There’s not a day that goes by where I don’t read about the CIO of some billion dollar company claiming that Big Data has saved his company untold sums of money. A one percent efficiency increase identified by Big Data resulted in some few million in savings. Segmentation of transactional sales mapped to a unified customer database revealed a 10% increase in alternate channel sales of related products. Well, at least we know they’re not liars; a recent Bain and Company study found that adopters of Big Data analytics have gained a significant lead over the rest of the corporate world. Companies with the most advanced analytics capabilities outperform competitors by wide margins. The leaders are:
- Twice as likely to be in the top quartile of financial performance within their industries
- Five times as likely to make decisions much faster than market peers
- Three times as likely to execute decisions as intended
- Twice as likely to use data very frequently when making decisions
However, this doesn’t mean that you should just jump into the deep side of the pool. For every story about accelerated financial performance, I can point to ten that talk about mismanaged investments and a loss of interest by leadership in Big Data. In my last blog, I talked about some of the things that Big Data isn’t. As promised, here is a follow-up blog to help you know when the time is right to begin your platform, services, and organizational investment into a multi-year, multi-disciplinary practice that will change the way your organizations makes decisions.
1) You have some degree of mastery over business analytics.
Companies looking to exploit the promise of big data have a proving grounds that must first be mastered; business intelligence. After all, if you don’t know what to do with the data you already have, chances are low that you will know how to proceed with mapping disparate data streams to find answers to questions you have yet to ask. Mastery of your own data has two principles values; one, it gives your organization the expertise it needs to take analytics to the next level, and two, it suggests that you have driven some (hopefully most) of the financial benefit out of the data you already have.
2) You are collecting streams of data.
Without some sort of historical perspective, it will be difficult to baseline potential efficiencies. Even with real-time operations, historical data sets can be mined for co-related data that can indicate operational anomalies or financial opportunities. Let’s imagine the following scenario:
As a part of the drilling and completions engineering team, you are looking to prevent differential sticking and possible twist-offs at a drill site. There are multiple vectors that needs to analyzed; the change in differential pressure as a function of time, fluid loss to formation as a function of the geology of the reservoir, the lubricity of the mud cake as a function of friction, etc. Operational components need to be kept in mind; the type of mud being used, collar shape, depth of drilling, and so on.
You decide to use Big Data to analyze what has happened in the past. After ingesting the data from ten thousand twist-offs into Hadoop, you find that torque as a function of time has certain upper and lower boundaries that are suggestive of a twist off 30% of the time. This increases to 60% when certain geological formations are present. Maybe it’s 80% when you include temperature, pressure, and mud loss characteristics. And maybe it’s even higher if you include the drill type, the manufacturer of the pipe, the incidence of earthquakes in a region, the age of the formation you are drilling into, the number of hours the crew has worked without a break, and the name of the drilling supervisor.
The data you are collecting in this scenario is manifold; SCADA devices are sending WITSML streams of data that give you down-hole information. Asset management systems are giving you information about the type and characteristics of the machinery used. Personnel records are giving you information about HSE and past performance of your crew. GIS data is coming in through multiple public domain data sources.
Each one of those streams of data is valuable, but finding where all of these streams intersect… well, that’s Big Data. But understanding where that data intersects cannot be properly identified without an initial analysis of historical data; essentially, understanding how something happens when it happens, and modelling that behavior for application to future use cases.
3) Your culture can embrace opportunistic analytics.
Big data does not have the same value proposition as other investments. It is not, as a rule, requirements based. There are no clearly-defined business problems that require resolution. You can’t build a standard ROI or NPV model; you don’t know what the expected return is going to be or what sort of organizational gymnastics you will have to do in order to support it. Big data is fundamentally opportunistic, best suited to solve problems that are ill-defined or not at all. Your business analysts are going to have apoplectic fits and your finance team is going to hit you with the general ledger. Be ready.
4) You have the nerd power.
Playing with Hadoop, NoSQL, and Splunk sounds like something your first grader might be doing with play-doh and an unfortunate sense of humor, but these toys are for big kids. These cutting edge tools have reinvented how data is assembled, analyzed, and correlated. However, nerds with platform expertise is insufficient; for a big data strategy to work, you must also have business analysts that are domain experts with intimate knowledge of the operations of whatever it is that your company does. Without this domain expertise, understanding the inter-relationships between disparate pieces of data will be impossible. Another challenge is finding the data scientist; these people are tasked with understanding how to drive value out of data using creativity, intuition, and the types of higher-order calculus that you definitely have no interest in understanding.
Big Data has a lot to offer, but I have seen many leaders invest money, resources, and reputations in Big Data strategies, only to fall on their face when it became clear that the fundamentals were missing. Maybe corporate data wasn’t being managed properly. Perhaps they lacked the in-house expertise to aggregate or crunch the data. Even worse were the organizations that expected results immediately – organizations that believed that Big Data could solve specific use cases or bring efficiencies into their organization, unaware that they probably already had the tools to solve such use cases using traditional analytics. You’ve probably heard of the proverb “learn to crawl, then walk, and then you can run.” Moving into Big Data without having a grasp on these four principles is like participating in a marathon when you’ve just learned how to scoot across a carpet. It’s just not going to happen.
Jamal is a regular commentator on the Big Data industry. He is an executive and entrepreneur with over 15 years of experience driving strategy for Fortune 500 companies. In addition to technology strategy, his concentrations include digital oil fields, the geo-mechanics of multilateral drilling, well-site operations and completions, integrated workflows, reservoir stimulation, and extraction techniques. He has held leadership positions in Technology, Sales and Marketing, R&D, and M&A in some of the largest corporations in the world. He is currently a senior manager at Wipro where he focuses on emerging technologies.
Interested in more content like this? Sign up to our newsletter, and you wont miss a thing!
[mc4wp_form]
