I’m pretty sure that you’ll are well aware of the buzz term ‘Big Data’ but let’s keep that matter to be discussed to another day. Let’s just concentrate on a basic matter in analytics, that’s Data Analytics. So here are 5 matters with regards to Data Analytics that a CEO needs to get her/himself familiar with!
We at Zepto help companies extract real business value from analytics that takes advantage of data. Whether you love data or see it as problematic, clearly it is too big to ignore. But how valuable is all this data and what does it mean to an organization’s success or survival? As a business that uses data for analytics projects internally, and also advises clients about how to use it, we believe it’s a potential oil rig.
As a CEO, you’ll have a central role to play in forming the conversation and guiding your company’s data analytics efforts. In the process, you may face misperceptions, cost restraints, and even competing for business agendas, not to mention C-suite colleagues whose ability to influence analytics decisions may surpass their subject matter expertise.
As CIO or CTO, take ownership of data analytics. They become the trusted advisor in an organization. “Tell” and “sell” the story of analytics broadly to the leadership team. In the process, helping the CEO and organization as a whole to avoid missteps. With our exposure and experience, we’ve distilled our own learning about data analytics down to five tips. They may seem obvious to you but, believe us when we say this, they may not be obvious to your C-suite colleagues.
1. Data has little intrinsic value
Data is not inherently organized or well-generated. It’s certainly not neatly structured in a database. It’s messy. It’s everywhere. It’s a morass of details of what we buy, where we drive, what we surf online, what we “like”. It’s transactions from financial markets and e-commerce sites, chats on social networks, cellphone conversations, tweets, texts, photos, video, your web search, and browsing patterns. ( we recommend you reading this article to get the whole picture) We human generate data every minute, without even us realizing it as we use a search engine, call up GPS or swipe our badges at work.
Data analytics is already changing our world in unprecedented ways, powering new products, business models, and scientific breakthroughs. But the data alone has no intrinsic value. None of the value that results from Data initiatives is created in a vacuum. You need to make sense of it and to find the relevance in it. That’s where the hard work comes in.
2. First, do no harm
Data makes sense, except when it doesn’t. More volume and variety are not necessarily better. If not pursued strategically, the data analytics can lead to missed opportunities, expensive distractions, even serious strategic errors.
Each organization needs to define a value for itself and its stakeholders. It may translate to increased competitive advantage, more efficient operations, or reduced risk. The litmus test for data analytics is whether it creates value, however, it varies from business to business as their own environment defines it.
Companies that effectively employ data analytics—in concert with other postdigital forces such as social business and mobile—are being rewarded by Wall Street. In a recent study, these companies were 26 percent more profitable than their industry competitors, generated 9 percent more revenue through their employees and physical assets, and enjoyed 12 percent higher market valuation ratios.¹
Analyzing big data can also help companies reduce their risk exposure. Waste, fraud, bribery, abuse—these are significant common problems for many organizations, especially large, global ones. With Zepto, AI, and machine learning, we help organizations detect issues in real time—or prevent them altogether. Limiting these exposures can translate to more profitable operations. A credit card company that could reduce fraud, waste, and abuse by even one percent of charged volume could significantly improve its bottom line.
Similarly, we help businesses to analyze data to help banks reduce their exposure to money laundering in ways required for regulatory compliance. Our bank clients are setting up systems to detect suspicious transactions. And we’re helping federal agencies apply deep analytics to move from “pay and chase” to prevention; in other words, anticipating and mitigating risk related to fraud, waste, and abuse rather than simply working to recover lost dollars. In an entity already burdened by high deficits and tight budgets, that translates to substantial value.
3. Data needs math
Until recently, analytics consisted of gathering existing data and using it to tell a story of what happened—to make sense of past events.
At present we can comb through large data stores—historical data, sure, but also current data and data from external sources—to model the future for planning, forecasting, and business decision support. This so-called predictive analytics is a “big math” endeavor.
For example, companies often struggle to achieve anticipated outcomes from large, transformative projects. These projects typically involve sizable financial, resource, and time commitments—and myriad data points along their life cycle. They also often suffer from overruns and delays; some even get totally scratched before delivery. There’s a real need to anticipate both roadblocks and opportunities to accelerate the process. Predictive analytics, along with traditional project management measures, can help organizations assess the likelihood of project success, mitigate risks, and improve their overall success rates. Big data analytics illuminate signals that inform decisions.
Importantly, this analytics can’t be tackled simply by purchasing hardware and software. Processing power thrown against a large, messy dataset will almost never produce a meaningful picture. Math requires specialized, analytical talent—data scientists with creativity and judgment, along with deep statistical and computer science knowledge. Building bench strength in this area should be on every CIO’s talent agenda.
4. Little data is fine for beginners
Little data refers to the data you own. Three-quarters of corporate executives in large companies say they are getting value from less than half the data they already own, but that hasn’t stopped many of them from moving ahead with attempts to mine large social media databases and other external sources of unstructured content.
Make sure your business is taking advantage of its available data before spending to acquire more. Get to know your data close up, and master the process of finding insights within it that support your organization’s mission and goals. The information you already own may be all that’s needed. That’s an important message for your CEO to hear.
Here’s another: You needn’t boil the ocean to get to the answer. Sometimes small data sets can readily reveal insights or trends that might be diluted or go undetected in huge datasets, especially if you’re unequipped to apply specialized analytic techniques to the data. For example, a fraud pattern may leap out at an analyst looking at a single week of transactions but remain hidden in 10 years of transaction data. Sometimes less is more.
Additionally, data analytics don’t need to start with an expensive, enterprise initiative to perfect your data. Select a handful of problems that matter to a CEO, board, and other decision-makers, and begin to solve them. Fiercely protect the scope of these early efforts. Operate on a small scale. Run pilots to demonstrate value and deliver early wins.
5. Data is not a strategy
Put your strategic objectives front and center; otherwise, your data analytics efforts will be interesting, probably expensive, intellectual exercises with limited or no strategic value. What problem are you trying to solve? What opportunities are you trying to discern or optimize?
Meanwhile, recognize that while direction and purpose are essential, so too is keeping an open mind. A narrow search for precise, predefined outcomes may cause you to miss the big Aha! Data has hidden in its top- and bottom-line results, but it’s folly to precise plan for those results—or worse, justify in advance investments based on hoped-for outcomes. Go where the data leads you.