Analytics and AI: Humans and Machines are Good at Different Aspects of Prediction

Mostly driven by growth in the IoT, and the widespread use of internet, social media and mobile devices to perform search, send text, email and capture images and videos, the amount of data that we are producing on a daily basis is startling.

Consequently, companies are turning to data analytics and AI technologies to help them make sense of all the data at their disposal, predict the future, and make informed decisions that drive enterprise performance.

Although adoption of analytics and AI systems is increasingly extending in more mission-critical business processes, the implications of these emerging technologies on busines strategy, management, talent and decisions is still poorly understood.

For example, the single most common question in the AI debate is: “Will adoption of AI by businesses lead to massive human job cuts?”

Borrowing lessons from historical technological advances, yes, certain jobs will be lost, and new ones also created. However, machines are not taking over the world, nor are they eliminating the need for humans in the workplace.

Jobs will still be there albeit different from the traditional roles many are accustomed to. The majority of these new roles will require a new range of education, training, and experience.

For instance, nonroutine cognitive tasks demanding high levels of flexibility, creativity, critical thinking, problem-solving, leadership, and emotional intelligence do not yet lend themselves to wholesale automation.

Analytics and AI rely on data to make predictions

As more and better data is continually fed to the machine learning algorithms, the more they learn, and improve at making predictions.

Given these applications search for patterns in data, any inaccuracies or biases in the training data will be reflected in subsequent analyses.

But how much data do you need? The variety, quality and quantity of input, training and feedback data required depends on how accurate the prediction or business outcome must be to be useful.

Training data is used to train the predictive algorithms to predict the target variable, while the feedback data is used to assess and improve the algorithm’s prediction performance.

Undoubtedly, advanced analytics and AI systems are only as good as the data they are trained on. The data used to train these learning algorithms must be free of any noise or hidden biases.

You therefore need to understand how predictive technologies learn from data to perform sophisticated tasks such as customer lifetime value modeling and profitability forecasting.

This helps guide important decisions around the scale, scope and frequency of data acquisition. It’s about striking a balance between the benefits of more data and the cost of acquiring it.

Humans and machines both have shortcomings

In the context of prediction, humans and machines both have recognizable strengths and weaknesses.

Unless we identify and differentiate which tasks humans and machines are best suited for, all analytics and AI investments will come to naught.

For instance, faced with complex information with intricate interactions between different indicators, humans perform worse than machines. Heuristics and biases often get in the way of making accurate predictions.

Instead of accounting for statistical properties and data-driven predictions, more emphasis is often placed on salient information unavailable to prediction systems.

And, most of the times, the information is deceiving, hence the poor performance.

Although machines are better than humans at analyzing huge data sets with complex interactions amidst disparate variables, it’s very crucial to be cognizant of situations where machines are substandard at predicting the future.

The key to unlocking valuable insights from predictive analytics investments involves first and foremost understanding the definite business question that the data needs to answer.

This dictates your analysis plan and the data collection approaches that you will choose. Get the business question wrong, conclusively expect the insights and recommendations from the analysis to also be wrong.

Recall, with plentiful data, machine predictions can work well.

But, in situations where there is limited data to inform future decision making, machine predictions are relatively poor.

To quote Donald Rumsfeld, former US Secretary of Defense:

There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we don’t know. But there are also unknown unknowns. There are things we don’t know we don’t know.

Donald Rumsfeld, former US Secretary of Defense

Thus, for known knowns, abundant data is readily available. Accordingly, humans trust machines to do a better than them. Even so, the level of trust changes the moment we start talking about known unknowns and unknown unknowns.

With these situations, machine predictions are relatively poor because we do not have a lot of data to ingest into the prediction model.

Think of infrequent events (known unknowns) that occur once in a while, or something that has never happened before (unknown unknowns).

At least for infrequent events or happenings, humans are occasionally better at predicting with little data.

Generally so because we are good at comparison and applying prudent judgement, examining new situations and identifying other settings that are comparable to be useful in a new setting.

We are naturally wired to remember key pieces of information from the little data available or the limited associations we have had in the past.

Rather than be precise, our prediction comes with a confidence range highlighting its lack of accuracy.

Faced with unknown unknowns, both humans and machines are relatively bad at predicting their arrival.

The simple truth is that we cannot predict truly new events from past data. Look no further than the current Brexit conundrum.

Nobody precisely knew the unintended consequences of the UK leaving the EU. Leavers and Remainers both speculated as to what the benefits and disadvantages of leaving the EU maybe.

Of course, nobody knows what will happen in the future but that doesn’t mean we can’t be prepared, even for the unknown unknowns.

In their book Prediction Machines: The Simple Economics of Artificial Intelligence, Ajay Agrawal, Joshua Gans, and Avi Goldfarb present an additional category of scenarios under which machines also fail to predict precisely – Unknown Knowns.

Per the trio:

Unknown knowns is when an association that appears to be strong in the past is the result of some unknown or unobserved factor that changes over time and makes predictions we thought we could make unreliable.

PREDICTION MACHINES: THE SIMPLE ECONOMICS OF ARTIFICIAL INTELLIGENCE

With unknown knowns, predictive tools appear to provide a very accurate answer, but that answer can be very incorrect, especially if the algorithms have little grasp of the decision process that created the data.

To support their point of view, the authors make reference to pricing and revenue analysis in the hotel industry, although the same viewpoint is applicable elsewhere.

In many industries, higher prices are analogous to higher sales, and vice versa.

For example, in the airline industry, airfares are low outside the peak season, and high during peak seasons (summer and festive) when travel demand is highest.

Presented with this data, and without an understanding that price movements are often a function of demand and supply factors, a simple prediction model might advocate raising different route airfares to sell more empty seats and increase revenues. Evidence of causal inference problems.

But, a human being with a solid understanding of economics concepts will immediately call attention to the fact that increasing airfares is unlikely to increase flight ticket sales.

To the machine, this is an unknown known. But to a human with knowledge of pricing and profitability analysis, this is a known unknown or maybe even a known known provided the human is able to properly model the pricing decision.

Thus, to address such shortcomings, humans should work with machines to identify the right data and appropriate data analysis models that take into consideration seasonality and other demand and supply factors to better predict revenues at different prices.

As data analytics and AI systems become more advanced and spread across industries, and up and down the value chain, companies that will progress further are those that are continually thinking of creative ways for machines to integrate and amplify human capabilities.

In contrast, those companies that are using technology simply to cut costs and displace humans will eventually stop making progress, and cease to exist.

Thanks for sharing:

Leave a Reply

Your email address will not be published. Required fields are marked *

Subscribe to get notified of new posts by email

Recent Posts

Categories

Finance Analytics: It’s Not About the Size of The Data

As the need to make impactful operational and strategic decisions in real time increases, CFOs are playing a greater role in the adoption and integration of data analytics in their organizations to support data-driven decision making.

Executives and business unit leaders are increasingly relying on insights produced by Finance to better understand enterprise performance. That is, what has happened, why it has happened, what is most likely to happen in the future, and the appropriate course of action to take.

In an era where data is proliferating in volume and variety, decision makers have realized it’s no longer enough to base key enterprise performance and risk decisions on experience and intuition alone.

Rather, this must be combined with a facts-based approach. Which means CFOs must set up modernized reporting and analytics capabilities with one of the main goals being the use of data as a tool for business decision making.

Appropriately analyzed and interpreted, data always has a story, and there’s always something to discover from it. However, many finance functions are failing to deliver value from their existing data analytics capabilities.

There is a misconception that to deliver actionable insights, the function needs more data for analysis. As a result, the supply of data keeps rising, while the ability to use it to generate informed insights lags badly.

Yet it’s not about the size of the data. It’s about translating available data and making it understandable and useful.

In other words, it’s about context and understanding that numbers alone do not tell the whole story. Finance leaders should connect the dots in ways that produce valuable insights or discoveries, and determine for example:

  • What is being measured, why, and how is it measured?
  • How extensive the exploration for such discoveries was?
  • How many additional factors were also reviewed for a correlation?

Further, to use data intelligently and influence better decision making, CFOs and their teams should recognize that most enterprise data is accumulated not to serve analytics, but as the by-product of routine tasks and activities.

Consider customer online and offline purchases data. Social media posts. Logs of customer communications for billing and other transactional purposes.

Such data is not produced for the purpose of prediction yet when analyzed, this data can reveal valuable insights that can be translated into action which delivers measurable benefits.

Often the company already has the data that it needs to answer its critical business performance questions, but little of it is being aggregated, cleaned, analyzed, and linked to decision making activities in a coherent way.

Exacerbating the issue is the mere fact that the company has a mishmash of incompatible computer systems and data formats added over the years ultimately making it difficult to perform granular analysis at a product, supplier, geographic, customer, and channel level, and many other variables.

There is nothing grand about data itself. What matters most is how you are handling the flood of data your systems are collecting daily. Yes, data can always be accumulated but as a finance leader:

  • Are you taking time to dig down into the data and observing patterns?
  • Are the observed patterns significant to altering the strategic direction of the organization?
  • Are you measuring what you really want to know, what matters for the success of the business?
  • Or you are just measuring what is easy to measure rather than what is most relevant?

CFOs do not need more data. What they need right now is the ability to aggregate, clean and analyze the existing data sitting in the company’s computer systems and understand what story it is telling them.

Before they can focus on prediction, they first need to observe what is happening and why. Bear in mind correlation does not imply causation.

Yes, you might have discovered a predictive relationship between X and Y but this does not mean one causes the other, not even indirectly.

For instance, employee training hours and sales revenue. Just because there is a high correlation between the two does not mean increase in training hours is causing a corresponding increase in sales revenue. A third variable might be driving the revenue the increase.

Jumping to conclusions too soon about causality for a correlation observed in data can lead to bad decisions and far-reaching consequences, hence finance leaders should validate whether an observed trend is real rather than misleading noise before providing any causal explanation.

Certainly, big data can be a powerful tool, but it has its limits. Not all data is created equal, or evenly valuable. There are situations where big data sets play a pivotal role, and others where small, rich data sets trump big data sets.

Before they decide to collect more data, CFOs should always remember data is comparable to an unexploited resource.

Even though data is now considered an important strategic asset for the organization, raw data is like oil that has been drilled and pulled out of the ground but not yet refined to its finer version of kerosene and gasoline.

The data oil has not yet been converted into insights that can be translated into action to cut costs, boost revenues, streamline operations, and guide the company’s strategic direction.

Thanks for sharing:

Doing The Right Thing For Too Long

Markets and business models are shifting, and so should you keep up with these market changes if your business is to survive and succeed. Compared with the past, the current era of digitization represents an inflection point.

Consider individual trends such as artificial intelligence, virtual reality, Big Data, cybersecurity threats, drones, the Internet of Things, driverless cars, blockchain technologies, and more.

These new technologies have significantly changed the way we connect and interact as individuals, including how businesses deliver products and services to their customers.

Reinventing your business will determine whether you succeed or fail in the digital age. As the saying goes, disrupt or be disrupted. No company, business, or industry is safe from disruption. Today, individual businesses have the potential to compete against multinational companies and win.

These businesses are quick to anticipate market changes and flexible to get ahead of the curve. Sadly, many companies are blinded by their successes and aren’t willing to disrupt themselves. They are not experiencing their desired growth trajectory because they are stuck doing the right thing for too long.

Don’t get comfortable with the status quo and allow your business to get stuck on a strategy and mindset that no longer fit the market.

Here are a few questions to ponder, the answers to which will determine the future of your business:

  • What is at the core of your strategy?
  • Are you in touch with the customers you want to serve? When customers give you negative feedback, how often do you listen and act on it?
  • Are you operating your business on the premise that you know what is best for your customers therefore they are supposed to buy whatever product or service you offer them?
  • Are you keeping up with market shifts or you only know how to grow under one set of conditions or products and services, but not how to survive and strive under another?
  • How robust and flexible is your IT infrastructure to help you innovate, perform your company’s Jobs To Be Done, and scale your business?
  • Are you creating a strong culture that is focused on customers, including a culture that not only embraces change but seeks it out?

Given our world is changing faster, it’s imperative to continuously look for signs that things are changing and think about how those shifts would play out in the short-term, medium-term, and long-term, not forgetting the impact on the execution of your strategy and enterprise performance.

The signs can reveal individually. At times, they are part of a wider trend.

Nonetheless, how you adapt will determine whether you succeed or fail. Keep learning. Learn about innovations in your industry and beyond. Try out new business models and technologies and embrace a philosophy of constant change.

Once you understand how the market is changing and evolving, you can develop the right product or service and strategy that will help you achieve your desired outcomes.

We often talk of the ability to “connect the dots” and “take a helicopter view of the business” as key ingredients for success. But how often are business leaders and their teams doing this?

Across the organization, a culture of “them versus us” prevails. Important decisions are made at a functional level with little or no consideration of their impact at the enterprise level.

Having the ability to grasp the big picture and see how different trends intersect is essential for determining the right path or course of action to pursue.

So, how do you spot market transitions and develop a clear sense of where the market is going?

  • Be curious and hungry for new ideas. Continuously ask tons of key performance questions and pay attention to what’s around you.
  • From time to time, challenge conventional wisdom. It’s easy to stick with what you know about your business model, customers, competitors, markets, or industry but dare to pivot when conditions change.
  • Don’t be nostalgic about the past or worried about protecting what you’ve built in the present. Always be curious about the future and develop a willingness to take calculated risks.
  • Ask existing and would-be customers how they feel about your company’s products, services, and strategy. Instead of turning to sources that reinforce your existing point of view, seek multiple perspectives and cross-reference them as new facts come in.
  • Develop an ability to handle multiple random data points at once. This will help you generate critical market, customer, and business performance insights and make smarter, informed decisions. Be careful to distinguish between the signal and the noise since data can be deceiving, especially when you’re looking for “confirmation” that protects your business model.

Data might not tell you why something is happening, but it does tell you what’s going on.

  • Look for patterns and abnormalities that might suggest something is going on, including any interdependencies.
  • Anticipate all the various scenarios of what could happen.
  • Plan your course of action in response to what’s happening in real time.

As the signals of a market shift increase, the need to act becomes more imperative. Note, monitoring and identifying market shifts, and effectively taking the appropriate course of action is a matter of timing.

If you continue doing the right thing for too long and lack the boldness to disrupt both the market and your own organization, you risk being disrupted and left behind. There is no company that is too big to fail. Neither is there a startup that is too small to succeed.

Thanks for sharing:

How Feasible Are Your Strategic Objectives?

Every organization sets out its goals and objectives, to accomplish its mission and vision. The two often seem like two interchangeable phrases but there is a distinction.

A goal is a desired result you want to achieve and is typically broad and vague. An objective, on the other hand, defines the specific, measurable actions each employee must take to achieve the overall goal.

It is every leader’s job to create a coherent set of feasible objectives or what Richard Rumelt calls proximate objectives. Objectives that define targets the organization is fairly expected to achieve, even overwhelm.

This is essential for ensuring energy and resources are focused on one, or a very few, critical objectives whose accomplishment will lead to a cascade of positive outcomes.

An effective strategy defines a critical challenge or opportunity and clearly articulates how the organization is going to play to win or perform customers’ Jobs to Be Done.

Thus, the objectives an effective strategy sets should stand a good chance of being accomplished, given existing resources and competence.

On the contrary, a bad strategy results in the setting of bad strategic objectives.

Long lists of “things to be done,” are often labeled wrongly as strategies or objectives. Or the desired outcome is simply rehashed with no explanation of how this will be accomplished.

It doesn’t matter how well-thought your strategy is in response to an identified challenge or opportunity. If the resultant strategic objectives are merely a list of things to do, or just as difficult to achieve as the identified key challenge, there has been little value added by the strategy.

In today’s highly competitive, uncertain, dynamic, and complex environment in which a leader’s ability to look further ahead is diminished, it is better to focus on a few pivotal items through taking strong positions, creating options, and building advantage.

First identify the key challenges or opportunities for the business. Look very closely at the changes happening within your business, where you might get an added advantage over competition.

Next, create a list of the issues, including the actions your company should take.

Then, trim the original list to a noticeably short list of pivotal issues and proximate objectives by identifying one or two feasible objective(s), when achieved, would make the biggest difference. Remember, the identified objectives should be more like tasks and less like goals.

Now, focus on the objectives by channeling skills and available resources to accomplish the overall goal.

Once accomplished, new opportunities will open up resulting in the creation of more ambitious objectives. This cycle will help you develop a system that enables the setting of feasible strategic objectives.

Thanks for sharing: