Observation

Most of our learnings come from direct and indirect experiences or observations, and I dare to say we absorb most of them through the eyes. It is said that more than 50% of the cortex is devoted to processing visual information. Even if you don't consider yourself a visual person compared to others, you are a visual being.

This is typically how businesses are run; a manager looks at what is visible to them, and then, many times based on a gut feeling or intuition, they make their decisions and hope they are right. The decision-making process is complex. There are many variables to take into consideration, and sometimes consequences are difficult to predict. Our mind is exposed to this kind of complexity every day and intuition helps us deal with that complexity while giving us solutions we can work with. Without it, we would be paralyzed as we would not be able to formalize everything needed for an optimal decision (if that even exists).

However, the same way humans ran up the hills to have a better clear view of their surroundings, decision-makers can also position themselves for better decision-making through communication and technology. By feeding their intuition with more data and sometimes even tests and simulations, they will make better decisions. It sounds cliche, but it's the decisions you make today that shape and define the company's success tomorrow.

There's no science without observation

The scientific process is not much different from any other learning process, really. It goes like this:

1. observation;
2. question;
3. hypothesis;
4. experience;
5. results;
6. conclusion.

That is very similar to the way people learn even the most basic things. You see something, create your own theories about it and then try to confirm or deny those theories. The difference is that the scientific process reduces biases and increases the accuracy of our learnings by implementing a formalized process. It is based on experiments, and even if you don't consider yourself a scientific person, you probably understand the value of experience. Typically, the more experienced a person is, the less likely they are to make mistakes, and the faster they reach their goals.

People that understand the value of being data-driven understand the importance of observation in decision-making. These same people will realize the power of improving those observations' quality and quantity.

Flying blind

"Flying blind" is an expression among pilots for flying in extremely low visibility and relying on the plane instruments instead.

Imagine tracking a plane's movement with just your natural senses: eyesight, smell, hearing, and so on. You would have to estimate its speed, altitude, engine temperature, cabin pressure, and even position without any instruments. Would you trust yourself or even the most experienced pilot to take you or your family on a flight like that? Probably not.

Nowadays, even if you're flying one of the most basic flying vehicles like a paraglider, you will probably be equipped with a variometer and a GNSS unit. And there's a reason for that: It reduces risks while improving efficiency and effectiveness, sometimes even a more comfortable trip or experience. And the "only thing" they do, provides pilots with the information they need.

Now think about your business. Just like a pilot, you're in charge of making decisions that put you responsible for the success of not just you but other people, perhaps your team or the business you are working with. Ask yourself, are you flying your decision looking through the window? Or are you checking your cockpit for any values that will help you make better decisions? It is your responsibility to make sure you are informed in the best way possible to make decisions. You're the pilot of that plane.

Reach your top performance

Top performance individuals and companies have one thing in common. They measure things.

Think of sports, for example. There was a time when talent and effort were everything. If you were talented and tried hard enough, you would reach the top. Those days are over. Now top athletes are not just working hard; they are working smarter. Today, top athletes keep track of everything: their performance, mood, calorie intake, sleep quality, you name it. They wouldn't be doing it if it wasn't for some solid reasons.

For both athletes, investors, business, or anyone else, some of these reasons can be:

• Identify what is affecting their performance - a top athlete will not be evaluated by their sleep quality, but their performance is most likely affected by it.
• Being able to keep focused on things that truly matter.
• Tracking improvement (and decline) in performance.
• Highlighting the weaknesses and bottlenecks.
• Exposing possibly unknown competitive edges.

To reach the highest performance possible, make sure you're observing, measuring, and tracking.

Collecting is not the same as tracking

You can collect data from everywhere in your business: website analytics, productivity, finances, social media analytics, operations, sales. Every activity that happens in your business is an opportunity for collecting data. Is it necessarily a good opportunity? No.

Choosing the correct data to collect is difficult and overwhelming at times, even for the most sophisticated manager. Perhaps even more so, as they are most likely to understand the opportunities in data. Don't try to collect everything, especially if you're just starting. Instead, focus on the things you want to track and begin by collecting these.

What should you track

Some people would say, "You have to track everything.". I disagree. You can collect data for many things, but tracking everything consciously and 24/7 is simply unproductive and might even damage your efforts.

With EAI, I've worked with companies that wanted to track everything and visualize everything, every time, everywhere. Bad idea. You should follow what is critical and driving your business. In the beginning, you won't know what this is, and that's ok.

One thing you should definitely track is your objectives. Fortunately, this is understood by most people in business, but many times not applied correctly.

How you do this is up to you, there are several frameworks with considerable overlap. You might have heard of SMART - Specific, Measurable, Achievable, Realistic, and Timely; OKRs - Objective Key Results and NSM - North Star Metrics. Learn about those frameworks and apply the one that fits you and your company the best. Give it some time to test it, and be willing to change: be it the objectives themselves or the framework altogether. Remember that a framework is just a framework.

What comes before objectives

The problem with measuring objectives alone is that you're measuring the results of previous actions.

Imagine a company that sells shoes and wants to maximize its profits as much as possible for the next quarter. Even if you had enough stock until the end of that quarter, you would never fire everyone from your factories just to minimize costs and maximize profits, right? I hope so.

It sounds absurd, and it is. The reason it sounds ridiculous is that it's super obvious that while in the short-term, firing everyone might lead to a higher profit, in the medium-term, this is unsustainable because the stock will be over soon. If you focus only on objectives, you will most likely make similar but less obvious mistakes, like reducing the budget for brand recognition or growing your social media, simply because the result isn't immediate.

By tracking objectives alone, you are tracking the past. You are looking at the results that your past actions delivered. That is not, however, enough to predict the future. For better predictions, you need to measure what took you to where you are today. You need to measure the things you did in the past that lead to where you are today. We call these "leading indicators".

Many of the decisions and actions you do today will have a delayed impact. That's why you need to measure these and their results to make better decisions. This is not easy. Sometimes you make decisions that seem to take the company nowhere, and other times you think the current results are coming from your recent choices when they are, in fact, coming from previous ones.

The solution is simple. Define your leading indicators and keep track of them. When you measure the results, make sure you look at how the leading indicators looked like back then. Now you'll be able to define better targets for the leading indicators you have today.

Detecting anomalies

While there's little value in personally tracking your indicators 24/7, that doesn't mean there is no value in collecting other data. One of the reasons to do that is to identify some kind of event in your data.

Before I explain what I consider an event, let's look at the following table:

Event	Cause	Evidence in Data
Increased sales	An influencer talks about your brand	1. Anomalous growth in the number of new followers today 2. Daily sales much above the daily average.
Broken website	Your website is down, and users can no longer use it.	A massive decrease in online traffic and sales.

Usually, if there's any kind of event in your business, you will want to know about it. Fortunately, it will most likely be represented in your data, one way or the other.

Being able to identify events can help in a number of different ways:

• You'll have a warning system, for when something fails;
• It will work as a filter, delivering you information that matters when it matters;
• It will help you keep track of the relevant changes instead of keeping you updated with every single metric every day.

Often, if you set up your anomaly detection system correctly, you'll know when something is happening in your business, good or bad. You might not know what or why it is happening, but you'll be informed that it is, and you'll be able to act on it.

If you don't understand how a system can identify what data is relevant enough to be presented to you, don't worry. There are several ways to implement this, which we'll talk about later in the book.

By detecting anomalies, you're closer to building a resilient system, but that is not enough. In the next chapter, we'll talk about creating and maintaining systems that can quickly and effectively recover from difficult conditions by talking about concepts such as microservices and anti-fragile designs. As always, we'll keep it as conceptual as possible in Part I and save any technical approaches for later in the book.