https://hbr.org/2010/07/the-execution-trap

“A mediocre strategy well executed is better than a great strategy poorly executed.”

Balanced scorecard can help senior managers systematically link current actions with tomorrow’s goals, focusing on that place where, in the words of the authors, “the rubber meets the sky".

Core to data science is the ability to connect a problem, or market, to data. Even today, many practicing data scientist are required to be "full stack": problem definition and research design; statistics and applied mathematics training; software engineering experience; and, product articulation and business impact. Having one person, or even a small team, being responsible for all of these operational components across a whole organization is unsustainable. As such, these roles need to begin being broken down into constituent parts within organizations.

Best advice for someone to wants to go into data science?

The most successful data scientists I know are those driven by a curiosity for solving problems through data. To that end, I recommend starting by thinking about what kinds of problems they find most interesting. From there, go out and try to find some data in the problem area and begin experimenting. Run some basic summary statistics on the data, plot some basic graphs, or even build a simple model. This is also a great way to learn the relevant tools and methods, and the result of which can become a nice asset in a person's portfolio. This also, typically, leads people to ask a lot of new questions, which creates a virtuous feedback loop of learning and community discovery.

On average, a 1 percent price increase translates into an 8.7 percent increase in operating profits (assuming no loss of volume, of course).

customer analytics dominate big data use in sales and marketing departments, supporting the four key strategies of increasing customer acquisition, reducing customer churn, increasing revenue per customer and improving existing products.

Big Data Use Cases:

48% Customer Analytics

21% Operational Analytics

12% Fraud & Compliance

10% New Product & Service Compliance

10% Enterprise Data Warehouse Optimization

18 Best Analytics Tools Every Business Manager Should Know

Business experiments: Business experiments, experimental design and AB testing are all techniques for testing the validity of something – be that a strategic hypothesis, new product packaging or a marketing approach. It is basically about trying something in one part of the organization and then comparing it with another where the changes were not made (used as a control group). It’s useful if you have two or more options to decide between.

Visual analytics: Data can be analyzed in different ways and the simplest way is to create a visual or graph and look at it to spot patterns. This is an integrated approach that combines data analysis with data visualization and human interaction. It is especially useful when you are trying to make sense of a huge volume of data.

Correlation analysis: This is a statistical technique that allows you to determine whether there is a relationship between two separate variables and how strong that relationship may be. It is most useful when you ‘know’ or suspect that there is a relationship between two variables and you would like to test your assumption.

Regression analysis: Regression analysis is a statistical tool for investigating the relationship between variables; for example, is there a causal relationship between price and product demand? Use it if you believe that one variable is affecting another and you want to establish whether your hypothesis is true.

Scenario analysis: Scenario analysis, also known as horizon analysis or total return analysis, is an analytic process that allows you to analyze a variety of possible future events or scenarios by considering alternative possible outcomes. Use it when you are unsure which decision to take or which course of action to pursue.

Forecasting/time series analysis: Time series data is data that is collected at uniformly spaced intervals. Time series analysis explores this data to extract meaningful statistics or data characteristics. Use it when you want to assess changes over time or predict future events based on what has happened in the past.

Data mining: This is an analytic process designed to explore data, usually very large business-related data sets – also known as ‘big data’ – looking for commercially relevant insights, patterns or relationships between variables that can improve performance. It is therefore useful when you have large data sets that you need to extract insights from.

Text analytics: Also known as text mining, text analytics is a process of extracting value from large quantities of unstructured text data. You can use it in a number of ways, including information retrieval, pattern recognition, tagging and annotation, information extraction, sentiment assessment and predictive analytics.

Sentiment analysis: Sentiment analysis, also known as opinion mining, seeks to extract subjective opinion or sentiment from text, video or audio data. The basic aim is to determine the attitude of an individual or group regarding a particular topic or overall context. Use it when you want to understand stakeholder opinion.

Image analytics: Image analytics is the process of extracting information, meaning and insights from images such as photographs, medical images or graphics. As a process it relies heavily on pattern recognition, digital geometry and signal processing. Image analytics can be used in a number of ways, such as facial recognition for security purposes.

Video analytics: Video analytics is the process of extracting information, meaning and insights from video footage. It includes everything that image analytics can do plus it can also measure and track behavior. You could use it if you wanted to know more about who is visiting your store or premises and what they are doing when they get there.

Voice analytics: Voice analytics, also known as speech analytics, is the process of extracting information from audio recordings of conversations. This form of analytics can analyze the topics or actual words and phrases being used, as well as the emotional content of the conversation. You could use voice analytics in a call center to help identify recurring customer complaints or technical issues.

Monte Carlo Simulation: The Monte Carlo Simulation is a mathematical problem-solving and risk-assessment technique that approximates the probability of certain outcomes, and the risk of certain outcomes, using computerized simulations of random variables. It is useful if you want to better understand the implications and ramifications of a particular course of action or decision.

Linear programming: Also known as linear optimization, this is a method of identifying the best outcome based on a set of constraints using a linear mathematical model. It allows you to solve problems involving minimizing and maximizing conditions, such as how to maximize profit while minimizing costs. It’s useful if you have a number of constraints such as time, raw materials, etc. and you wanted to know the best combination or where to direct your resources for maximum profit.

Cohort analysis: This is a subset of behavioral analytics, which allows you to study the behavior of a group over time. It is especially useful if you want to know more about the behavior of a group of stakeholders, such as customers or employees.

Factor analysis: This is the collective name given to a group of statistical techniques that are used primarily for data reduction and structure detection. It can reduce the number of variables within data to help make it more useful. Use it if you need to analyze and understand more about the interrelationships among a large number of variables.

Neural network analysis: A neural network is a computer program modeled on the human brain, which can process a huge amount of information and identify patterns in a similar way that we do. Neural network analysis is therefore the process of analyzing the mathematical modeling that makes up a neural network. This technique is particularly useful if you have a large amount of data.

Meta analytics/literature analysis: Meta analysis is the term that describes the synthesis of previous studies in an area in the hope of identifying patterns, trends or interesting relationships among the pre-existing literature and study results. Essentially, it is the study of previous studies. It is useful whenever you want to obtain relevant insights without conducting any studies yourself.

In Resume Building for data science - Communication and prioritization is important so that interviewer is aware that we are knowing of this fact

6 key analytics skills used by successful analyst/data scientist are:

DTD framework: Understanding and hands-on experience of the basic “Data to Decisions” framework
SQL skills: Ability to pull data from multiple sources and collate: experience in writing SQL queries and exposure to tools like Teradata, Oracle etc. Some understanding of Big Data tools using Hadoop is also helpful.
Basic “applied” stat techniques a.k.a. Business Analytics:Hands-on experience with basic statistical techniques: Profiling, Correlation analysis, Trend analysis, Sizing/Estimation, Segmentation (RFM, product migration etc.). If you are in a consumer business, this list would include hands-on comfort in A/B Testing (also called Design of Experiments)
Working effectively with business side: Ability to work effectively with stakeholders by building alignment, effective communication and influencing
Advanced “applied” stat techniques (hands-on) a.k.a. Predictive Analytics and machine learning: Hands-on comfort with advanced techniques: Time Series, Predictive Analytics – Regression and Decision Tree, Segmentation (K-means clustering), machine learning - neural networks and Text Analytics (optional)
Stat Tools: Experience with one or more statistical tools like Python, R, SAS, SPSS, Knime or others.

4 key data science skills needed by business professionals are:

DTD framework: Understanding and hands-on experience of the basic “Data to Decisions” framework.
Basic “applied” stat techniques: Hands-on experience with business analytics: Profiling, Correlation analysis, Trend analysis, Sizing/Estimation, Basic Segmentation and basics of A/B Testing (also called Design of Experiments)
Working effectively with analysts: Ability to work effectively with Data Scientists/Analyst
Advanced “applied” stat techniques (intro): High-level understanding of predictive analytics: Time Series, Predictive Analytics – Regression and Decision Tree, Segmentation, Neural Networks.

Key Attribute of an Actionable Insight:

1. Alignment

When an insight is closely tied to your key business goals and strategic initiatives, it’s more likely to drive action. If you don’t know how to react to a particular metric when it significantly increases or decreases, you might be looking at an unnecessary vanity metric. Insights based on key performance indicators (KPIs) and other key metrics inherently engender a sense of urgency that other data won’t. It’s easier to interpret and convert strategically-aligned insights into tactical responses because they often relate directly to the levers in your business that you control, influence or are focused on.

2. Context

It is hard to move forward on an insight if you are lacking ample background to appreciate why it’s important or unique. We often need to have a comparison or benchmark to give data proper context. For example, if your company generated 1,400 leads this week your reaction to this result could change entirely with a dash of context. If you know your marketing team typically generates 1,250 leads each week you might do nothing. However, if your company just sponsored and exhibited at your industry’s major convention last week, you may be wondering why it didn’t translate into significantly more leads. Without accompanying context, an insight can end up raising more questions than action. Having ample supporting details ensures the insight results in action and not unwarranted skepticism and objections.

3. Relevance

A single insight can be both a strong signal for one person and just more noise for another. There’s a level of subjectivity when it comes to the relevance of insights. In order to be relevant, an insight needs to be delivered to the right person at the right time in the right setting. If insights aren’t routed to the right decision makers, they will not receive the attention they deserve. If insights aren’t timely, they might be too stale for stakeholders to act on. If insights are trapped in an analytics tool that managers never access or delivered to devices they use infrequently, the insights may never reach the intended audience.

4. Specificity

The more specific and complete the insight is, the more likely it can be acted on. Sometimes insights based on KPIs and other high-level metrics can highlight interesting anomalies but lack sufficient detail to drive immediate action. For example, knowing your revenue is up 35% this month may be a cause to celebrate, but party planning might be premature without deeper insights. You might discover a massive fraudulent order messed up your online revenue numbers or a big customer win was based on promised product functionality that won’t exist for the foreseeable future. If an insight doesn’t adequately help to explain why something occurred, it’s not yet actionable. Deeper probing may be required before it’s ready for primetime.

5. Novelty

With so much competing data and information to digest, novel insights will have an advantage over more familiar insights. The first time your company spots a particular pattern will be more interesting and compelling than the tenth time, especially if you feel you already have a good handle on what’s driving the behavior. Curiosity can drive people to test or verify an unusual or unexpected finding in the hope that it sheds new light on a key subject area. This criterion speaks more to human nature than to how valuable an insight actually is. We become numb to certain insights if we feel as though they reinforce rather than challenge or evolve our current knowledge and beliefs.

6. Clarity

If people don’t clearly understand an insight, why it’s important and how it can help them—the insight will be overlooked and forgotten. Communicating insights effectively is important to their adoption and fruition. The right data visualizations and messaging can help explain insights so they are more easily understood and correctly interpreted. However, poor communication can cause the signal to be lost in the noise. A clearly communicated insight creates a strong signal that is hard to miss or ignore, and it prepares a pathway for action to occur.

With these six criteria you can weigh how “actionable” the insights are that you receive from your analytics and business intelligence tools. Regardless of the source of the insights—humans or machines—the more they line up with these attributes, the more actionable they will be for your business. Strategically-aligned tops random; relevant beats extraneous; novel trumps familiar—you get the idea.

Harvesting more insights from your data can yield tremendous returns for your company. However, as author Richard Bach said, “Any powerful idea is absolutely fascinating and absolutely useless until we choose to use it.” While the increased actionability of an insight doesn’t guarantee its adoption or application, it should motivate more individuals within your company to think more deeply about the data and encourage them to act on a more consistent basis. Make sure your hard-earned insights are as actionable as possible so they’re primed to drive value for your organization.

https://www.forbes.com/sites/louiscolumbus/2018/06/08/the-state-of-business-intelligence-2018/#7f7650297828

https://www.forbes.com/sites/metabrown/2016/10/31/5-big-data-and-analytics-learning-resources-that-most-people-miss-but-shouldnt/#7ec64af7ecd5

https://www.forbes.com/sites/metabrown/2016/05/26/choosing-data-analytics-training-thats-worth-the-investment/#cc671a555ba9

http://www.aprapros.com/#

https://www.smartdatacollective.com/

https://www.smartdatacollective.com/why-data-analytics-insurance-industry-is-major-game-changer/

https://www.smartdatacollective.com/5-brands-using-big-data-to-brilliantly-disrupt-tech/

https://gotrg.com/

https://junkcharts.typepad.com/numbersruleyourworld/

https://fivethirtyeight.com/

http://www.dataminingblog.com/

https://junkcharts.typepad.com/junk_charts/