Silicon Valley State of Mind, a blog by John Weathington, "The Science of Success"
  • @SVSOM
    @SVSOM

    Welcome to a Silicon Valley State of Mind, thoughts tips and advice based on the consulting work of John Weathington, "Silicon Valley's Top Information Strategist."

  • bio
    bio

Silicon Valley State of Mind

Tips, thoughts, and advice based on the consulting work of John Weathington, "The Science of Success."

  • Home
    Home This is where you can find all the blog posts throughout the site.
  • Categories
    Categories Displays a list of categories from this blog.
  • Tags
    Tags Displays a list of tags that has been used in the blog.
  • Archives
    Archives Contains a list of blog posts that were created previously.
Archive, May 2015. Switch to list view

    Posted by on in Information Exploitation

    If your products and services don’t serve the data science community; however, you’re using data science in your products and services for a competitive advantage, you’re in a popular but challenging situation with your customers. I call what I’ve just described: using data science as a supporting strategy. For instance, the people at Graze.com incorporate data science into their snack business to develop and deliver the next box of goodies their customer will get. Let’s be clear though: they’re in the snack business, not the data science business. In this situation, I recommend keeping your data scientists as far away from your customers as possible. If you’re using big data as a supporting strategy, make it a priority to keep your customers insulated from your data science.

    Buffering

    Buffering is an important strategy for leaders using data science as a supporting strategy. In short, buffering is structuring at least one organizational layer between your data science team and your customers. Contrast this to leaders using data science as a core strategy–selling products and services to other data scientists, like RapidMiner, Kitenga (now part of Dell), and Cloudera. In this case, it’s a great idea to put your data science team in front of your customers, because like attracts like. However, Graze.com’s snackers have no interest in data science, so in this case, keep the analytics out of the conversation.

    Instead, have your customers interface with other people in your company who are like them. The same “like attracts like” concept applies. If you’re in the business of wearables for athletic people, put a layer of athletic-minded people between your customers and your data science team. A good friend of mine is a triathlete that runs analytics to help other triathletes compete. Although he’s an analytic, he wears his triathlete persona when addressing his customers. Since he’s a one-man shop, that’s his only choice. In a larger company, this concept should obtain as a sales and marketing layer comprised of athletes–not engineers.

    Translation

    One important job of the buffering organization is to translate what the data scientists are trying to accomplish, into terms your customers understand. The reason why you don’t put data scientists in front of non-analytics, is that they’re typically difficult to relate to. Imagine a group of pro football players showing up at Comic-con. The first time a trekkie introduces themself to a linebacker in Klingon, there will be a problem. Before a product or service is introduced to your customers, it must be sanitized from its analytic underpinnings.

    When Progressive talks to its clients about its SnapShot device, there’s no discussion about analytics. Their marketing may allude to the scientific prowess that goes into their product for effect; however, in practice they call it usage-based insurance. This is a perfect example of translation. Most drivers understand the term usage-based insurance. You’ll quickly lose them if you start talking about behavior-based digital profiling using a synthesis of regression and machine learning algorithms.

    It may take multiple layers within the organization to successfully translate your analytic-based competitive advantage into customer-facing language. I’ve worked with several organizations where the developers are three or four levels removed from the customer. When I worked with Visa, there was a product development group, product function group, business analyst group, and then developers and architects. Sometimes it takes multiple translations to get it right for the customer.

    Curating

    Curating is a special requirement for those integrating advanced analytics into their products and services. A special challenge the buffering organization has with their analytic brain trust is information overload. Curating sifts through the piles of brilliance to extricate the golden nuggets that will appeal to your customers. That’s no easy feat.

    Consider a museum curator whose job is to process archeological findings into a display of wonderment. Piles and piles of ancient bones, tools, and artifacts must be reduced, organized, and displayed in a way the appeals to the masses. Curators do more than just translate–they manage and oversee their body of work, and interact with the viewing public.

    In a similar fashion, your curators must own the body of work produced by your data science team. Whether or not you put your curators in direct contact with your customer (both ways work), they should synthesize the wealth of information produced by your data scientists into a concise, attractive package that your customers will relate to. Even if you translate well, if you don’t curate, you’ll hit your target market with too much information and they’ll find a competitor that’s easier to understand.

    Summary

    There’s no doubt your data scientists are brilliant; however, too much brilliance for your uninitiated customers will drive them away. If you incorporate fancy analytics into your products, but your customers aren’t really jazzed by math and science, save the tech-speak for your in-house design team. As you structure your organization, ensure there’s a buffer between your data science team and your customers, who can translate and curate their findings. If I’m a Graze.com customer, I don’t want a lecture on how to design the perfect meal–I just want a snack.

    Submitted for Publication in TechRepublic’s Big Data Analytics Blog

    This is the sneak peak of my latest contribution to TechRepublic’s Big Data Analytics Blog. As editors do, when this gets published, some of the words and content may be arranged or deleted for a variety of reasons including SEO. What you’re looking at here is the uncut, unabridged, unedited version of the article that was submitted.

    Rate this blog entry:
    0

    Posted by on in Information Exploitation

    If you can’t get your data scientists and other analytics to be concise, you’ll never get anything done. To make the most effective use of your time, educate and coach your analytics on how to be concise.

    As much as I love working with data scientists, this has to be the most frustrating part of my job. Analytic managers and consultants like me are responsible for getting things done; however, the very talented resources we deal with value brilliance over deadlines.

    Notwithstanding their analytic disposition, everyone–including your data scientists–wants to succeed. To succeed, they’ll need to be concise: in their speech, in their writing, and even in their approach to solving difficult problems. Here are my seven best tips for making that happen.

    Tip # 1: Analyze the Impact

    You must do your homework before you broach the subject of concision. At the onset concision is very uncomfortable with analytics, so they’ll need to rationalize it for themselves before they unseat their de facto behavior. So when in Rome, do as the Romans. Do some research on the benefits of concision and the costs of not being concise, and prepare some analysis.

    In my experience, you can double or triple your productivity, when your team effectively practices concise behaviors. You’ll need the numbers for your specific situation to make it relevant. High-level studies are interesting, but when the analysis is brought into their reality, it becomes impactful.

    Tip # 2: Communicate the Need

    Armed with your analysis, you must let them know what your intentions are. You can do this formally or informally, depending on the structure of your organization. I like informal–it’s better for engagement; however, do whatever you feel works best. Double- and triple-check your analysis; remember, you’re dealing with people who can spot a hole in your analysis a mile away.

    This should be an engagement, not a communication. Engagement implies dialog and discussion. Listen to what they have to say: their feedback and concerns. Make them understand that you understand. If they don’t voice any concerns, they’re either not listening or not internalizing the implications of the message. Continue the dialog until they stop head-nodding and start sharing.

    Tip # 3: Teach Them How

    Concision is a skill that needs to be taught. Work with your team coach, Human Resources, or an external consultant to design a program that teaches concision. The facilitator should be familiar working with analytics—they are a special breed when it comes to this type of instructional design. Analytics have always been good at whatever they try to learn; you’re asking them to learn something they won’t initially be good at. It takes finesse to navigate through this human dynamic.

    Tip # 4: Show Them How

    Modeled behavior should follow education. Once your analytics have some guidelines to ponder, they’ll want to see it modeled in exemplars. The analytic manager on a data science team should be the paragon of concise behavior. Shorten one hour working sessions to thirty minutes and eliminate status meetings altogether. When documents are created or reviewed, focus on communicating the most amount of information in the least amount of space, with the question, point, or thesis within the first few sentences.

    Tip # 5: Help Them Build

    Be encouraging and supportive, not critical or condescending. Analytics are especially sensitive to skills they can’t quickly master. Give them time to grow and they’ll eventually come around. In addition to modeling concise behavior, I suggest introducing them to a well-written newspaper like the New York Times or the Wall Street Journal. I receive regular email alerts from the Wall Street Journal. They’re usually a hundred words or less do a great job of communicating breaking news within a few seconds.

    Tip # 6: Give Them Feedback

    Give them positive feedback when concision is done right. They won’t do it right for some time, so here’s where you have to be very careful. Criticizing an analytic for rambling or producing a tome when a brief will suffice, is a natural tendency that should be avoided. Even when it’s in the spirit of improvement, highlighting any shortcomings should be done with care. In this situation, just ask them to produce a more concise version, and be specific. I once had a data scientist give me a 50-page PowerPoint of all words. My feedback was that it had a lot of great content, but I’d need a 2-page process visual to call it done.

    Tip # 7: Give Them Kudos

    When you see your analytics exhibit concise behavior, whether a brief response or a quick turnaround on a priority deliverable, make a big deal out of it. Constructive feedback should always be done in private, but exemplary behavior should be well publicized. Leaders should support analytic managers in this effort; even a handshake from a higher-up is a big deal to most people. Everyone appreciates kudos, but more importantly when analytics see their peers getting rewarded, they take notice.

    Summary

    Be concise, and coach your analytics to do the same–enough said.

    Submitted for Publication in TechRepublic’s Big Data Analytics Blog

    This is the sneak peak of my latest contribution to TechRepublic’s Big Data Analytics Blog. As editors do, when this gets published, some of the words and content may be arranged or deleted for a variety of reasons including SEO. What you’re looking at here is the uncut, unabridged, unedited version of the article that was submitted.

    Rate this blog entry:
    0

    Posted by on in Information Exploitation

    Are you noticing anything wrong with your data science team?

    I'm sure you are; it's human nature. A client recently told me that she came home one day and noticed that there was a water-filled glass sitting directly on a wood table. She asked her husband, "Where is the coaster for this glass?" Her husband responded, "That's what you noticed? I just finished cleaning the entire house!"

    I see a lot of leaders frustrated with their data science team. They've spent a lot of money so the have very high expectations. In consulting, we call that White Knight Syndrome, and I deal with it all the time. So when things don’t go as expected, they go down a very classic route of identifying gaps and solving problems. Not only is this enervating, but it's a reckless abuse of your data science team's potential. It's far better to build on the strengths of your data science team, than it is to improve on their weaknesses. Here are five things to absolutely love about your data scientists.

    They Fuel An Uncatchable Competitive Advantage

    Your data science team is a key ingredient for a breakthrough competitive advantage. This is no joke; so don't ever overlook this fact. They tackle unsolvable problems for fun, in a way no other profession can. Most people take for granted how the data scientists at Google have changed the world, with a search engine that was late to the party. Sure, the leaders had the vision that powerful search capabilities would equate to market domination; however, it was the data scientists that figured out to jump into our brains, figure out what we were trying to find, and bring back the most relevant results. Google's data scientists made it one of the most powerful organizations in the world.

    They're A+ Students In School and Life

    Data scientists learn fast and retain extremely well. They've done it their whole lives. Most data scientists you encounter excelled in school—4.0 GPA in high school and college. And although you would expect them to get good grades in computer science and math, remember that a computer science degree has more than just computer science classes. Data scientists don't only get good grades in math and science; they get good grades in everything. Don't be shy about bringing them into your business world. They'll start contributing real value faster than you realize.

    They Deliver No Matter What

    Data scientists are extremely loyal under the right conditions--sometimes to a fault. I can't count the number of times I've been roped into an all-nighter because of situations far out of my control. We dig in and we deliver anyway; it's part of that excellence gene that I referenced earlier. The only thing you need to do is setup the right conditions, which has more to do with job satisfaction than money (although a good paycheck doesn't hurt either). Data scientists love to create data masterpieces with people they enjoy. With the right environment and the right challenge, they'll stay with you all the way.

    They Are A Magnet For Other Talent

    It seems like everybody's having a hard time finding good data scientists, except for other data scientists. If you're a leader, you probably know a lot of other leaders; so, guess who data scientists hang out with? You guessed it--other data scientists. This is important to you on a number of levels. If you ever need to extend your team, the best source for finding more data scientists is the team you already have. Also, the data scientist community is very supportive. So if your team actually gets stuck on a problem, there's a huge brain trust at their disposal that's ready and willing the help.

    They Save Your From Yourself

    Data scientists think through everything before making a decision. This will and should drive you crazy if you're an impulsive leader. Impulse is good for immediate action, but like all things the best results come from Aristotle's golden mean--the desirable middle between two extremes. At one extreme is a knee-jerk reaction that gets you into trouble (sound familiar?) and at the other extreme is analysis paralysis. The trick is to get the right balance, and you won't do that without the counsel and reason of your data scientists. You may think you have a good idea, but it won't sit right with your data scientists until there's data, research, and analysis. This voice will save your assets more times than not.

    Summary

    Identifying problems and closing gaps with your data science team will only bring you status quo; however, identifying strengths and raising the bar will catapult you to a place nobody can catch. Instead of obsessing about what's wrong; invigorate your organization by using the strengths within your data science team. There's a lot to love: they're extremely bright, loyal, and precise. Make this your starting point and enjoy your immaculate house, instead of worrying about a missing coaster.

    Submitted for Publication in TechRepublic’s Big Data Analytics Blog

    This is the sneak peak of my latest contribution to TechRepublic’s Big Data Analytics Blog. As editors do, when this gets published, some of the words and content may be arranged or deleted for a variety of reasons including SEO. What you’re looking at here is the uncut, unabridged, unedited version of the article that was submitted.

    Rate this blog entry:
    0