An Investor’s Lesson to Entrepreneurs

Business has the power to impact society for the better. But that doesn't mean entrepreneurs can't go wrong—and investors know that. Entrepreneur and venture capitalist Doug Mellinger shares some tips for realizing meaningful change.

Investing & Diversity: The Changing Faces of Venture Capitalists

Is the venture capital industry embracing diversity in investors? Watch global venture capitalists from around the world discuss the state of things and what needs to be done for a more inclusive future.

Servant Leadership

There's more to leadership than driving a team to profit. In fact, there's a word for looking beyond self-interest to prioritize individual growth: servant leadership. Try this course for a quick breakdown of what that is, how it works, and how it can lead to organizational success.

Organizational Behavior and Leadership

Ever wonder what makes a great leader? Whether your role requires leadership or not, understanding organizational behavior is useful for your career. This course from GLOBIS Unlimited can set you on your way.

Leadership vs. Management

Leadership and management are different skills, but today’s leaders must have both. Try out this course from GLOBIS Unlimited to understand the difference, as well as when and why each skill is necessary for motivation, communication, and value.

Strategy: Creating Value Inside Your Company

Have you ever wondered why certain companies are more successful than others? The answer is strategy: internal processes that control costs, allocate resources, and create value. This course from GLOBIS Unlimited can give you the tools you need for that strategic edge.

Strategy: Understanding the External Environment

To plan strategy on any level, you need to understand your company's external environment. In fact, your level of understanding can impact hiring, budgeting, marketing, or nearly any other part of the business world. Want to learn how to do all that? This course from GLOBIS Unlimited is the perfect first step!

Using Japanese Values to Thrive in Global Business

Japanese companies have unique cultural, communication, and operational challenges. But they also have values that have led to remarkable longevity. Check out this seminar to hear how these values help earn trust from overseas head offices and develop employees.

Turnaround Leadership: The Differences Between Japan and the West

What's the best way for leaders to communicate a shift in corporate strategy? How do you even know when it's time for such a change? This course explains how Japan might have one answer, Western companies another.

Conflict Management

Conflicts in the workplace are inevitable. But they can lead to positive outcomes if they’re managed well. Check out this online course for a two-step process that can help you manage conflict successfully.

Evernote Founder: How Tech Startups Can Break through in Japan

Can startup models from Hollywood and Silicon Valley succeed anywhere? Phil Libin, cofounder and CEO of startup incubator All Turtles, explains how AI can solve everyday problems to bring products to market.

Women Empowerment: Lessons from Cartier

How can women overcome gender inequality and reach their leadership goals? Cartier Japan CEO June Miyachi shares her secret in this special course from GLOBIS Unlimited.

Marketing: Reaching Your Target

Every company works hard to get its products into the hands of customers. Are you doing everything you can to compete? In this course, you’ll find a winning formula to turn a product idea into real sales. Follow along through the fundamentals of the marketing mix and see how companies successfully bring products to market.

Marketing Mix

Seeing good products into the hands of customers is no easy task. The marketing mix can help. It's a collection of strategies and tactics companies utilize to get customers to purchase their products or services, and is an essential part of the overall marketing process.

The Principles of Negotiation

With the proper skills and attitude, anyone can become a successful negotiator.  But first, you'll need to learn the basics to prepare for, assess, and respond to offers for the best results. GLOBIS Unlimited can help.

Negotiation: Creating Value

Want to create more shared value between yourself and your negotiation opponent? Discover how cognitive bias affects the judgment of others. Try this course from GLOBIS Unlimited to master the value of negotiation.

Finding Your Life Purpose with Ikigai

Ikigai can guide you in your quest for self-discovery. Listen to Japanese brain scientist Ken Mogi explain why and how.

Maslow’s Hierarchy of Needs

Want to leverage Maslow's Hierarchy of Needs as a leader? Try this short course to see how the theory can be applied in practical work scenarios.

Confirmation Bias

We all subconsciously collect information that reinforces our preconceptions. It's natural . . . but it does lead to a kind of flawed decision-making called confirmation bias. To become more objective and impartial, check out this course from GLOBIS Unlimited!

An Investor's Lesson to Entrepreneurs

Entrepreneurs have the power to transform societies for the better. But how do you attract investors to start or grow a business? Or to sell one? Check out this seminar for the answers to these and more, straight from a master venture capitalist!

Managerial Accounting

Managerial accounting is a powerful way to measure progress, identify problems, and meet your goals. Check out this course to learn how data-backed decisions can help you run your business.

Finance Basics: 1

For a healthy mix of quantitative planning, evaluation, and management, you need solid decision-making. And finance is the secret sauce! Get the essentials of finance in this two-part course from GLOBIS Unlimited.

Basic Accounting: Financial Analysis

Want to compare your performance vs. a competitor? Or evaluate a potential vendor? Then you'll need to conduct a financial analysis. This course will teach you how to use three financial statements and evaluate financial performance in terms of profitability, efficiency, soundness, growth, and overall strength.

Career Anchors

What drives you to be good at your job?

Career anchors are based on your values, desires, motivations, and abilities. They are the immovable parts of your professional self-image that guide you throughout your career journey.

Try this short GLOBIS Unlimited course to identify which of the eight career anchors is yours!

Digital Marketing Psychology to Transform Your Business

How does digital marketing really differ from traditional marketing? How is social media changing things really? And what's going on in Asia?

Pyramid Structure

Having the pyramid structure in your communication toolkit can not only help you approach a problem, but convince others that your solution is valid. Break away from linear thinking and test your logical thinking with this course from GLOBIS Unlimited!

Leadership with Passion through Kokorozashi

The key ingredient to success? Passion.

Finding your kokorozashi will unify your passions and skills to create positive change in society. This GLOBIS Unlimited course will help you develop the values and lifelong goals you need to become a strong, passion-driven leader.

AI First Companies – Implementation and Impact

AI is changing the way companies operate. How do you structure teams to increase efficiency?

Technovate in the Era of Industry 4.0

Is Industry 4.0 is the next step of human evolution human civilization? Dr. Jorge Calvo seems to think so. Join him to learn how the past can help you set goals for an exciting future of digital innovation.

Technovate Thinking

Business leaders of tomorrow need to harness the power of technology and innovation. That means understanding algorithms and how they drive business results. Discover opportunities to make technology work for your competitive edge.

Product Life Cycle

Every product takes a natural course through the market—there's a how, when, and why customers adopt products at different stages. Check out this course from GLOBIS Unlimited to find out how a product you use every day is part of this cycle.

Logic Tree

Logical thinking is the most valuable asset any business professional can have. That's why logic trees are such a valuable tool—they can help you identify a problem, break it down, and build it back up to a solution.

MECE Principle

Using the MECE principle can help ensure you categorize without gaps or overlaps. Check out this course from GLOBIS Unlimited for a practical demonstration of how it works!

Have you got a burning question? Want insight from the experts?

Quora is a Q&A platform where users can share and gain knowledge about anything from European political history to 1960s movie-making to the latest advancements in tech. With over 300 million users, the site offers seventeen language options and is continually expanding.

Quora founder and CEO Adam D’Angelo visited GLOBIS University for a dialogue with GLOBIS President Yoshito Hori. Together, they discussed using AI to share knowledge with an international audience. D’Angelo also fielded questions from both Quora users and the audience about the Quora platform, entry into the Japanese market, and the expectations for the future.

GLOBIS president Yoshito Hori talks with Quora cofounder Adam D'Angelo
GLOBIS President Yoshito Hori (left) with Quora Cofounder & CEO Adam D’Angelo (right) | ©GLOBIS

Transcript:

Yoshito Hori:
What is the business model [of Quora]?

Adam D’Angelo:
[On the English site], we run ads. An advertiser can come to Quora, and they can pay to run an ad on a specific question or topic where they want to target people. So it looks similar to other self-service ad products.

Hori:
I see. You know, Adam started [creating tech products] in high school and college, then joined Facebook, and then quit. So [he’s a] serial entrepreneur. And right now, he has started a company which we call a unicorn. Which is a private company with an over $1 billion market gap.

And what is your goal? What do you want to achieve through Quora? How big or how far do you want to reach using Quora?

D’Angelo:
I’d like to get to the point where everyone in the world is using Quora. I think you’ve listed a lot of these different valuations. And I think people tend to fixate on those because it’s the only concrete number you can get from the outside. But we’re really focused on our mission [which is] to share and grow the world’s knowledge. We think that there’s just a very positive impact on society when there’s more access to knowledge. So we want to build a sustainable business. And while making a lot of revenue and raising money is an important step in that, what we really we want is to just greatly increase the amount of knowledge that’s available in the world.

Hori:
And what keeps you up late at night? What are your concerns now?

D’Angelo:
When we started Quora, we really focused on quality. We focused on making sure that as Quora got bigger and bigger, the quality would stay high. So the thing I kind of worry the most about and focus the most on is keeping the quality high as we get to more and more people using the product. With 200 million people every month coming to Quora (circa 2018), there’s a lot of pressure, right? And that attracts spammers to the system. It attracts people who might not actually be interested in helping other people and truly sharing knowledge. They just have something that they can gain by sharing something. It takes a lot of work to sort of stay on top of the quality control as we have a greater and greater user base.

Hori:
I’m a venture capitalist myself, and at the same time an entrepreneur. I’m quite curious: what kind of added value do you get from venture capital?

D’Angelo:
So maybe I can just talk about our series A investors of benchmark capital. They invested in 2010, and Matt Cohler is our partner from there who’s on our board. We’ve gotten an incredible amount from them over the last eight years.

The first thing I’d say is [that] just having raised money from a highly reputable investor, even independent of any work they do, having that stamp of approval is very important to things like recruiting certain employees, especially when a company is new. Employees can’t always tell for themselves how successful [a company] is going to be later. And so they look to [investments] as external signals of credibility. Especially employees who have conservative parents. The employee themselves might want to join, but they need to get their parents bought into the decision. And so having this reputable firm there, that adds a lot of value just by itself.

Then on top of that, we just got a ton of advice. They refer employees to us, they do reference checks. If we’re hiring someone new, we can ask through their network if the person worked at another benchmark company. They’ve been very helpful with later rounds of fundraising. In our later fundraising rounds, they can connect us with investors. They can vouch for us to investors. So those are some specific examples. But really, there’s just a ton that we’ve gotten from them.

Questions from Quora Users

Hori:
Okay. Very good. Well, let me go to the questions now [that] I picked up from Quora the day before yesterday and yesterday.

Why do you think that a Q&A is the best method in collecting and sharing knowledge? This is a question by Itaru Hamamoto.

D’Angelo:
This is a good question. You can think of a question as a request for knowledge. That’s kind of like an interface between one person who wants knowledge and another person who has knowledge. The question is an expression of what someone wants to know. What a question allows is someone who has a lot of knowledge in their head—[when] there’s a lot of different things that they could share, it helps them focus on one particular thing. It helps them and it gives them a sort of motivation.

We can collect 200 people who are all interested in the same question, and then we can go to someone who is an expert and say, “Hey, 200 people want to know the answer to this.” That’s a lot easier for an expert than if they had to just write a blog post out of nowhere. People run into writer’s block. They sometimes look presumptuous if they’re writing about something and it’s not clear that a lot of people want to know about it. So the question serves as basically this request for knowledge, and that’s been very important for us over time.

Hori:
The next question is from Kensuke Inoue: What do you think is the advanced format of knowledge sharing beyond Q&A?

D’Angelo:
Beyond Q&A, I think Wikipedia is a really valuable resource for the world. It’s not always perfect and it has limits, but it’s very useful to a lot of people. So if I was going to say besides Q&A, I’d probably say [Wikipedia’s] format.

Hori:
And [another question] from the same person, Kensuke Inoue: Do you think that AI will answer questions in the future?

D’Angelo:
I think that’s a long way away. A lot of the time when a human is answering a question, it’s not just that they need to be smart to be able to write the answer. It’s that they need to have specific experience in the world that they’re drawing on. The AI is not usually going to have access to the information that’s in everyone’s heads out there. But to the extent that it’s possible, we want to support that. You could imagine a future where there are answers on Quora from AI, maybe AI that Quora creates, but also that other companies create. We can aggregate that knowledge. But I think this is a long way away.

I’m much more optimistic about AI being used for helping to route questions to the right people [or] AI and humans working together than I am about AI doing the whole thing [by] itself.

Hori:
Another question from Kensuke Inoue: Why do you prohibit or discourage emojis?

D’Angelo:
In our questions, we are pretty strict about canonizing questions. So we want to make sure there’s only one version of each question on Quora. That’s how we can rally 200 people onto the same question—because we don’t spread everyone’s energy over a hundred different versions of the same question. And so, as part of that, we have a pretty strict policy on merging questions together. If they’re the same, we don’t want unnecessary formatting in questions. So that may be what they’re talking about. Then that’s the answer.

Hori:
I see. This is a question from Sakura Suzuki: Are the types of questions different depending on country or language?

In the case of Instagram, they say that in Japan, there are more photos of pets (like dogs and cats) than any other country. And at the same time, so many people like to put up photos of food. . . . What about in the case of Q&A? Is it different depending on [whether it is] Germany or the UK, the US or Japan, Italy? Are there any differences in the types of questions that people raise?

D’Angelo:
Yeah, there’s lots of differences. One of the differences is that in Italian, there’s just a lot of questions about Italy and local events in Italy. And one of the top topics on Japanese Quora is Japan.

And it’s natural, it makes sense. And maybe it’s not totally obvious if I hadn’t told you, but if there’s a political election happening, there will be questions about whatever’s going on. If there are current events, some are more relevant in some countries than others.

I think in Japan so far, we’ve done well with questions about the technology industry, startups, and programming. But we’ll see. I think a lot of our early traction in the US, as well, was in the technology industry, programming, and startups.

Hori:
Question from Satoshi Suzuki: When every [kind of] knowledge can become searchable and it can be storable in the outside memory, what kind of people will become more intellectual? Do you think the definition of an intellectual person changes depending upon the technological advancement?

D’Angelo:
I don’t know if there’s ever going to be a state where all knowledge is searchable and accumulated. I think there’s always new knowledge being formed every day as people have experiences out in the world. And so I don’t really expect a state where everything will be in computers and searchable.

I think being intellectual (intellectual as an adjective) means more than just being able to search through knowledge. I think it means that you really like knowledge. I think that’s probably something that will continue.

Questions from the Audience

Hori:
Okay. Let me go on to the questions from the floor. You can see there have been fourteen votes for this particular question: What are some of the best AI-related books that you recommend to technologically focused people and people from non-tech backgrounds, as well?

D’Angelo:
That’s really hard. I think AI is moving so fast that I don’t get a lot of my information about AI from books. Usually books are many years out of date. So for a non-tech person, I don’t have a great sense of what books to read.

I would say if you want to learn about AI, I think the way to go is to try actually building something. If you know how to program, you can just dabble in it [for] a little bit of experimentation. I think [that] would go a long way in building your intuition about what’s possible.

Hori:
Because AI itself is changing, advancing so rapidly, and books are not catching up with it, who do you respect a lot? Who do you think we should follow to know about AI [and] how it is advancing? Or who do you respect the most in AI?

D’Angelo:
I recently joined the board of OpenAI, which is an AI research lab. That’s an institution that you can follow. You can follow all the publications that they’re putting out. It may be hard to follow if you don’t have too much of a technical background, but I think some of it is relatively accessible. They have a Twitter account that you can follow.

Hori:
Okay. So [the next question has] ten votes: What was the hardest thing you ever had to do through Quora’s business course?

D’Angelo:
Let’s see. We had a period where we grew really quickly. This was in the beginning of 2011. There was this huge spike in our usage, and there was a kind of series of events where we got a bunch of press, and then we got endorsed by some famous people on Twitter. Then we [were] a trending topic on Twitter. And then we have these digest emails, [and] we fixed some problems in those that kind of reactivated a lot of existing users.

So there was a spiral of greatly increasing usage over the course of a week. This was before everyone was back from the holiday break, and we obviously had to just stop what we were doing over the holidays to just focus on making the system scale.

We were able to keep things running, but it resulted in the quality, I think, going way down as our systems at the time were just not ready to cope with this increase in scale. We had a ton of these huge backlogs of manual work that needed to be done to manage the quality. And it took months really, I think, to dig ourselves out of that.

It was a really tough time for everyone at the company from the beginning of 2011 until maybe June 2011. I think, in retrospect, we wish that we had set things up to be more scalable and ready for more growth. But then, I also think in retrospect that if you spend too much time planning for the future, you don’t know what’s going to happen. It’s really hard to know how much to plan ahead versus [how much to] just react.

Hori:
Okay. This is a question that you have been asked quite a few times in Japan: Considering that Japanese people have already been familiar with Yahoo’s Chiebukuro for years, what is your strategy to penetrate the Japanese market?

D’Angelo:
I don’t know. I’ve looked at some of the competition in Japan, but I know the US market a lot better. So let me just describe that because the analogy is pretty relevant.

Before Quora launched in English, there were a lot of other Q&A products and other forums—other places where people would share knowledge. We noticed that the ones that were small could maintain high quality, but any [platform or forum] that got really big would always end up with very low quality. My understanding is that that’s also the case in the Japanese market. And I think it’s been true in other markets outside the US, as well.

Our goal is to just focus on quality. We focus on quality using artificial intelligence and also a lot of the ways we’ve designed the system.

And so we expect that, by focusing on high quality, we’re providing a product that’s pretty differentiated from what’s out there. That’s what we’ve seen so far. It seems to be working. We keep growing, we keep getting more users, they write more questions with more answers. It gets more useful for everyone. And it seems like all the competition and other places where people are sharing knowledge are not doing well enough. It’s serving this core need that people have in order to sort of stop us from growing.

Hori:
Fifty-five questions already! I’ll be quick asking some more questions. It says: What are Quora’s plans for fake news filters, and how will it avoid bias, especially as it wants to be a knowledge-based business system?

D’Angelo:
These are really important concerns. For fake news, we don’t really focus so much on news. I think there’s a whole news industry and other products that are made for distributing news. We focus on answers to questions which tend to be more evergreen. They tend to be valuable for many years once someone writes an answer. The fact that we’re focused on this evergreen content, I think, makes some of these problems a little bit easier. One of the challenges with fake news is that the news moves so fast that there isn’t time for it to get reviewed before it gets distributed.

In our case, we have answers that are going to live on for many years. And so we have more of a chance to adapt and react if there’s a bad answer. Then, in terms of fairness and bias, we want to make sure that we have lots of users with all different kinds of perspectives using Quora. We want to make sure that our algorithms aren’t biased, as well. And so we have a few different things we do when there are problems like that cropping up. We’re not perfect, but we just kind of do our best to try to make sure we don’t have a system that [will have a] bias against people with a particular viewpoint.

Hori:
Is your AI model in English and Japanese the same or not?

D’Angelo:
In general, we try to build one model that works across all languages. The model has features in it so that it’ll know what language it is. And it can learn [that] people in a certain language have different preferences than people in other languages.

Hori:
Okay. So what do you think will take AI from the current state of AI (artificial narrow intelligence) to AGI (artificial general intelligence)?

D’Angelo:
That’s a huge question. That’s what I think most AI researchers are thinking about. I can’t say what’s going to be the “one thing,” but I think computing power is still a major limiting factor. And so as computer hardware gets faster and faster, we can train larger and larger models, and train on more and more data. I think that may not be all we need, but I think that’s one kind of prerequisite before we get to more general algorithms that really work as well as narrow algorithms,

Hori:
Any future for Quora to use blockchain technology?

D’Angelo:
I think blockchain gets a lot of hype, and it’s not always clear exactly what it means. Quora is a centralized product. And we really believe in centralization. A lot of the value in Quora is that we can use machine learning to match people with the right other people to share knowledge, and blockchains usually make sense in a distributed, decentralized world. They’re kind of a different approach to things. Usually when you’re doing AI and machine learning, you want everything centralized and these blockchain applications are for cases where there’s no centralization at all. Or you’re trained to evade some central authority. I’d say, in general, no, [but] there may be interesting things we could do with the technology right now.

There’s a lot of uncertainty around how the whole technology and legal treatment is going to be for cryptocurrencies. So depending on how all this stuff works out in the future, we may. There are some blockchain products that try to enable knowledge sharing, [such as] compensate users for sharing their knowledge. But none of them have been very successful. If some of them were more successful, then we might see that as an indicator that there’s a lot of potential there.

Hori:
What is the incentive for experts to write answers?

D’Angelo:
That’s a good question. Why do people like writing answers? Different people are motivated by different things. I’ll just try to list some different motivations that different people have.

Some people just intrinsically enjoy writing for the same reason why someone might write a diary, an email to their friend, or a blog post. Other people, they really want to help people. [It makes them] feel good. They help someone for the same reason they might be motivated to work in a job where they’re helping people, the same reason they might work for a nonprofit. And Quora provides a very efficient way to help a lot of people. For a lot of people, where their motivation is to help other people, and there’s some subject that they know a lot about, answering questions on Quora is a very efficient way to help large numbers of people. Whereas, if you’re helping people one-on-one, that’s not a very efficient process. [On] Quora you can help a hundred thousand people. So those are some of the motivations.

Other people benefit because they can build their reputation. We have lots of people who have gotten jobs because of answers they’ve written on Quora. There’s a venture capitalist, Jason Lemkin, who has publicly credited Quora [for] building up his reputation to the point where he was able to raise a big venture capital fund. [All] because of the answers he wrote on Quora that established this reputation and put him in the position to be able to do that. Other people are motivated by the feeling that they’ve learned a lot in their life and they want to give back to society. It’s not necessarily about helping a specific person, but they want to sort of repay society. We see this motivation especially in some retired people who feel that Quora is a good way to share with future generations.

But our stance as a platform is different. People have different motivations, and we want to enable and support all of them.

Hori:
Okay, Adam, I have sixty-one questions for you. Instead of me selecting questions, why don’t you go ahead and try whatever questions you feel like answering.

Audience Questions Selected by Adam D’Angelo

D’Angelo:
The question is, “How come we decided to add Japanese?” Okay, I’m going to interpret this as, “Why did we prioritize Japanese?”

So our mission is to share and grow the world’s knowledge. But most of the people in the world don’t speak English. And so it’s very important for our mission to be able to expand to other languages. Because we were so focused on quality and keeping the quality high in English was very challenging, we didn’t internationalize at all until we were about six years into building Quora. In order to fulfill our mission, we got to the point where we were confident enough in the quality in English, and we felt like we were ready to take on some more challenges. And so we internationalized.

The reason we prioritized Japanese was a few things.

  1. Japan is a big market. There’s a lot of people in Japan. I think it has a very deep culture, and there’s a lot of knowledge that has been accumulated here through hundreds of years. And being able to get that knowledge into the world is a valuable thing.
  2. Then we thought there’s a particular fit between our product and Japan. I’ve really enjoyed coming to Japan. Just as a tourist personally, I’ve been here I think over 10 times, and one of the things I’ve noticed—I’m not an expert on the culture, but it seems that people here particularly value quality. There’s a real focus on quality, and our product focuses on quality.

And so we think that there should be a good fit between Quora and the Japanese market. We’re going to learn and adapt. We don’t think we’re going to get everything perfect right away, but we do think that there should be a particular compatibility here.

Hori:
Adam came to Japan ten times. Eight times, he came as a tourist, and he came only twice as a businessperson. This Japanese version was launched only in November last year. And I was quite amazed that there’s no office in Japan yet. Some of you may become an employee of the Quora Japan office. And for those of you who speak Japanese and who read Japanese, I encourage you to download the Quora app. To ask questions or contribute by answering questions.

So what’s the next topic that you would like to go for? This is interesting because I’m curious as to what kind of question Adam will pick.

D’Angelo:
So there’s a question that’s basically asks, “What makes us different from Reddit and Stack Overflow?”

Relative to Reddit, Quora is focused on questions and answers. Reddit is more like a forum, and it focuses just on recent content. Quora focuses on building up this big database of knowledge, and Reddit focuses on community interactions. Reddit is very decentralized. So Reddit is organized into these subreddits, which can lead to great communities. But it can be kind of disorganized. There’s a lot of redundancy, and it’s very difficult to find the right subreddit. Whereas Quora is just a flat space, so it was more like Wikipedia.

Reddit is all pseudonyms. People don’t use their real names on Reddit, whereas on Quora, everyone has to use their real names. [With] a lot of different subtle choices, I think you end up with a totally different product at the end. Stack Overflow is also vertically oriented. [It] focuses on programming Q&A, and it’s really good for programming Q&A. There’s a lot of features that are specialized for that.

Quora is a horizontal platform. We don’t focus on any particular topic, and we intentionally only invest in features and work that’s going to benefit sharing knowledge about any topic.

Hori:
Interesting. Okay. The first question picked was about why he came to Japan, why he launched the Japanese market. The second was about differentiation compared to other competitors.

D’Angelo:
The next question is [about how] I said that we evaluate the quality of an answer through AI and voting, but this question says people who vote are probably not experts. So we actually specifically try to direct a new answer toward experts to see what they think of it.

Not everyone’s vote counts equally. We will pay a lot more attention to a vote from an expert than a vote from someone else. It’s almost like you can think about how an academic peer review works. If you’re going to publish an article in an academic journal, you’ll submit it and the editor will send it out to some other experts who they trust to see what they think about it as a way to sort of gauge whether it’s worth publishing. Anyone can vote on any answer, but the votes we pay the most attention to are votes from other people who actually are experts.

Hori:
So what you mean is that even though you get ten votes, nine may not be experts. So the one vote by an expert is more appreciated. How do you quantify that? How do you calculate the differences?

D’Angelo:
There are degrees. There are a lot of different machine learning algorithms. It’s hard to tell you the exact rules for what it does. But for example, if someone has a PhD in a field, then that’s a signal that maybe their vote should count more. Or if, historically, their votes have lined up with answers that we know are good, then that’s a signal that they have better judgment as someone who’s voting.

Hori:
Voting. Very interesting. Do you have any other questions that you feel like answering? Or if you don’t, we can ask other questions.

D’Angelo:
“Do I think Quora AI can replace universities or MBA schools as a physical place to share knowledge?”

Hori:
Okay, that’s a good one.

D’Angelo:
I don’t think we actually think of the education space as a different market from what we are doing. There may be online education products that could potentially threaten an MBA program in the future, but it won’t be Quora.

So, of the characteristics [that] you need for education, one is motivational aspect. Quora relies on people just wanting to spend their time reading answers naturally. Often with education, it’s going to be hard. You’re going to have to really focus, and you need a stronger level of motivation than just randomly checking something. Often, educational products or schools will have a commitment aspect where you have to [for example] pay the whole tuition up front. And so now that you’ve committed, you’re going to lose that [investment] if you don’t actually study, or you’re going to get bad grades. And you only get the degree at the end if you complete all the courses.

We don’t have any kinds of systems like that. If we added those, it would complicate things in a way that we don’t want. There’s a lot more to education than just this commitment mechanism, but that’s something that makes it a different market than sharing knowledge on the internet.

Hori:
Interesting. In the case of education, it has an aspect of meeting friends—like you met Mark Zuckerberg. Do you add any kind of community aspect to Quora so that people can meet each other?

D’Angelo:
People make friends on Quora. There’s a lot of people who I meet who say that they have met a cofounder of a business, or they’ve recruited employees, or they’ve made friends on Quora. So there is some of that. I think it’s hard to compete with real world socializing, which is what you can get in a physical school. But there is a social aspect of Quora as well.

Hori:
Do you have meetups quite often? Actual physical ones?

D’Angelo:
In English, we’re so big that it’s just kind of unwieldly to manage. So we’ll have meetups for the top contributors to Quora in English, which is a very exclusive event. It’s very hard to get to that level. But in other markets outside of English, we have more frequent meetups, and we rely a little bit more on that

Hori:
In case of all the web services and apps, they have a stickiness to them. They want the users to come back. What kind of stickiness do you have on Quora? Why do users keep coming back?

D’Angelo:
We have a lot of stickiness, especially for our knowledge-sharing product. I think we end up being a lot more sticky than other knowledge products.

A lot of it comes down to personalization. We’re very good at showing people answers that are relevant to them, that they’re going to find interesting. And that is through, again, machine learning. But people feel like when they come to Quora, we often show them something that helps them out in some way, or is interesting. I think there are some studies about why people read the news, and a lot of it doesn’t actually have much to do with the fact that it’s about current events. They want some kind of intellectual stimulation. They want to learn something new. If you look at the articles in a typical newspaper, there are actually often a lot of articles that are not related to breaking news. I think the same kind of motivation that causes people to want to read also leads a lot of people to check Quora regularly.

Hori:
How many people will come to Quora and ask about what’s going on?

D’Angelo:
People also just check Quora because, [for example], maybe someone’s really interested in cooking, and then there are new answers to read about new recipes or new cooking answers. For those people, sometimes that’s more engaging than reading about what’s going on with some war somewhere across the world. So news is not the only motivation that people have.

Work-Life Balance

Hori:
Interesting. While you search, let me ask you more questions. How long do you work every day, and what do you do besides work?

D’Angelo:
So I’d say I work pretty hard, but I think I try to keep it at a sustainable pace. I’ll go into the office usually around 9 or 10, and then I will stay until 6 or 7, and then I will also do work at home after, and I’ll go eat dinner usually. Then I’ll do work at home for—well it all depends on how much I have, but maybe another few hours. Maybe not if I’m tired. I think, over the course of a week, I am probably working like 60-65 hours a week. Which I think of as working hard, but not like killing myself. I usually try to say, on the weekend, [that I] have one day where I don’t do work.

Hori:
So what do you do on the weekends?

D’Angelo:
I like reading. I really like just using the internet. That’s why I like building internet products—[because] I like using them too.

Hori:
What do you like using on the internet?

D’Angelo:
I like Twitter, for example. I’m in Silicon Valley, and there’s a news product called Tech Meme, which is basically like a news digest. And I find that it’s a very good way to keep up with the technology trends. I do angel investing. I just sort of learn a lot about what’s going on in the industry outside of Quora.

Hori:
So that’s where you keep up with the trends and also the advancement of technology. You sit on the board of an AI company, and you do some investments. What else do you do to keep up with technology or AI? Do you still program?

D’Angelo:
[Not really.] I program occasionally. I will if I want to see if something’s possible or to test something out. But no, I don’t program serious projects. On Twitter, I follow a lot of machine learning researchers. I will read new papers when they come out. Sometimes I’ll look through different machine learning courses.

I think the other thing is I hang out with a lot of other CEOs and I see other people in Silicon valley. I think just that is just very stimulating and I get a lot of information from that. That is one of the great things about Silicon Valley: everyone is right there, and there’s this whole social scene.

Hori:
So what do you do when you hang out?

D’Angelo:
Often have dinner. Not a lot of drinking. People have parties, someone will have a housewarming party, and they’ll invite people over, maybe drink a little, but it’s mostly about socializing.

Hori:
So here’s a question: What is your motivation for what you’re doing right now?

D’Angelo:
For me, I think there’s a lot of different motivations that come together for Quora. When I was in college, as I mentioned before, I made internet products, and that was something I just did when I had time. And I just intrinsically enjoyed the work involved in making a product like this. I think it’s an interesting challenge to create and run a system like this. I really like working with other good people, and I learn a lot from the other people that I work with.

So a lot of my motivation is about working with other good people. And part of what you get out of [having] a startup is that it’s a chance to work with other good people.

And then, I also care about my impact on the world. I feel like I’ve learned a lot in my life, and I’ve benefited from all the technology that society has built up. And all the education and institutions that exist in the world. I feel like I have some responsibility to contribute back to society in some way. And so I think Quora is a way to do that. I think it’s particularly leveraged because Quora allows other people to help society by sharing their knowledge. But by focusing on creating Quora itself, that can have a much bigger impact than if I just like wrote a book myself or something like that.

Hori:
Interesting. Okay. So last two questions.

Copyright, Coaching, and Other Questions about Quora

D’Angelo:
“Who owns the copyright of the answers?”

The person who writes the answer retains the copyright. They grant us a license to show the answer to other people. But we don’t take the copyright.

“Could Quora be useful for counseling or coaching functionalities?” I think what the question means is, “Would we have a feature where you can talk to another user instead of just asking a question and then getting a one-shot answer? Could we let you get advice, like a back-and-forth dialogue, maybe with someone who would write an answer?”

The answer is no. We define the market around answers that are going to help a large number of people, or at least multiple people. If you’re building a product for one person to help a single other person, then you want a totally different kind of product. Usually those products involve the person who needs help having to pay money to get the service.

So there are some consulting products like this in the US. There’s this thing called Gerson Lehrman Group, hedge fund investors. And people who are familiar with a company’s products can use this network to get in touch with experts and then have a phone call with them. We think that’s just a different market.

The market is sharing knowledge in public so that it can benefit a lot of people and not just a single person. So it ends up being a different market. I think an interesting thing I’ve learned is that Q&A, or this network of connections between experts and non-experts, ends up being a very vaguely defined market. And in practice, there are these totally different companies that are in the space that I wouldn’t even think of as competitive at all. It turns out that that, from the outside, it looks like everything is Q&A, or there there’s a lot of similarities, but they don’t actually fit together.

Hori:
Wow. There’s lots of questions still. I will ask one or two questions and then we’d like to close. How would your life be different if you’d never met Mark Zuckerberg?

D’Angelo:
Oh, I don’t know. I wouldn’t have joined Facebook as early. I’m sure I would’ve ended up in Silicon Valley otherwise, and I probably would’ve started a company. I got a lot of valuable experience at Facebook that really helped me. I think that I still would’ve been interested in social networking. I probably would’ve gotten into social networking, but I wouldn’t have had as much experience. I probably would’ve made some worse decisions at Quora if I had [even] started Quora. I might have started a different social networking company, but who knows? I don’t know. It is really hard to say.

Hori:
Now this is a question: “What would you recommend doing to improve programming skills?”

D’Angelo:
I would say it depends on how much time you have to invest.

So one thing you can do is just take courses. I think that’s useful if the courses are a good, if you like courses, and you learn well from courses. Then that’s a good way to learn to program.

If you’re not as motivated by the academic structure, then I think just doing projects and making something that you like. So if you like games, then make some simple games for yourself. If you like internet products, then try making a very simple internet product. I think playing around and doing little experiments is very important to learning. Don’t try to make one massive project, just do a series of like 10 small projects. Then I guess I’d also say, if a programming contest appeals to you, then that’s a really efficient way to accelerate your learning.

Finding and Hiring Talent

Hori:
Okay. The last question is, “How do attract, select, and hire talent?”

D’Angelo:
I could talk for a long time about this. Silicon Valley’s an extremely competitive environment for talent. It’s very important to differentiate relative to other employers. At different stages in our history as a company, we’ve differentiated in different ways. So when we first started out, a lot of it was around scale. The fact that someone would join a small company was a lot different. We recruited people who were working at Google, for example, and they felt like they would have a lot more impact at Quora. And they were going to be a lot closer to the actual decisions getting made than they would be at Google.

As we’ve gotten bigger, we’ve focused a lot on our mission.

A lot of people feel like Quora is something that benefits the world, and maybe they like Quora themselves. So through the mission and through the product, that differentiation helps us a lot. I guess I’d say another thing is, as you start to build a strong culture and you have some really good employees, they start to be people who other people can learn from. People get drawn to the company because they know about the reputation. When people come to interview, certain times we’ve had the reputation of having the hardest technical interviews out of any of the companies in our category that we’re competing with. And so people want to come interview at Quora just because they feel proud of being able to get an offer from us, or to work with a high-performing team.

Then there’s all the standard stuff, too, that every company does. We have a recruiting team, and we are constantly reaching out to people. [As for] the retention of employees, we have good managers [who] make sure we’re making good decisions. We run engagement surveys to help get data on what we could do better from all around the company. We spend a lot of time just talking to employees, understanding what’s motivating to them. Developing good leaders is really important. We’re constantly trying to do better. And I think Silicon Valley is competitive, it can be challenging that it’s competitive because, you know, your employees, anyone who’s working at Quora, have lots of options for other places. They could easily go get a job that would pay just as well. So that presents a challenge, but it has forced us [to become] very good at people management, I think, relative to other companies our size. We’ve made a choice to get good at that, to build a strong culture and to have a strong mission.

Hori:
Great. Adam, thank you very much for coming. You know, we have learned a lot about you. I have asked lots of personal questions as well, from high school, and then college, and then Facebook and Quora, startups, the future of AI, and leadership, and also about motivation and so forth. And so I hope you’ll keep coming back to Japan. We’ll be hosting the G1 Global conference in October, so that’ll give you a good reason to come back to Tokyo. I hope you can come.

I’d also like to thank the audience here for giving us very good questions. They were very helpful. I hope that this two-hour session has been quite fruitful for all of you. So thank very much. And thanks to Adam for coming.

Connect with Insights

Trouble keeping up with all the insights? Subscribe to our newsletter for monthly career inspiration right in your inbox!
Your newsletter subscription with us is subject to the GLOBIS Privacy Policy.