll be talking about specifically, and what about the ownership question? Who will own it, and if everyone owns it, how will we make decisions around it?
A. Mann:
I agree with Bill to extent, and Bill I hope you realize, you’ve just ordered cereal [all laughing] for tons of people.
J. Maguire:
There are eight boxes of Cap’n Crunch coming to your house!
B. Schmarzo:
Good, I could use the extra calories.
A. Mann:
I agree with Bill to an extent, and I see the patterns that we’ve seen in the past. Databases were the pervue of someone very very deep and significantly into the organization. And slowly it started to build out into other areas, and we had better ways of using data.
“Digital” was another one. We now have chief digital offices, and they “own” digital, but it’s part of a coordination function in the best businesses. And in the best businesses, everyone owns digital. It’s like saying, “Who owns mobile? Does sales own it, does marketing own it, does I.T. own it? Hell, does networking own it?” Everyone has to have a piece of mobile because that’s how we do business today.
And I think, to that extent, Bill is right. Everyone is going to have to own analytics to an extent. Some people will be analytics consumers, some people will be analytics providers within the business. I think there will still be centers of excellence around analytics, you’ll still have scientists who can do the really in depth, Ph.D. level number crunching, building the algorithms for example.
While you’ve got data consumers, who will be everyone to Bill’s point, they need to be able to consume that data or make business decisions based on that data. Just as a sales person needs to understand the financing of their product, just as a finance person needs to understand the tempo of marketing campaigns, so they can fund them and get the ROI on them. Everyone is going to need to understand, “How do I use data in my role,” and pull that data from the providers within their business. Maybe that’s a “chief data officer” or a “chief data department,” maybe it’s not. But I think everyone is going to have to consume but not everyone will provide.
J. Maguire:
Well, but of course, if you are a data consumer, by necessity you would have to be a data analyst, because you’ve got to understand what that data means and how to analyze it. True or false?
A. Mann:
Yes and no. Historically we would have said that you need to be an IT person to use information technologies. And we went through consumerization of IT. And now it’s more the case that there is more computing power out in the hands of our employees, then there is in our data center.
Now we may not even have a data center. I think that, yes there will need to be data analytics professionals, but the process of doing data analytics is going to get so much easier. It’s going to be like an iPhone in our hand, as opposed to the I.T. department obvious to you.
B. Schmarzo:
I love that phrase, “The consumerization of analytics.” I love that. I think that’s a really powerful concept. Think about how relevant the iPhone is. We no longer call an operator to make phone calls. It used to be a point and time in day when if you had to make a phone call, you’d call an operator and they’d plug things in. The consumerization of communications. Now I think you’re spot on. It’s the consumerization of analytics, I love that. I love that.
J. Maguire:
I hear a blog post coming.
B. Schmarzo:
Amen, I’m working on it right now!
J. Maguire:
I thank the two of you for your expertise. Thank you very much!
B. Schmarzo:
This was great. Andi, great to meet you, I love how you think.
A. Mann:
Thank you, Bill! Always great to meet a fellow traveler in the area of analytics. Really enjoyed the conversation, James, thanks so much for having us both.
