<ul><li>Graduation Versus Salary</li><li>Stocks Versus Cash</li><li>Vacation Versus training</li><li>Paying off debt Versus Spending on Welfare by the government</li><li>Entrepreneurship versus steady job</li><li>Selling Stocks now and 2 months later</li><li>Investing in stocks or a higher degree</li></ul>

Top 7 Examples of Opportunity Cost

A person named X is currently working in a company and drawing some salary. X is getting an option to do the graduation for 2 years, but for that, he/she must leave his/her job. If he/she doesn’t go to graduation, the opportunity cost will be a higher degree plus the additional salary that he/she might get because of this degree. On the other hand, if he/she opts for the job, then the opportunity cost will be 2 years’ salary that has to be foregone.

Graduation Versus Salary

Let’s say you got $50,000 in your bank account, which you don’t need immediately. You can have many options with this cash. For example, you can just keep this cash, or you can invest this money in stocks. Let’s say you would have invested wisely that money in some good stocks, and after a year, 50,000 would become $60,000. On the other hand, if you keep this money idle instead of investing in cash, then your opportunity cost will be the difference between 60000 and 50000, i.e., $10,000

Stock Versus Cash

Your school has announced a vacation. You finally get a chance to go on holiday for the next 1 month. But suddenly you come to know that training is scheduled for your favorite sports which you do not want to miss. So, if you go on vacation, your opportunity cost will be missing the training session, and if you stay for training, then your opportunity cost will be enjoying the vacation.

Vacation Versus Training

A country’s government is preparing its budget. It has some surplus which can be used for paying off its debtor or it can be used for introducing some welfare schemes for its citizens such as subsidies.If it pays off its debt instead of a welfare scheme, it would be classified as an opportunity cost for its citizens.

Paying off Debt Versus Spending on Welfare by the Government

You have a steady job with a good income, but your passion is to open your own business, which requires you to leave your current job, and you have to spend a lot of money on opening a business. You have both options. If you are choosing a steady job instead of opening a new business, then your opportunity cost will be not having the work you wanted and maybe success because of the new business. If you start your business, the opportunity cost would be a steady job and a regular paycheck.

Entrepreneurship Versus Steady Job

You are having shares of a company worth $5,000. You are thinking of selling that shares now or waiting for 2 more months. Let’s say, after 2 months, shares would be worth $6,000. If you sell the shares now, then your opportunity cost would be 6000-5000= $1,000, which you could have gotten if you would have waited for 2 more months.

Selling Stocks Now and 2 Months Later

You have got $20,000 which you are thinking of either investing in shares of some company, or you can invest this money in getting a higher degree in a good university. If you invest in stocks, then your opportunity cost would be higher degrees and a higher salary because of the degree. But if you invest in a higher degree, then your opportunity cost would be the profit gained from those shares.

 Investing in Stocks or Higher Degree

Learn to evaluate life and business choices via the lens of Opportunity Costs

problemsolving

Problem Solving

business

Business

moneyandinvestments

Money & Investments

What Is Opportunity Cost

Opportunity cost examples that provide an outline of the most common Opportunity Cost.

Guide to Opportunity Cost Examples. Here we discuss its definition and the top 7 examples along with detailed overview and explanations.

Opportunity Cost Examples

Human and animal excrement is a natural, renewable, and sustainable resource - if only we can overcome our visceral disgust of it!An average adult produces about a pound (or half a kilo) of poo a day. That means that New York City, with its official census population of more than 8 million, pumps out more than 8 million lbs (or 4 million kg/4,000 tonnes) of excrement a day. Tokyo surpasses that slightly with 8.3 million lbs daily. China’s capital Beijing, a huge urban conglomerate of 21.3 million dwellers, beats NYC and Tokyo combined.

Holy Crap!

<ul><li>In the developed world, we built flushing toilets, underground pipes and gigantic sewage treatment plants to protect ourselves from our massive sh*t piles</li><li>Yet, these marvels of modern engineering have done significant damage to Earth's ecology.</li><li>Soils grow barren, so we use synthetic fertiliser, which isn't anywhere near as good as the real shit and also is very polluting to produce.</li><li>Restoring that broken link is essential to our food security and healthy ecology.</li></ul>

Those Sewage Systems Aren't Environmentally Friendly

<ul><li>For the most part, we try to distance ourselves from it as much as possible. The exact mechanisms of that process depend on where you live. </li><li>In the Western world, we flush it down the toilet. </li><li>In the less fortunate places, it’s left to decompose in pit latrines or underneath trees. </li><li>But no matter the country or the culture, the common thread is that we try to move our ordure as far away from us as we can. </li><li>We’re universally disgusted by it. It’s excrement. It’s yucky by definition. It’s appalling by sight and smell.</li></ul>

What The World Does With Poo

Dumping waste in a local body of water has proved dangerous. The infamous <a href="https://aeon.co/essays/the-next-pandemic-will-be-nothing-like-ebola">pandemics</a> of cholera that swept through Europe in the 19th and early 20th centuries were started by faecal contamination of drinking water. Even today, diarrhoeal diseases sicken and kill about 827,000 people a year in the developing world, <a href="https://www.who.int/en/news-room/fact-sheets/detail/sanitation">according</a> to the World Health Organization.

Dumping Poo In Water Isn't A Great Idea

If you think about where our food comes from, particularly in colder climates, you’ll realise it’s grown someplace else. As our bananas, apples, lettuce, corn and rice grow, they extract nutrients from the land. That food is then trucked, shipped and flown to where we live – and where we eat it and excrete it. Yet, we don’t return that organic matter back to where we got it from, as my grandfather did. We don’t drive, sail or fly to return this organic bounty to the land. We flush it down the drain.

We Don't Let Poo Go Back To Where It Can Do Good

Our local sewage treatment plants clean that water from pathogens, but not from the nitrogen, phosphorus and potassium this wastewater is rich in. These potent fertilisers usually flow into a body of water nearby, constantly overfeeding the lakes, rivers and the ocean. That results in toxic algae blooms, dead fish and decaying waterways, which are not biologically designed to absorb so many fertilising chemicals. Likewise, our Earth is not biologically designed to keep growing food without us feeding it.

The Drainage System Can Collapse The Earth

<ul><li>Because we don’t ship our shit back to where the food comes from, we keep perpetuating the redistribution of nutrients on the planet. </li><li>Soils grow barren, so we use synthetic fertiliser, which isn’t anywhere near as good as the real shit, and also is very polluting to produce. </li><li>In our quest to rid ourselves of our dangerous dark matter, we’ve broken the essential rules and laws of Mother Nature. </li><li>By taking our poo out of the equation, we altered not only our agriculture but the entire planet’s ecology.</li></ul>

The Soil Needs Poo But We Are Depriving It

<ul><li>With so much smart tech, why haven’t we closed our metabolic rift? The problem is that we have to mend another huge fissure in the excremental ideology – not the metabolic but the mindset one. </li><li>Unlike people in ancient societies, we still think of our excrement as the ultimate waste product that needs to be dealt with. We still don’t view it as an extremely valuable and versatile asset. </li><li>We spend our efforts and money to remove the dangerous filth rather than to acquire and use a superb product of our metabolic bodies. And that’s the leap of thinking we must achieve to fix the problem fully.</li></ul>

A Problem To Solve For The New Generation

Human and animal waste has been misunderstood all along. It is what keeps the earth's ecology balanced.

Our excrement is a natural, renewable and sustainable resource – if only we can overcome our visceral disgust of it

scienceandnature

Science & Nature

A short biography of human waste and its value | Aeon Essays

This plant can endure a lot of neglect. Sansevierias can tolerate very low light if there is some natural light in the room. They have leaves that store water and can go for longer stretches between watering. There are over 60 different species of sansevieria. Shapes can range from the classic “snake plant” look, to more interesting options like starfish .

Sansevieria (Aka Snake Plant And Mother In Law's Tongue)

These waxy-leaved plants need lots of bright, filtered light but no sunlight. The soil needs to dry out almost completely between waterings for the best result. These vines will climb, or you can let them trail over the sides of its pot. They put out numerous beautiful little flowers.

Hoya (Aka Wax Plant)

Ficus elastica is a large plant that can fill a not-so-bright corner. It will thrive in lower-light conditions and is very forgiving when it comes to missed watering. Its large, glossy leaves might need an occasional wipe-down to get rid of dust and water spots. You need to repot your ficus elastica once every couple of years.

Ficus Elastica (Aka Rubber Plant)

The dracaena marginata is sometimes called the dragon tree. This drought-tolerant plant is great for beginners. It prefers medium light but will also do well in low light to partial shade. This plant can grow fairly tall, often over six feet, even indoors, which can be a stunning addition to spaces with high ceilings.

Dracaena Marginata

It requires bright to medium indirect light and medium water but can tolerate a little drought. A happy Monstera can result in a voluminous plant that might require pruning, but this plant is also easy to propagate from cuttings.

Monstera Deliciosa (Aka Swiss Cheese Plant)

The ZZ plant is an excellent pick for forgetful caregivers, withstanding periods of drought and lower light conditions. Over time the leaves can arch in a graceful fashion. If kept in good light and cared for well, the plant can grow to be quite large.

Zamioculcas Zamiifolia (Aka ZZ Plant)

These similar-looking plants are often confused for one another because they're both vines that can be trellised or left to trail. Both do best with bright, filtered light and watering about once a week, and both offer a variety of leaf shapes and patterns that can add style to any space.

Pothos And Philodendrons

Best Plants That Are Hard to Kill

The startup organisation is one of the greatest ways to make the world a better place. You can unlock human potential if you organise a group of people in a startup with the right equity incentive. Yet, many startups still don't succeed because they lack the right timing.

The startup

<ol><li>Timing accounted for 42 percent of the difference between success and failure.</li><li>Team and execution came second</li><li>The uniqueness of the idea came third.</li><li>The business model and funding are the least important. If you're underfunded at first but then gain traction, is very easy to get intense funding.</li></ol>

Timing is the most important

<ol><li>The idea. However, the idea is not everything. </li><li>The team and execution. Mike Tyson once said, "Everybody has a plan until they get punched in the face." Much about a team's execution is the ability to adapt and to get punched in the face by the customer.</li><li>The business model.</li><li>Funding. Some companies with intense funding don't succeed.</li><li>Timing. The idea is too early, and the world is not ready for it.</li></ol>

Factors that  contribute to start-up success

Aside from a good business model, a good idea and excellent execution, timing determines if a company succeeds or fails.<ul><li>Airbnb. People thought that no one would rent out a space in their home to a stranger. But the company came out during the recession when people needed extra money.</li><li>Z.com (1999-2003) was an online entertainment company with a great idea. But broadband penetration was too low, and it was too hard to watch video content online.  In 2005,  Adobe Flash solved the codec problem, and broadband penetration reached 50 % in America, perfect for YouTube to start and become a success.</li></ul>

Examples of timing

Bill Gross has founded many start-ups and incubated many others. He got curious about why some succeeded and others failed.

Bill Gross has founded a lot of start-ups, and incubated many others — and he got curious about why some succeeded and others failed. So he gathered data from hundreds of companies, his own and other people's, and ranked each company on five key factors. He found one factor that stands out from the others — and surprised even him. TEDTalks is a daily video podcast of the best talks and performances from the TED Conference, where the world's leading thinkers and doers give the talk of their lives in 18 minutes (or less). Look for talks on Technology, Entertainment and Design -- plus science, business, global issues, the arts and much more. Find closed captions and translated subtitles in many languages at http://www.ted.com/talks/bill_gross_the_single_biggest_reason_why_startups_succeed Follow TED news on Twitter: http://www.twitter.com/tednews Like TED on Facebook: https://www.facebook.com/TED Subscribe to our channel: http://www.youtube.com/user/TEDtalksDirector

personaldevelopment

Personal Development

entrepreneurship

Entrepreneurship

startup

Startups

videos

Videos

Bill Gross has founded a lot of start-ups, and incubated many others — and he got curious about why some succeeded and others failed. So he gathered data from hundreds of companies, his own and other people's, and ranked each company on five key factors. He found one factor that stands out from the others — and surprised even him.

TEDTalks is a daily video podcast of the best talks and performances from the TED Conference, where the world's leading thinkers and doers give the talk of their lives in 18 minutes (or less). Look for talks on Technology, Entertainment and Design -- plus science, business, global issues, the arts and much more.
Find closed captions and translated subtitles in many languages at http://www.ted.com/talks/bill_gross_the_single_biggest_reason_why_startups_succeed

Follow TED news on Twitter: http://www.twitter.com/tednews
Like TED on Facebook: https://www.facebook.com/TED

Subscribe to our channel: http://www.youtube.com/user/TEDtalksDirector

The TED Talks channel features the best talks and performances from the TED Conference, where the world's leading thinkers and doers give the talk of their lives in 18 minutes (or less). Look for talks on Technology, Entertainment and Design -- plus science, business, global issues, the arts and more. You're welcome to link to or embed these videos, forward them to others and share these ideas with people you know.

TED's videos may be used for non-commercial purposes under a Creative Commons License, Attribution–Non Commercial–No Derivatives (or the CC BY – NC – ND 4.0 International) and in accordance with our TED Talks Usage Policy: https://www.ted.com/about/our-organization/our-policies-terms/ted-talks-usage-policy. For more information on using TED for commercial purposes (e.g. employee learning, in a film or online course), please submit a Media Request at https://media-requests.ted.com

The single biggest reason why start-ups succeed | Bill Gross

Technical debt refers to a rushed development process or a lack of shared knowledge among team members. However, in many cases, technical debt is inevitable and is part of a normal software development process.The negative impact of technical debt on businesses is massive.Here are the best ways to prevent and manage technical debt that both startups and enterprise companies can implement today.

Defining Technical Debt

Imagine two software engineers have worked on a new feature. They own all knowledge about this specific feature. However, other software engineers in your team don't have this knowledge. When they have to work on this feature or have to do an implementation that makes use of this feature, they won't have the knowledge required to correctly implement a new feature.Therefore, it's essential that you actively share knowledge about features and important codebase changes.

Knowledge-Based Tech Debt

Design debt often occurs in highly competitive markets or startups where speed to market often is the highest priority.When delivering new features quickly, you're not thinking about the structure of features or the architecture of your codebase. It often gets neglected, making it increasingly harder to add new features. And now you have to fix it as a team.Therefore, design debt is closely linked to structuring features and adhering to design patterns.

Design Debt

Code debt is about writing bad code and not fixing this in time. For instance, a developer wants to quickly merge code without writing sufficient tests or adhering to code standards.Many organizations use automation tools such as pre-commit hooks with code linting to verify code quality. When you don’t implement such code checks, bad code can quickly decline the overall quality of your codebase.

Code Debt

A refactoring week allows your team to resolve open bugs, evaluate the current architecture, and prepare the architecture for the upcoming product features. Benefit: A refactoring week gives the needed breathing room for developers to evaluate and reflect on the code before implementing a new set of features. It’s great for solving big pieces of debt.Drawback: The development process slows down while you are doing refactoring and your team is not solving debt continuously.

Three Best Ways to Fight and Prevent Technical Debt: Code and Architecture Refactoring

A retrospective meeting addresses what went well and what didn’t. It’s an open stage to share feedback without placing blame. It would be best if you focused on improvement.Benefit: You can use retrospective meetings to share updates about the code. Engineers can show what they have accomplished.  Drawback: Stakeholders and managers need to be on board with the vision and give engineers time to organise these meetings.

Start Regular Technical Debt Discussions

Tracking technical debt in the editor allows engineers to:<ul><li>Get full visibility on technical debt</li><li>See context for each codebase issue</li><li>Reduce context switching</li><li>Solve technical debt continuously</li></ul>Benefit: Developers already spend most of their time in the editor so it's the best place to track and report technical issues. Implementing a process for managing technical debt will positively influence engineering team morale and customer satisfaction.Drawback: Starting a new habit takes time and effort from team members and it's impossible to do without a tech debt hero.

Start Tracking Technical Debt in Your Editor

Track metrics related to different types of tech debt.<ul><li>Overall code coverage percentage and code coverage per feature</li><li>Number of failed CI/CD builds</li><li>Bug count per week or month</li><li>Feature throughput</li><li>Issues with non-functional requirements</li></ul>

How to Detect a Growing Technical Debt

<ul><li>Startups can easily solve technical debt by implementing automated tools that verify code quality.</li><li>Large enterprises need to use project management tools to understand which features are in development and who’s working on specific areas of the codebase.</li><li>They should use automated code quality tools to ensure overall codebase health.</li></ul>

How to Deal with Tech Debt Differently

Leaving technical debt unaddressed can result in several issues for your organisation, such as higher total cost of ownership, slower time to market, reduced agility because of a poorly designed architecture, and poor security.Make sure to have a process that makes it easy to share knowledge with team members and resolve technical debt continuously.

The Bottom Line

career

Career

technologyandthefuture

Technology & The Future

computerscience

Computer Science

Kill Technical Debt Before it Kills Your Project

Data bias can have notable implications for research and practical applications. For example, in a Facebook scandal, its AI shockingly asked users if they wanted to continue seeing videos about primates after watching a video featuring Black men. Data bias refers to data sets that don't represent the population in study. Models trained on biased data could contain prejudice. That's why AI Researchers and Data Scientists must be vigilant to ensure models don't contain any bias.

AI models trained with biased data

The most common avenues for collecting training data:<ul><li>Paying for data sets</li><li>Using public data sets</li><li>Sourcing open source content</li><li>Using in-person or field-collected data sets</li></ul> A combination of all four will provide the best training data.

Collect data from a variety of sources.

Ensure the data is diverse. Speakers in the audio or video files should possess a range of characteristics, including locations, dialects, genders, sex, race, and nationality. Sourcing such data could be difficult if you only rely on open-source data.

Ensure diverse data

Once you've ensured your initial data set is diverse, your model can still have bias. That's why you should monitor your model's real-world performance. For example, does your model better predict female speech over male speech? If so, retrain with new datasets to overcome any problem areas.

Monitor real-world performance

This article will look at three ways to limit data bias: collecting data from a variety of sources, ensuring data is diverse, and monitoring real world performa

artificialintelligence

Artificial Intelligence

What Is Data Bias and How to Avoid It

You may have been to a presentation that is rich in data, but there's no story or anything compelling to help you connect the data to the larger insights. Without a story, the audience's eyes may start to gloss over as they try to remain present. You may struggle to understand what the presenter is trying to say.

Presenting data without a story is painfully boring

The story helps to frame communication. Descriptors like people, pictures, colours, sounds, and points of view give data meaning. For example, a politician may tell a story about Casey, whose life was improved because of specific legislation. As a listener, you may not know the details of the legislation, but you will learn how it helped Casey. Consider adding an overview or summary, so the reader knows where you're taking them. Adding visuals can bring the story to life.

Start with a story

Data is just information. It needs to be organised, distilled and interpreted so the audience knows what the message is and what actions can be taken. Creating a story gives data a voice. Consider framing your data in an amusing or comedic way.

Make the data entertaining

When stories accompany data, it allows us to see into other perspectives. Stories make data relatable and memorable. It connects the audience to the data.

Put things into context

When data is connected to an effective story, it becomes a supporting actor in a larger production. As a result, people will feel connected to the data and be more likely to take action.

Craft connection

The type of story depends on your audience. Some people will need more granular information, while others are only comfortable with the highlights. For example, when you are presenting to the board, a story about how customers are impacted is more compelling than a story about Casey.

Know your audience

Data plus a narrative makes for a better presentation.

communication

Communication

How to Tell a Good Story with Data

Evidence-based management is a practice of management and decision-making based on critical thinking and reliable evidence.If you have made a decision based on so-called “best practices” and if you haven’t questioned the culture, myths, and ideologies that contributed to those practices—your decision-making framework is fundamentally flawed.

Evidence-Based Management

Six steps you can follow to increase the likelihood of a favorable outcome from your decisions:<ol><li>Asking: Translating a practical issue or problem into an answerable question</li><li>Acquiring: Systematically searching for and retrieving the evidence </li><li>Appraising: Critically judging the trustworthiness and relevance of the evidence</li><li>Aggregating: Weighing and pulling together the evidence </li><li>Applying: Incorporating the evidence into the decision-making process </li><li>Assessing: Evaluating the outcome of the decision.</li></ol>

How to Use Evidence-Based Management

<ul><li>Scientific literature. Knowledge from scientific evidence is more accurate than the opinions of experts or their best practices. </li><li>Internal data. Professional judgments based on hard data are far more accurate than judgments based on individual experience. </li><li>Professional expertise. The professional experience of many people results in more accurate decisions than the personal experience of one or two individuals.</li><li>Stakeholders’ values and concerns. Consulting the people affected by our decisions can be done with focus groups, internal surveys, or enterprise discussion management tools.</li></ul>

Sources of Evidence for Leaders

<ul><li>Subscriptions to academic journals. JSTOR, Directory of Open Access Journals, and ScienceDirect are just a few to consider.</li><li>Business intelligence tools. Leveraging your organization’s business analyst or data science team is a good way to ensure the data sets you’re using to make accurate decisions. </li><li>Membership to industry associations: they give you access to perspectives outside of your experience.</li><li>Enterprise discussion management: they are critical in helping us connect our workforce and gauge stakeholders’ values and concerns about a specific topic. </li></ul>

Tools to Enable Evidence-Based Management

Evidence-based management is an approach that involves consciously setting aside the accepted conventions and hierarchy of opinion, and instead, using critical thinking and the best available evidence to make decisions.

leadershipandmanagement

Leadership & Management

What is Evidence-Based Management?

Not only do we typically work with very large data sets, but those data sets are extremely rich. That is, each row of data typically has many, many attributes. When you combine this with the temporal sequences of events for a given user, there are an enormous number of ways of looking at the data. 
Contrast this with a typical academic psychology experiment where it's trivial for the researcher to look at every single data point. The problems posed by our large, high-dimensional data sets are very different from those encountered throughout most of the history of scientific work.

New Frontiers In Data Analysis

Most practitioners use summary metrics (for example, mean, median, standard deviation, and so on) to communicate about distributions. 
However, you should usually examine much richer distribution representations by generating histograms, cumulative distribution functions (CDFs), Quantile-Quantile (Q-Q) plots, and so on. These richer representations allow you to detect important features of the data, such as multimodal behavior or a significant class of outliers.

Data Distributions

Examine outliers carefully because they can be canaries in the coal mine that indicate more fundamental problems with your analysis. 
It's fine to exclude outliers from your data or to lump them together into an "unusual" category, but you should make sure that you know why data ended up in that category.

Consider The Outliers

Randomness exists and will fool us. Some people think, “Google has so much data; the noise goes away.” This simply isn’t true. Every number or summary of data that you produce should have an accompanying notion of your confidence in this estimate (through measures such as confidence intervals and p-values).

Look At The Noise

Anytime you are producing new analysis code, you need to look at examples from the underlying data and how your code is interpreting those examples. Your analysis is abstracting away many details from the underlying data to produce useful summaries.
How you sample these examples is important:
<ul>
<li>If you are classifying the underlying data, look at examples belonging to each class.</li>
<li>If it's a bigger class, look at more samples.</li>
<li>If you are computing a number (for example, page load time), make sure that you look at extreme examples</li>
</ul>

Look At Examples

Slicing means separating your data into subgroups and looking at metric values for each subgroup separately. We commonly slice along dimensions like browser, locale, domain, device type, and so on. If the underlying phenomenon is likely to work differently across subgroups, you must slice the data to confirm whether that is indeed the case. 
Even if you do not expect slicing to produce different results, looking at a few slices for internal consistency gives you greater confidence that you are measuring the right thing.

Slice Your Data

With a large volume of data, it can be tempting to focus solely on statistical significance or to home in on the details of every bit of data. But you need to ask yourself, "Even if it is true that value X is 0.1% more than value Y, does it matter?" 
This can be especially important if you are unable to understand/categorize part of your data. If you are unable to make sense of some user-agent strings in your logs, whether it represents 0.1% or 10% of the data makes a big difference in how much you should investigate those cases.

Consider Practical Significance

You should almost always try slicing data by units of time because many disturbances to underlying data happen as our systems evolve over time. (We often use days, but other units of time may also be useful.)
During the initial launch of a feature or new data collection, practitioners often carefully check that everything is working as expected. However, many breakages or unexpected behavior can arise over time.

Check For Consistency Over Time

Looking at day-over-day data also gives you a sense of the variation in the data that would eventually lead to confidence intervals or claims of statistical significance. This should not generally replace rigorous confidence-interval calculation, but often with large changes you can see they will be statistically significant just from the day-over-day graphs.

Check For Variation

Almost every large data analysis starts by filtering data in various stages. Maybe you want to consider only US users, or web searches, or searches with ads. Whatever the case, you must:
<ul>
<li>Acknowledge and clearly specify what filtering you are doing.</li>
<li>Count the amount of data being filtered at each step.</li>
<li>Often the best way to do the latter is to compute all your metrics, even for the population you are excluding. You can then look at that data to answer questions like, "What fraction of queries did spam filtering remove?"</li>
</ul>

Acknowledge And Count Your Filtering

The most interesting metrics are ratios of underlying measures. Oftentimes, interesting filtering or other data choices are hidden in the precise definitions of the numerator and denominator. For example, which of the following does “Queries / User” actually mean?
<ul>
<li>Queries / Users with a Query</li>
<li>Queries / Users who visited Google today</li>
<li>Queries / Users with an active account (yes, I would have to define active)</li>
<li>Being really clear here can avoid confusion for yourself and others.</li>
</ul>

Ratios Should Have Clear Numerators And Denominators

Validation: Do I believe the data is self-consistent, that it was collected correctly, and that it represents what I think it does?Description: What's the objective interpretation of this data? For example, "Users make fewer queries classified as X," "In the experiment group, the time between X and Y is 1% larger," and "Fewer users go to the next page of results."Evaluation: Given the description, does the data tell us that something good is happening for the user, for Google, or for the world?

Three Stages Of Data Analysis: Validation, Description, and Evaluation

Before looking at any data, make sure you understand the context in which the data was collected. If the data comes from an experiment, look at the configuration of the experiment. If it's from new client instrumentation, make sure you have at least a rough understanding of how the data is collected. 
You may spot unusual/bad configurations or population restrictions (such as valid data only for Chrome). Anything notable here may help you build and verify theories later.

Confirm Experiment And Data Collection Setup

As part of the "Validation" stage, before actually answering the question you are interested in (for example, "Did adding a picture of a face increase or decrease clicks?"), rule out any other variability in the data that might affect the experiment. For example:
<ul>
<li>Did the number of users change?</li>
<li>Did the right number of affected queries show up in all my subgroups?</li>
<li>Did error rates change?</li>
</ul>
These questions are sensible both for experiment/control comparisons and when examining trends over time.

Check For What Shouldn't Change

When looking at new features and new data, it's particularly tempting to jump right into the metrics that are new or special for this new feature. However, you should always look at standard metrics first, even if you expect them to change. 
For example, when adding a new universal block to the page, make sure you understand the impact on standard metrics like “clicks on web results” before diving into the custom metrics about this new result.

Standard First, Custom Second

Especially if you are trying to capture a new phenomenon, try to measure the same underlying thing in multiple ways. Then, determine whether these multiple measurements are consistent. 
By using multiple measurements, you can identify bugs in measurement or logging code, unexpected features of the underlying data, or filtering steps that are important. It’s even better if you can use different data sources for the measurements.

Measure Twice, Or More

Both slicing and consistency over time are particular examples of checking for reproducibility. If a phenomenon is important and meaningful, you should see it across different user populations and time. But verifying reproducibility means more than performing these two checks. If you are building models of the data, you want those models to be stable across small perturbations in the underlying data.
If a model is not reproducible, you are probably not capturing something fundamental about the underlying process that produced the data.

Check For Reproducibility

Often you will be calculating a metric that is similar to things that have been counted in the past. You should compare your metrics to metrics reported in the past, even if these measurements are on different user populations.
You do not need to get an exact agreement, but you should be in the same ballpark. If you are not, assume that you are wrong until you can fully convince yourself. Most surprising data will turn out to be an error, not a fabulous new insight.

Check For Consistency With Past Measurements

If you create new metrics (possibly by gathering a novel data source) and try to learn something new, you won’t know if your new metric is right. With new metrics, you should first apply them to a known feature or data.
If you have a new metric for where users are directing their attention to the page, make sure it matches what we know from looking at eye-tracking or rater studies about how images affect page attention. Doing this provides validation when you then go to learn something new.

New Metrics Should Be Applied To Old Data/Features First

Typically, data analysis for a complex problem is iterative. You will discover anomalies, trends, or other features of the data. Naturally, you will develop theories to explain this data. Don’t just develop a theory and proclaim it to be true. Look for evidence (inside or outside the data) to confirm/deny this theory.

Looking For Evidence

Good data analysis will have a story to tell. To make sure it’s the right story, you need to tell the story to yourself, then look for evidence that it’s wrong. One way of doing this is to ask yourself, “What experiments would I run that would validate/invalidate the story I am telling?” Even if you don’t/can’t do these experiments, it may give you ideas on how to validate with the data that you do have.

A Story To Tell

When doing exploratory analysis, perform as many iterations of the whole analysis as possible. Typically you will have multiple steps of signal gathering, processing, modeling, etc. If you spend too long getting the very first stage of your initial signals perfect, you are missing out on opportunities to do more iterations in the same amount of time. 
Further, when you finally look at your data at the end, you may make discoveries that change your direction. Therefore, your initial focus should not be on perfection but on getting something reasonable all the way through.

Exploratory Analysis Benefits From End-To-End Iteration

We typically define various metrics around user success.
You can not use the metric that is fed back to your system as a basis for evaluating your change. If you show more ads that get more clicks, you can not use “more clicks” as a basis for deciding that users are happier, even though “more clicks” often means “happier.” Further, you should not even do slicing on the variables that you fed back and manipulated, as that will result in mixed shifts that will be difficult or impossible to understand.

Watch Out For Feedback

<ul><li>As you work with data, you must become both the champion of the insights you are gaining and a skeptic of them. You will hopefully find some interesting phenomena in the data you look at. When you detect an interesting phenomenon, ask yourself the following questions:</li><li>What other data could I gather to show how awesome this is?</li><li>What could I find that would invalidate this?</li></ul>

Skeptic And Champion At One Go

There’s always a motivation to analyze data. Formulating your needs as questions or hypotheses helps ensure that you are gathering the data you should be gathering and that you are thinking about the possible gaps in the data. Of course, the questions you ask should evolve as you look at the data. However, analysis without a question will end up aimless.
Avoid the trap of finding some favourite technique and then only finding the parts of problems that this technique works on. Again, creating clear questions will help you avoid this trap.

Opportunity Cost Examples

A short biography of human waste and its value | Aeon Essays

Best Plants That Are Hard to Kill

The single biggest reason why start-ups succeed | Bill Gross

Kill Technical Debt Before it Kills Your Project

Antonio Y.