Big Data & Data Experts, are often making headlines as the fresh angle of business competitive advantage.
This should not come as a surprise. In an Harvard Business Review article back in 2014, it was documented through a widespread study that, high performing organizations seem to share specific common traits. In particular,
- Purposeful Positioning.
- Enhancing the value of products and services by creating Customer Experiences.
- Use of Big Data and leveraging of insights stemming from it.
As we know, companies are overflowing with data. It is important to define Big Data and Analytics to avoid misinterpretations:
- Big Data is a term that describes the large volume of data –both structured and unstructured– that swamps an organization on a day-to-day basis.
- Analytics refers to the analysis & insights extracted from this Big Data termed as Big Data Analytics or just Analytics. Analytics rely on processes based on statistics, computer programming and operations research.
There is a multitude of cases on the payoff of Big Data Analytics in various sectors e.g. Financial institutions use it to detect fraud, Police to predict and detect criminal activity, big Cities to optimize traffic control, etc.
The idea is to shift through data to arrive at meaningful insights that can help one to make more accurate predictions and more informed decisions. In doing so, companies get an edge over their rivals.
First things first. Using data to its full potential is more about management than it is about technology.
Putting data to work requires structure, discipline and governance in the continuity from data, to insight, to profit. Hence, there needs to be a strategy i.e. a Data Strategy. A solid framework to support decision making. This responsibility falls squarely on a Company C-suite executive and the CEO.
It should be recognized that Data comes in two forms structured and unstructured. This has important implications in the decision-making process.
Internal company data is normally structured and typically consists of numbers. In contrast, external data is unstructured and typically comprises text in various styles and formats, videos, etc.
Decision making in our times should not rely solely on internal KPIs e.g. a Company’s financials, quarterly reviews, annual plans. These structured data sets revolve around historical events which have earlier taken place in a Company’s life. They help to understand a Company’s past but are increasingly a poor indicator to its future. This data analysis which is based on traditional Business Intelligence (BI) models and subsequent decision making, is rather retrospective and reactive.
Modern decision-making i.e. Data Science, shifts attention to external KPIs, relating to unstructured data sets associated, for instance, with competitors, clients, suppliers and other players within the ecosystem a Company operates and sometimes outside of it as well. This data analysis based on Data Science and subsequent decision making based on it, is both proactive and predictive.
In our times, to survive and pull ahead of the pack, an organization needs to shift from the traditional BI to Data Science. For this, a new data strategy is required.
Towards a Data Strategy
Information is data endowed with relevance and purpose. Converting data into information thus requires knowledge. And knowledge, by definition, is specialized. – Peter Drucker.
As conveyed by Leandro DalleMule & Thomas H. Davenport, there are two aspects to a Company’s Data Strategy I.e. Data Architecture and Information Architecture.
Data Architecture, is a set of rules, policies, standards and models that govern and define the type of data collected and how it is used, stored, managed and integrated within an organization and its database systems.
Should be the sole repository of data within a Company, flexible enough to support the needs of every function and department.
No other sources of data should exist.
Information Architecture, guides the processes and rules that convert data into useful information, which in turn, is used as input to decision making.
For instance, Data Architecture feeds raw Advertising share of voice & share of spending, as well as, Sales figures into an Information Architecture system such as a Marketing dashboard. This is subsequently analyzed e.g. through graphs and tables, to uncover relationships between advertising spend and sales, by channel and geography.
Data Science and Data Experts
Data Science is an interdisciplinary field comprising statistical and mathematical techniques, involving scientific processes, algorithms and systems.
Data Scientists/Experts are those who use their abilities to analyze large, noisy and complex data sets, recognizing patterns which potentially lead to valuable insights. The rest of a Company’s organization, should in principle, pick up on these insights and turn them into opportunities e.g. products, processes and services; leading ultimately to growth and profit.
For instance, analyzing a competitor’s patent filings or the profile of those it hires, one can accurately predict the areas a competitor is interested in and potentially a threat soon. This way the forward looking Company i.e. one that its decision making process is informed by external KPIs, may have the time to act proactively vs reactively at a later date.
There are some important challenges worth mentioning working with Analytics:
- Very large data sets usually contain noise e.g. corrupt or incomplete data, leading to pseudo-correlations i.e. spurious correlations. To have actionable insights we need causal relationships.
Correlation does not imply Causation. - The Data Expert owes to have the curiosity to leave its desk and validate with business acumen the results of his/her theories and hunches, in real life e.g. investigate with decision makers, wander in the factory floor, do market visits pretending to be a customer, etc.
Data should never replace intuition. - The Analytics function within a Company should be centrally managed, with its leadership reflected in the C-suite. The Data experts, within the Analytics function, may belong to different departments e.g. Marketing, Supply Chain, Development etc. A sound Data & Information Architecture, as described above, will make sure ease in data share, based on consistent formats definitions and standards, as well as, use of commonly approved tools converting data into useful departmental information and insights.
- It may be surprising but as of now there are no university programs offering a degree in Data Science even though many are fervently working in putting a curriculum together. So how do you go about recruiting them? What are the characteristics a Company should be looking for?
It should be noted from the beginning that Data Expert is not a back-office job, creating reports and presentations. Data Experts need autonomy and need a position at the helm of the organization assisting their C-suite colleagues -when they are not themselves part of the C suite- with responses, in real time. They are co-building a Company’s future, actively.
- The Data expert is a hybrid. His/her skills are relating to computer science, math and economics; while competencies relate to strategic thinking, business acumen and effective interpersonal communication. In particular, possessing the art of lucid data storytelling, to the main stakeholders, seems to be the sine qua non of his/her DNA.