Personalize Relevant Insights through Reliable Big Data
Data volumes show no sign of abating. Although storage and processing is now affordable by companies of all sizes, turning data into actionable information is an increasing challenge. Not only is the volume making finding the relevant information requiring the assistance of highly skilled, very scarce “data scientists”, it makes things even worse when the data is not reliable and you are essentially gambling with your insights. Once the data is made reliable and frontline business users can get relevant insight, the next step is to get help by providing guidance through personalized recommended actions.
Taking a Hit
Take a small but smart data example.
Blackjack is a game that many people believe offers the best chances to win in a casino, because it is a game where the odds of any particular outcome can be determined accurately if you have the right data. The best players in Blackjack do not make winning any particular hand a priority, but strictly make the best moves based on the probable outcomes, and the data they have through card counting.
“To get a 360 view of people, products, organizations, places and their interactions, graph technology must be used”
If you played "basic strategy" in blackjack, you would be executing based on odds and probable outcomes, not gut feel. If you are familiar with the rules of blackjack you'll likely make the correct decision in this following scenario:
Let's say you are one-on-one against the dealer at the table and have been dealt two cards: 10 and 6 totaling 16.
Let’s say the dealer’s card showing is a Jack, value of 10. Do you “Hit”, (Blackjack term for take another card) or do you stand? (Stop and take your chances and hopes the dealer goes over 21)
The odds favor you taking a card.
Why? Because there are 5 cards that can improve your hand. Ace, 2, 3, 4, 5 (giving you 17, 18, 19, 20, 21 respectively). Only one card, a 6 can really hurt you. The cards 7, 8, 9, 10, J, Q, K are irrelevant because if you take them you bust (go over 21), but if the dealer gets them he/she gets 17 or better and you lose anyway because you only have 16.
So the odds are 5 cards to 1 in your favor. Even though the dealer's advantage is that you are going first, you must hit 16 against a dealer 10 or face card, otherwise you are basically giving your money away to the casino.
Under what circumstances might you decide to go against these odds? If you happen to be a card counter and you've kept track of all the cards that have been dealt, and know exactly what's left, allowing you to "predict" at a higher-level of certainty to go against the odds and conventional wisdom of the hand.
Ironically, this simplified blackjack strategy highlights how you might be gambling with your data.
Firstly, if you don't have reliable data, you don't really have a good handle on exactly what cards have gone before, or are still remaining in the deck. So you can't be sure any decision you make will yield your preferred outcome.
Second, you can collect as much data as you want, such as the history of the dealer, the skill of your fellow players, the way the table has been paying out, and even extrapolate that across all the tables in the casino. But that doesn't offer the relevant insight at the specific moment in time you need to beat the dealer.
Why does the casino offer Blackjack as a game, if it's common knowledge that if you play according to the odds you stand a chance of winning? Well, apart from the fact that it still retains the edge because the dealer gets to go last, and there are other esoteric rules that tilt the odds back in the house's favor, they know that most people don't have the knowledge, or the recommended action to take based on the situation.
Just like playing blackjack without reliable data, your business decisions are a gamble, regardless of the analytics and visualization tools you've invested in. Most applications and data warehouses end up operating on inaccurate data. BI and analytic tools still require you to sift through mounds of irrelevant data to find what you need. And process-driven apps like CRM don't encapsulate the best practices or offer recommended actions of what to do given even the most common scenarios.
Reliable Data at Big Data Scale
What does this mean for enterprise IT where the volume, variety, and velocity of data is exponentially higher? The complexity of making data reliable in a fluid environment seems a million miles away from a simple blackjack example, but the premise is still the same. Data needs to be clean, relationships discovered, and personalized recommended actions need to be delivered, in context with a user’s role and goal in real-time.
To get a 360 view of people, products, organizations, places and their interactions, graph technology must be used. Like graphs that power the most popular consumer data-driven applications such as LinkedIn (Economic Graph), Google (Knowledge Graph) and Facebook (Social Graph), enterprise-class graph technology powers a new generation of data-driven applications for the enterprise.
The twist is that most off the shelf graph databases are not designed for big data scale. Fortunately a new breed of data management platforms are combining technologies such as NoSQL horizontally scalable databases such as Apache Cassandra, with graph technology to deliver the best of both worlds. These technologies also seamlessly synchronize and integrate with analytic platforms such as Apache Spark, solving the dilemma of keeping reference data and analytic environments synchronized. Often the biggest barrier to companies being able to obtain the insights they need in a flexible and agile manner.
Sales and marketing teams armed with a 360 view of their customers and products can then use the power of reliable, accurate data, combined with relevant insights to improve engagement and customer service.
Personalization is one of the fastest growing techniques in B2B to capture the attention of customers who are very much used to that level of service in their daily consumer interactions. The dangers of B2B personalization of course are that if the data is incorrect, the personalized message falls flat, and worse may cause a negative perception or reaction. The complete opposite of the goal in the program or campaign.
Take the recent popularity of personalized video. With a reliable data foundation, the attributes related to an individual can be confidently delivered in context and personalized.
Making it Less of a Gamble
Companies need to stack the cards in their favor by leveraging modern data management technologies. Starting with a reliable data foundation business teams can ultimately benefit from recommended actions that in turn will allow them to confidently leverage the information for personalized customer engagement.
Getting the Most out of Big Data
Big Data: Separating the Hype from Reality in Corporate Culture
Maintaining Maximum Relevancy for Buyers and Sellers
Building Levies to Manage Data Flood
By Chris Tjotjos, VP, Cisco Solutions Practice, Black Box...
By Laura Jackson, Sr. Manager-Risk Management, ABS Consulting
By Jason Cradit, VP of Information Systems, Willbros Group
By Steve Garske, Ph.D., Senior Vice President & Chief...
By Roman Trakhtenberg, CEO, Luxoft
By Renee P Wynn, CIO, NASA
By Mike Morris, CIO, Legends
By Louis Carr, Jr., CIO, Clark County
By Andrew Macaulay, CTO, Topgolf Entertainment Group
By Dominic Casserley, President and Deputy CEO, Willis...
By Dave Nelson, SVP-Portfolio Lead, Avanade, Inc.
By Michael Cross, SVP & CIO, CommScope Holding Company Inc.
By Pauly Comtois, VP DevOps, Hearst Business Media
By Dan Adam, CIO, Extreme Networks
By Matt Schlabig, CIO, Worthington Industries
By David Tamayo, CIO, DCS Corporation
By Scott Cardenas, CIO, City and County of Denver
By Marc Kermisch, VP & CIO, Red Wing Shoe Co.
By Brian Drozdowicz, VP, Digital Services, Siemens...
By Les Ottolenghi, EVP and CIO, Caesars Entertainment