'You have to take this mass of information and distill it down...'
Jim Stikeleather is chief innovation officer at Dell. Lucy Handley speaks to him about managing data in a large business, how to make the most out of data produced by social media, and cyber crime.
Marketing Week (MW): How do you manage data in such a large business?
Jim Stikeleather (JS): That is probably the $64bn question. The next big issue for IT is how you are going to deal with ‘big data’ [which means large quantities of data from every source].
This is not just true of Dell. Even the smallest companies are being inundated with data.
Traditional business intelligence systems aren’t going to work. By the time you have amassed the data, put it into the database, run the analysis and done the queries, it is already obsolete because you have this huge real-time feed of data coming in.
Traditional market research might involve a survey, analysis and then six months to a year later, you have the information to make a decision about your new product or service.
This contrasts with social media, which is all real-time and right there. The way most companies - and, to a degree, ourselves - are doing market research is long-term. But we have to move to embedding real-time analytics right into the data stream.
We haven’t quite got it figured out yet, but we have a lot of people monitoring the comments made on social media sites, so we still have a very heavy human intervention component in our analysis of large data.
MW: How can businesses best use social media and crowdsourcing?
JS: Everyone talks about the term ‘social media’, and from that perspective it’s more closely related to television, radio and print [as a marketing channel].
But it is really a social medium. It is a platform that is good for generating and sharing knowledge. But the problem is that there is lots of noise, it is very disorganised and there is a lot of polarisation that goes on.
The problem with an organisation participating in social media is that you have the issue of governance without control. There are legal and regulatory [considerations]. You can’t make forward predictive statements, for example.
You have to look at social media in terms of what it is good for, what it’s not good for and how you are going to use it
You have to look at social media in terms of what it is good for, what it’s not good for and how you are going to use it. It is really good for idea synergy. People and companies are talking to their customers, but also their competitors and suppliers, and they should keep that in mind.
There are synergistic explosions of creativity and that is the reason why companies want to be involved with it. What social computing allows you to do is get more diversity in ideas than you could otherwise. You can start hearing small voices.
There is a famous quote by Henry Ford that goes: “If I’d asked my customers what they wanted, they would have said a faster horse.” That is the fallacy of using only traditional things like market research or consultants, which might do a statistical analysis and lose those small voices. Social media brings them back in.
You’ll hear the person who isn’t asking for the faster horse, but is asking for a non-biological form of transportation, for example.
Also, because there are so many people looking and talking about what is going on, there is the general wisdom of the crowd. You’ve got lots of intelligent, motivated people talking and they generally have better judgement than individuals do.
MW: How can brands work out what data is useful from social media interactions?
JS: You have to be very careful because the content is very disorganised. You’ve got to have a process for taking this mass of information and distilling it down.
A lot of companies make the mistake of dealing with the noise because there will be one or two people who are screaming and yelling, but they are totally inaccurate. They might try to deal with those comments instead of looking for the nuggets of gold.
Companies can get caught up in the quantity rather than the depth of information. You have to find some way to maintain your governance, when, where and how you are communicating, and at the same time recognise that you have almost no control of whatever is going on.
MW: Dell implemented 400 of 10,000 ideas suggested on your crowdsourcing platform IdeaStorm. How did you pick those?
JS: There is a sense of organisation you have to attach to any type of social media function. This is the process of how you let this wisdom of the crowd distill your ideas.
People post ideas, complaints or recommendations and as others are looking at them, they give them the thumbs up or thumbs down and they can add comments.
What happens is you get a filtering effect - the cream rises to the top. You have to have a mechanism to allow that to happen. If you are basically creating an electronic suggestion box, it is not going to be terribly useful.
MW: How can companies contribute to the discussions on social media?
JS: You have to observe what is going on, not engage. It’s almost as if you are a teenager, who has moved to a new city and school. You would just watch who is in the ‘in crowd’ and behaviours that are seen there. Then you can start orientating yourself and ask: “Considering what the culture is, what do they expect and what will they let me legitimately do?”
You can jump up and down, saying you are the world’s most innovative company, but if people don’t have any expectation of innovation from you, it won’t accomplish anything.
People post ideas, complaints or recommendations and as others are looking at them, they give them the thumbs up or the thumbs down and they can add comments
Then you can begin assuming the character that the environment, culture or ecosystem has painted you into. Once you have done that, you can try to change a perception.
If [the group] has misinformation, you have to find a way to give them the valid information, or [it may be that you] have to change something about yourself.
Then you can accordingly go in and start sharing your information. If people are critical, you can say: “Here’s what I think I heard you say and here’s what I think I can do about it.”
If you don’t agree with what is being said, you can say: “Thank you very much. We think that is a misperception and maybe came about because of this.” So you observe, act and process.
MW: How can humans intervene with data analytics?
JS: In researching my [forthcoming] book, I came across something that suggested that the US Air Force had sent out a solicitation for someone to build avatars for them, so that they could observe what was going on [online], interact with them and manipulate the network.
To be able to do that, you have to have real-time analytics in place that match your timeframe - that is the big change that is going to take place. You have to let those analytics start making decisions for you.
But the issue is how you get humans back into the analytics to provide governance when you don’t have control. There is work going on looking at complex adaptive systems theory, such as at the Santa Fe Institute in New Mexico. It is working on how form and substance can emerge from chaos, which is really the governance without control issue.
MW: How do you feel about Dell’s ability to protect data and prevent cyber crime?
JS: At a recent conference, someone said: “We have had crime since we could climb down from the trees and someone stole someone else’s squirrel. It is going to be there, so get over it.” There is no such thing as 100% perfect protection.
It becomes an issue of risk management, [so it means considering] what the level of risk is, versus the value of the loss and adjusting yourself accordingly.
We recently acquired a company called SecureWorks and one of the things it is excellent at is helping people figure out where their valuable data is and how to protect it. The big thing in the innovation world is the concept of open innovation, where you say: “I’m going to give all my data to the world and they are going to come back and give us their best suggestions of what I can do.”
So you give all your information to customers, suppliers or partners and then you sit down together and create value. That is a very different mindset versus proprietary intellectual property. [People need to consider] what they have to do from a legal or regulatory perspective and then work out what information they need to take advantage of open innovation and co-creation.
At the same time, they must protect those things they really don’t want the world to know about. It is all much more complicated than it used to be.
Stikeleather on innovation
Marketing Week (MW): What does your job involve?
Jim Stikeleather (JS): I focus on innovation and the process of it within Dell. It’s about how you innovate the way you work, rather than specific products and services. The biggest single challenge is that no-one has a really good definition of what innovation is.
I also face the task of changing the misperception that innovation is when you take a lot of weird creative people, throw them into a room, send them lattes to drink and out come these great ideas. Innovation can be a systemic process and you can measure and manage it.
MW: How do you make sure you have the right people to work on innovation?
JS: Most businesses, especially technology companies, tend to be dominated by engineering mindsets. To be successful at innovation, you have to take the detailed specifics and data points of that mindset and combine them with creative divergent thinking.
It is about trying to bring two cultures of thinking and management together, almost like throwing together TBWA\Chiat\Day - one of the most creative ad agencies in the world - with Intel, one of the most engineering-driven companies in the world. It is not exactly the easiest thing to do.
MW: How do you manage innovation?
JS: We have been working on something called the ‘innovation architecture’. We talk about what conditions are needed to increase the probability of being innovative.
This is the engineer in me trying to make a systemic architecture out of it. One example is foresight, when you look at what is going on in the economy, demographics, business and social systems. You don’t so much project what is going to happen but you project how people are going to think and make decisions.
The people who are really good at that are science fiction writers. They have been much better than business or economic forecasters at predicting what the future would look like - unlike the engineers, who were predicting flying cars and that kind of stuff.
Data 3: Jim Stikeleather outlines the three most crucial skills any innovation specialist should possess
1. There is a new book coming out called The Future of the World Belongs to the Polymath, taking the really varied and different backgrounds [people have]. I think the reason I’m having some success is that I am kind of like that. I’m almost 60 years old and I race motorcycles. My background is engineering but I do calligraphy. My hobby is complex adaptive systems theory. My undergraduate degree is in computer science and my graduate work is in marketing.
2. Success has got to be more driven by the ability of generalists who are able to find and bring together specialists. This is opposed to what the past few decades have been about, which is increasing levels of specialisation and ‘verticalisation’.
3. If you are talking to people about getting into social media, you have to observe without interfering, understand what is really going on, what the perception of you is. Then you will know what role you can legitimately take, and then decide what change you want to make in influencing that culture. Gradually infuse yourself into it.
It is said: “The definition of intelligence is the ability to hold two diametrically opposed ideas in the mind at the same time.” For me, that is training. There is the whole concept of foresight, the creative, marketing, advertising movie script-writing capability. Then there is insight. So it is a combination of what your research and development tells you is going to happen and what your marketing trends tell you. The training is how you can be a ‘managed schizophrenic’ and be able to be creative and holistic one minute [then use insight the next]. It’s about being able to take them together.
A career in dates
1972-1977: Jim Stikeleather completes degrees in physics and computer science and works as a systems programmer and sales engineer at various software companies. 1978” Becomes a director at GTE Data Services.
1987: Completes MBA at the University of South Florida. Job roles include vice-president of marketing at a start-up, director of strategic planning at Comtech and applications development director
1993: Becomes chief technology, marketing and quality officer at Technical Resource Connection (TRC), helping to grow revenues from $250,000 in 1993 to $48m in 1998.
1996: Publishes book, Next Generation Computing.
1998: Becomes a director at Perot Systems after it buys out TRC.
2002: Joins packaging and systems developer Meadwestvaco in various board advisory roles, rising to new ventures chief technology officer.
2009: Rejoins Perot Systems after it is acquired by Dell, rising to chief innovation officer of Dell in July 2010.
2012: Stikeleather’s new book, Business Innovation in the Cloud, is due out this year.