First Principles About When to Use First Principles

Summary: in this post you will learn when to take the time to use first principles and the three rules for thinking.   

There are tons of people that claim you need to use first principles for all things. People I follow and greatly respect (Naval Ravikant, Shane Parrish, Julia Galef, Eric Weinstein, Scott Alexander, Peter Thiel, Elon Musk, etc.) regular promote first-principle thinking.

The problem is that you cannot use first principles to determine everything. You don’t have the time to do that. You need to rely on proxies who you believed have figured things out and believe in them (until you eventually figure out that the proxies are wrong, frauds, etc.).

For instance, I have never actually done the full proof that the world is round. I don’t actually know, with 100% certainty, the shape of the earth. I use proxies to help me determine that. It might not be round. There might be a conspiracy. Or we might be living inside a simulation. I’m not 100% sure. But I rely on proxies and make an assumption that the world is round (at least for my purposes).

I don’t know (with certainty) that the moon landing in 1969 was real. Some people believe it was faked. But I use proxies who I respect and therefore adopt the belief that the moon landing was real. I believe this even though I have not taken the 100+ hours to prove it myself.

Therefore, I believe the world is round and also believe the moon landing was real. Am I 100% certain? No. But I live life believing it and know that I will likely never take the time to prove either to myself.

So when should one go to first principles?

Yes, you do not have time to go to first principles on everything. But you cannot just blindly follow the “experts” on everything — because the experts are not always right. So when do you dive in on the data yourself?

One heuristic you could use is to go to first principles when the decision REALLY matters to you. 

Your life would not be much different if it turned out the world was flat, so you might as well believe that the world is round. 

But what about your health? Or your child’s health? 

But here too, going to first principles is very costly. Should you eat meat? Should you completely cut out sugar? These questions might take thousands of hours for you to thoroughly research via first principles. 

Of course, if it only took one hour to go to first principles figure out an important decision, it would be worth it. 

I originally misspelled this post “first principals” but then the grammar checker caught my mistake and I was able to quickly rename it “first principles.” That was a good use of time to get to the truth. Took about 3 seconds. 

But what happens if something takes 10 hours, 100 hours, 1000 hours, or even 10,000 to “know” — what should you do?

The answer (and this is incredible unsatisfying) is, “I don’t know.” But I will try to give some guidance:

Most people’s problem is not they try to go to first principles on TOO MANY things. Yes, a few people overdo it. But most people go to first principles on too few things. In fact, most people go to first principles on NOTHING. They rely on proxies for EVERYTHING. 

The first rule of thinking: at any given point in time, you should be working on something (at least one thing) via first principles thinking. It might take you a really long time to think through. It might take you the rest of your life. But having at least one thing to go deep on is a life well-lived.

The second rule of thinking: while you are likely going to rely on proxies for over 99% of what you believe, it is really important to vary your proxies. Relying on only a handful of proxies to tell you how to think is dangerous. 

By increasing the number of proxies, you are more likely that the proxies end up disagreeing with each other on important truths. 

And this comes to the third rule of thinking: make sure that many of the proxies that you use conflict with each other at least some of the time. If all the proxies that you use agree with each other, you might never have the opportunity to question your assumptions. And some (but likely less than 25%) of your proxies should be heretics in their field. 

Elon Musk on first principle thinking

What is the rate you should be scammed?

Let’s say you are investing money in something, what is the rate you want to be scammed?

You could, of course, say that rate should be zero. That you will tolerate no loss due to scams, unethical practices, etc. But that puts an extreme due diligence burden on you before you make an investment. You can’t be 100% on BOTH precision and recall. If you have fewer false positives, you will inevitably have fewer false negatives. 

Being skeptical of everything will allow you to avoid investing with Madoff, but it will also have you miss that angel investment in Facebook and Airbnb. Many ideas seem very crazy (until they aren’t). 

This is also true in life.

You can distrust every taxi driver and every construction contractor … but that might lead do you distrusting most people which could lead to a lot of unhappiness.

Or … or … or … you can accept that you will have some rate that you will be scammed.

You should have a rate you want to be scammed.

A good rate is likely 1-3% of your interactions. This can be on taxi cab drivers, investments, hires, etc. If your scam rate is under 1%, you are likely not taking enough chances. If your scam rate starts approaching 10%, you might lose all your money. 

If you never get into a car, you will never die in an auto accident. But you will also have a lot of trouble living life. So you need to have some guide-rails (wear a seat belt, don’t get in a car with a drunk (or sleepy) driver, etc.). The same is true for investing or doing anything else in life. 

Retail stores as a model.

Retail stores have figured this out. They know that people will steal things from them. Customers will steal things. Employees will steal things. Customers will abuse return policy (like people getting the big screen TV the day before the Superbowl and returning it the day after). 

Retail stores factor in shrinkage. They assume it will happen. Maybe they assume its 2%. They are essentially assuming a 2% scam rate. And yes, they COULD get that rate much closer to zero — but that comes at a large cost. They could make sure that employees don’t steal from them — but that means they need added surveillance, more bureaucracy, and they need distrust the good employees (which are the vast majority of the people that work for them). The same is true for getting the customer shrinkage rate to zero — efforts to do that might inconvenience the good customers (which are the vast majority of people that buy from them). 

Retailers could ensure that customers do not take advantage of them. But allowing for a liberal return policy also increases sales.

Retailers could ensure that items cannot be stolen from the store … but tagging all the items comes at a cost … and customers don’t like to be frisked when leaving. 

So instead retailers target a shrinkage rate. If the shrinkage rate gets too high, they are losing money. BUT … and this is the most important sentence of this post … if the shrinkage rate gets too low, retailers also lose money. If the shrinkage rate was 0.2% rather than 2%, it means they are optimizing for fraud protection rather than revenue growth.

The same thing is true in investing.  

If you never get scammed, you are likely not taking enough risks. 

Of course, if you get scammed too much, then you are also in trouble.

For instance, let’s look at bitcoin. A lot of people think it is a scam with prices pumped up by wash trading. (I don’t personally have an opinion on it). But it would certainly not be responsible to put a massive amount of your investing capital in Bitcoin — whether it is a scam or not, it is a very risky asset that has the potential to go to zero. 

But if you have been a professional investor in the last 10 years and you invested $0 in bitcoin, that also might not be great (especially if you took the time to do the work on it). The important thing is that if the price of bitcoin goes to $0 or your bitcoin gets stolen by a hacker (both of which have high probabilities), how will that affect your life? If 30% of your net worth is in bitcoin, then you’ll be hurt a lot more than if it was 1%.

We will all be scammed … a lot.

If you think you are not being scammed, you are just ignorant. Everyone gets scammed. Scams happen to all of us. And they happen a lot. Sometimes they are big scams. Sometimes they are small scams. But we all get scammed.

The simple thing to remember in investing is that you will be scammed. It will happen. The more you invest, the more you will be scammed. The more people you interact with, the more you will get scammed. The more you do, the more you will be scammed. Walmart gets scammed a lot more than the local corner store because WalMart has a lot more customers. 

The problem isn’t that someone was so stupid to put their money with Madoff. That can (and will) happen to everyone. The problem is that some people put ALL their money with Madoff. 

If you put 2% of your assets with Madoff and found out it was a Ponzi scheme you’d be super bummed but you would survive. If you put 50% of your assets with Madoff, you are in deep trouble. 

If you have to make a very large investment, then you need to do massively more due diligence. For instance, you do NOT want to get scammed by someone you marry. In the case of marriage, you want to have an extremely high likelihood of success. Summation: do a LOT of due diligence on a potential spouse.  

Beware of the macro-scam. 

One of the tough things with investing and scams is that many investments are highly correlated with each other. For instance, you might believe that the entire last ten years is a bubble propped up by the scam of quantitative easing. If that is the case, many of one’s investments (from stocks, bonds, real estate, etc.) could come crashing down. So if the scam is a meta scam (i.e. bubble) then you might have a much higher percentage of your assets vulnerable if the scam is eventually uncovered (the bubble pops). Summation: it is significantly harder to avoid macro scams than micro scams. 

Advice on what do to after being scammed.

When you do get scammed, here are a three pieces of advice:

1. Move on. Don’t stew on it, just move it. Salvage what you can (and take the tax write off). 

2. Warn everyone you know. If you were scammed by somebody or something, you have a moral duty to warn everyone you know. If it was a person, you have the duty to tell the authorities and to do everything possible to significantly reduce the scammer’s reputation so that they are not in a position to scam others again. Just walking away is immoral (in my opinion) because it means the number of victims will increase.

3. Never, ever, interact with this person again. “Fool me once, shame on you. Fool me twice, shame on me.” If you are scammed by a person, never let them back into your life for any reason ever again. Yes, people do change and people do get reformed … but the psychological impact of getting scammed twice is too great to take the risk. That does not mean you cannot eventually forgive the scammer. You can forgive (I have tried, and failed, to forgive). But even if you forgive, it does not mean you need to interact with the person. 

What is the rate that scams should take place in society?

Maybe a better question is what is the ideal rate of corruption in society? What is the crime rate? It is not zero. Zero would require the panopticon looking over you and watching your every move. 

Again, likely an ideal society scam rate is in the 1-3% range. Where 1-3% of your interactions are scams. Most taxi rides are above-board. But a few taxi drivers are figuratively going to take you for a ride. If you are taking a taxi in Beirut, the number might be more like 80% which can be incredibly frustrating as every ride becomes and a negotiation. If it is closer to 1% (or maybe as high as 5%), then you can be more relaxed.

Certain industries are more likely going to take people for a ride than others. Those industries are generally ones where there are highly unscrupulous people that go into them (illegal drug dealers) or ones that have high information asymmetry (healthcare, construction, etc.).

What do you do if the decision IS really important?

What do you do if you or a loved one get a serious disease? You don’t want to be scammed by the system. This is the case that makes sense for you to do lots of research on every provider. Don’t just take someone’s recommendation. Do lots of reference checks on people. Ask very probing questions. If you can afford it, hire a private investigator. If you can afford it, hire an advocate.

Anyone that has actually gone through life accepting a scam rate has saved up thousands of hours over the years. This is your chance to deploy those time savings to make sure you make the right decision.

Summation: have an internal base rate that you want to be scammed.

In marketing, social proof is king, queen, and emperor

Much of marketing is social proof. You use products because you see other people that you admire using products. This is especially true in B2B marketing.

Social proof, when it works well, is a feedback loop. Actions create evidence which create relevance and then create consequences.

This is true in products you buy personally and products you buy for your business. It is true for homes, schools, medical procedures, and even political candidates. Social proof is the number one thing that convinces you to choose any product that is out there.

If you are a marketer, you need to acknowledge the power of social proof and use it to your advantage.

Social proof is a very good short-cut for people who are doing due diligence of a product. They want to understand who else is using a product and what they think of it.

In marketing, social proof is king, queen, and emperor.

Social proof looms large in B2B software

Is Stripe the best payment processor for your business? I have no idea. But I do know that many companies I respect use Stripe and like Stripe … so if I need a payment processor I am going to check out Stripe. I’ll likely check out Square too because many companies I respect use them.

I’m going to assume a LOT about Stripe and Square because of the social proof. I am going to assume they are reputable. I am going to assume they are technically proficient. I am going to assume they have good fraud protection. I am going to assume tons of positive things about these services (which might not actually be true) because of the vast social proof surrounding them.

Review sites and social proof

One of the reasons reviews are so important is because of social proof. Here the raw number of review is key for social proof. That charger review on Amazon, the database review on G2 Crowd, or the company culture review on Glassdoor. And marketers spend a lot of time managing these reviews and sometimes gaming these reviews.

Of course, when these reviews are too easily gamed, they are massively discounted.

Celebrity endorsements are so powerful.

This is why celebrity endorsements are so powerful. You might admire Denzel Washington and if he endorses Toyota, that might make you more likely to get a Toyota. Of course, if you know the endorser was paid for the endorsement you might discount the recommendation … but even then the endorsement can be powerful (because you know Denzel Washington has standards and does not just endorse every product).

One of the hottest things right now in marketing is influencer marketing (paying a popular YouTuber to talk-up your product). The more genuine the endorsement is perceived, the more likely you are engage with the product.

I particularly like the endorsements that Malcolm Gladwell makes on his podcast, Revisionist History. He talks so eloquently about his love for ZipRecruiter. I have no idea if ZipRecruiter is a good or bad product, but Malcolm Gladwell’s endorsement (even though I know it was a straight-up paid endorsement transaction) makes me want to check it out.

Picking a school to send your kids to

People that have achieved a decent level of wealth have a massive number of schools to choose from for their kids. San Francisco alone has hundreds of schools (public and private) — and many of them are extremely highly rated. Which one should you choose for your kids?

Most people use social proof. They look to people they admire and see where those people sent their kids … and then they make their choice accordingly.

It would take too long to do deep due diligence on 200 schools. In fact, it takes too long to do deep due diligence on even one school. A much easier short cut is to look to social proof.

This is why recruiting at start-ups is so hard

If you a start-up, by definition very few people have joined your company. That means most candidates will not know anyone that has joined your company (or even interviewed with your company). So getting social proof is rather difficult). Larger companies like Google, Goldman Sachs, and McKinsey don’t have this problem — most top candidates will have known many people that have joined these organizations.

Summation: in marketing, focus first on social proof.

Feedback loop in social proof marketing

This was originally posted on Quora.

How do I make over $200,000 / year?

If you are not already making $200,000 compensation in your job, there are five steps to getting you there.


(1) Do everything you say you are going to do.

(2) Manage your boss and colleagues — don’t make them spend time managing you.

(3) Proactively help the organization.

(4) Be positive (don’t complain). Be a “yes, and” person.

(5) Report to someone making over $200k.

Even if your goal is not money, following these steps (save the fifth one) will help you achieve success in any organization you are in (including teaching in a school, being a soldier in the military, being a firefighter, working at a non-profit, and more).

100% of 10Xers do the first four things. Or maybe it is 98%. And these are things ANYONE can do — you do not need to have some sort of superhuman skill to achieve the first four things. If you do these things well, you will likely be a 10Xer to your organization.

(1) Do everything you say you are going to do.

One of the rarest things to do in the work world (and this is also true in the social world) is simply to do what you say will do. Be dependable. When you say you will do something, you do it. You meet expectations. Almost nobody does this. Just doing this one step puts you in the top 10% of employees.

Of course, that does not mean that you will never miss deadlines. That does not mean that you will never fail. But when you do miss a deadline, it means that you are on top of it and let the other stakeholders know ahead of time (well before the deadline). “I told you the report would be done on Tuesday. I underestimated the research involved. It will instead be done by Thursday (two days later). Let me know if this is not ok or you want to dive in more.”

(2) Manage your boss and colleagues — don’t make them spend time managing you.

If your boss or your colleagues are managing you, that is a bad sign. You need to manage them. You need to be asking them for help in achieving your goal. “Sue, please send this pre-written email to Customer Y to help my BD deal get unblocked.”

You also need to be telling your boss the things you should do. That means you should be setting the one-on-one agenda with your boss and also coming up with the ideas of things you can do to help the organization. Don’t put the burden on your boss to come up with things for you to do. You should instead have weekly, monthly, quarterly, (and sometimes yearly) project plan that you can go over with your boss and get her advice, input, and mentorship on it.

In fact, don’t think of your boss as a boss that gives you direction. Think of her as a mentor who can help guide you.

(3) Proactively help the organization.

The best employees take action. They are extremely action-oriented. If you see a problem, fix it. If you see an opportunity, go grab it. If it involves getting resources, understand the organization’s trade-offs. Do not wait to be assigned something. Do not wait to get the core priority from the big boss. Ask for forgiveness, not permission.

Determine what needs to be done to help the organization and set things in motion to achieve the goals. And this need not be the defining thing that propels an organization forward. It can be a really small thing (like ordering extra toilet paper for the bathrooms). Look to solve problems.

(4) Be positive (don’t complain). Be a “yes, and” person.

Most smart people are “no, but” people. They quickly identify problems. They are really good at finding flaws in other people’s arguments. They see the bugs in the code. They see all the inherent inconsistencies in the marketing plan. They see the product limitations.

The things the “no, but” people identify are really important. There are important flaws in every organization. But too many Cassandras are not helpful in organizations. You need to take what you have and make it better.

Better to be a “yes, and” person. Instead of shooting down other people’s ideas, build on the ideas. Give them positive feedback. Take what you like about an idea and build on it. If someone has an idea for the organization, it means they are putting themselves out there. Especially if the person is proactive (see #3 above). No need to stomp on that idea. No need to quash enthusiasm.

Instead, be “yes, and.” (note: “yes, and” comes from improv comedy where the rule is to always agree, and add something to the discussion). Take the facets of an action and continue to build on it. Be enthusiastic.

Negative-Nates are Debbie-Downers. Negativity is an infectious disease — it will infect many others in the organization. One needs an extra dose of positivity to combat against those that are negative. And this is particularly important if you are in a high-performance organization that recruits a lot of high-IQ people. Because high-IQ people are really good at seeing the problems (and they loudly complain about them).

(5) Report to someone making over $200k.

And if your goal is not just to impact the organization, but also to have high compensation, reporting to someone making over $X is very important if you want to make over $X. Of course, that does not mean you need to report to a super-well-paid person today. If you do the first four things, you will immediately be in the top 0.3% of desired employees in the world.

Note: this post originally appeared on Quora.

The unintended consequences of rising stock prices – decreased risk taking among employees

Rapid raise in stock prices result in some people in the company being overpaid.  This can be very bad for the overpaid employee and also very bad for the company.

Many tech companies are going public right now and many tech companies have seen significant share price increases in recent years.  We can expect that most of these are facing real internal motivational challenges that could be extremely hard to overcome.  
The weirdness of RSUs in public companies

Let’s say that a company gives you an offer of $100k salary and $500k in RSUs vested over 5 years.  That essentially means that the company values you at $200k per year (as stock and salary are fairly fungible in public companies).  

Let’s say the stock goes up by 20% after six months.  The RSU grant (over 5 years) is $600k and your yearly comp goes from $200k to $220k (a 10% increase).  No big deal for the company as you are probably worth more than 10% more than what they originally offered you because you now have been at the company for 6 months, understand the processes there, have grown your skills, etc.

But now let’s look what happens when they stock goes up by 300% after 3 years (which happens in the tech world).  Now the original grant of $500k is now $2 million (over 5 years).  So the stock alone is $400k per year.  Add in the salary (with assuming some raises is now $150k/year) and you pulling in $550k per year.  

This is when things get a bit hairy.  Because likely the company only values you at $350k so you are making $200k more than you are worth. In fact, if you quit the company and went to work for its top competitor, you might have a hard time getting more $300k.

So now both you and the company are in a bind.

First, you cannot leave the company.  In fact, your biggest fear is getting fired.  So unless you are a big risk-taker, you are likely to “play it safe.”  You are going to work on things that have a high chance of success.  You are not going to rock the boat.  You are not going to stick out your neck.  So you are likely going to spend the next two years doing very average work, not push the envelop, not ruffle any feathers.

And if you are in this position, it means that many people in the company are in this position (because the stock went up 300%) so that means many people are going to start acting just like you. And that means things get ultra political (because people are going to try to point blame away from them).  

From the company’s perspective, they are in a bind.  They can’t fire all the overvalued employees because then no one would join their company in the future (as the promise of the stock going up is one of the reasons that high-performers join companies).  But if they do nothing, they lead to a paralysis in their organization.

Of course, when your five year vesting is over, you are 100% going to leave the company.  Because your new RSU grant is going to be a lot lower that the value of the current grant.  You might go from getting $400k/year in stock to getting $250k/year in stock.  And even though the new lower comp is still higher than your market worth, it is very hard for the ego to take a lower comp from the same company.
This presents lots and lots of problems.  

From the company’s perspective, it is overpaying tons of people and those exact people are performing worse than the fairly paid employees (because the overpaid employees are “playing it safe”).  

So the people who the company is paying the most are often just propping up the status quo and not moving the needle.

For the overpaid employees’ perspective, it might be worse.  Sure, they are making a lot more money than they are worth.  But they are settling.  They are atrophying.  They are made to be just mediocre employees and feel that they are preventing from reaching their full potential.  And they could be in this state for 2-3 years — which is a really long time.

Summation: The companies with the fastest rising stock prices will also have a high number of employees who will want to play-it-safe (because their biggest fear is getting fired). 

Everything I know about data companies (in 30 minutes) and one viral tweet-storm

Over the last year I have been steadily putting together everything I know about running (and investing in) data companies. These companies are trendily known as DaaS (Data-as-a-Service).

I posted it to the SafeGraph blog:

It is long (will take a good 30 minutes to read) but there is also a summary tweet-storm.

The tweet-storm about the DaaS Bible went viral so including it here:

How do you determine the best business to start?

So you want to start a business eh?
“Yes,” you say. “It has always been my dream to start a business. I just can’t figure out what to start.”

The answer is right in front of you. Literally. It is in this post. (just keep reading)…

The best business to start is to figure out the join of:
1. Something that will be very valuable in the future.
2. Something most smart people do not think will be very valuable in the future.
3. Something you have a real advantage doing.

If you find a business that fits all three criteria, you have a very good chance of building a massive business.

Of course, it is really hard to know what will be valuable in the future (criteria #1). As the future is very hard to predict.

But the good news is, it is much easier to figure out criteria #2 (something people do not think will be valuable) and criteria #3 (something you have a unique advantage in) … if you are honest with yourself. If you can get those two things right, you have a real shot on changing the world.

How do you tell a great business from just a good business? (spoiler: key metric: acquiring new customers gets easier over time)

When you are evaluating a business (to invest in or join), one simple heuristic is to understand how easy is it for the business to get new customers.

In B2B businesses, the metric that companies track is CAC (Customer Acquisition Cost). But this metric in itself isn’t that interesting and companies typically track LTV/CAC ratio where LTV is the LifeTime Value of customers. The problem with this ratio is that many companies are constantly focusing on the numerator rather than on the denominator.

The cost of acquiring the next marginal customer should be less than the cost of acquiring the last customer. And you should see this cost decline over time.

The CAC itself should decline each month. If it does, it means you likely have a great business. If it doesn’t, the business is a good business at best.

Of course, CACs should be declining for a specific cohort of customer. If your business was only focused on small businesses and now you are selling to enterprises, your CAC will increase dramatically. In this case, the key thing is to track the CACs for SMBs and enterprise customers separately (with is why so many firms use the LTV/CAC ration to simplify this step).

The best way CACs will decrease over time is if you haver some sort of network effect. LiveRamp (my last company) is a middleware company … which means it is essentially a marketplace of buyers and integration partners. It is a classic network effect business that makes it easier and easier to acquire new customers over time. Once we hit about $10 million, the CACs started dropping fast.

One other way to think about this when selling to enterprise is to track the quota for a full ramped sales rep. Is the quota for an average sales rep going up over time? If so, you have a great business. If not, the business still has some work to get to great.

All platforms follow this logic. Companies like Plaid, Segment, Marqeta, LiveRamp, and Carta are classic platforms where acquiring new customers gets cheaper over time (disclaimer: I’m either an investor or friends with the CEOs of all these companies). These types of companies can take the savings (from not having to invest as much in sales and marketing) and put them into the product. So the product can get better and better over time (which is the double-edge flywheel that all great companies have).

Other companies that have declining CACs are ones with great brands. Essentially every time a company buys their service (and raves about it), other companies are more likely to use it. Twilio and Stripe have declining CACs because they have become the default go-to companies in their space. There is a LOT of power in being the default.

Summation: Once a business gets over $10M ARR and it has declining CACs, it has the makings of a great business.

What VCs really means when they say “your TAM is not big enough”

When venture capitalists tell you “your TAM is not big enough” what they are really saying is “I don’t think your team is smart enough to move to an adjacent market once you dominate your initial niche.”

They are not really saying your TAM is too small. Great VCs invest in companies with small TAMs all the time. They might believe that the founders’ think too small or that the founders just are not very good.

Many great companies started in markets where the TAM (Total Addressable Market) is small. In some cases, the companies under-estimated the TAM (the TAM got way bigger over time). In other cases, the team was smart enough to move to other adjacent markets.

Of course, it is really hard for a venture capitalist to tell a founder “we do not think you are talented.”

Even when a VC truly believes that, they can never actually say it. But founders want to know why a VC is passing and the VC wants to preserve some optionality to invest in the founder in the future (in case the VC’s assessment about the founder was wrong) or in the founder’s friends. So VCs come up with another reason not to invest. A good one is that the TAM is small — that usually satisfies the entrepreneur (who thinks the VC is just not smart enough to see the bigger picture) and satisfies the VC (who wants to preserve the relationship).

Investing in companies that are initially focused on a smaller niche is actually easier to do than investing in companies that are going after a giant market.

If the company is going after a giant market, then there is usually massive competition in the market and you really have to spend a great deal of time understanding the market (and each competitor) before investing.

For instance, if we were thinking of investing in Ford Motor Company (which competes in the giant market of automobile sales), we need to understand a ton of things:

  • What are the future of of cars? Will demand increase in the short term? What about the long-term?
  • How does the rise of places like China and India change the demand curve for autos? Even if it greatly increases the demand for cars, will Ford be able to capitalize on it? What about auto tariffs?
  • Ford makes much of its income on selling trucks (like the F150 — one of the most amazing vehicles). How does the demand curve for trucks change in the future?
  • What will happen to emissions policies? Is Ford investing in enough green vehicles to take advantage of potential policies?
  • While Ford is a big company, its market share in the auto industry is really low (because there is SO MUCH competition). So now we need to know about ALL the other car companies (and even potential car companies like Apple) to understand the future competitive dynnamics.
  • And many, many more things (like the financial profile of Ford, its labor contracts, its capitalization structure, and more).

It is much simpler to invest in smaller businesses that are tackling a smaller niche. We can get our head around the niche faster. We can assess the competition faster.

The essential questions we need to answer when investing in a niche business are just four:

  1. Will this company be able to dominate the niche? Sometimes the company is already dominating the niche. Sometimes there is a network effect reason to dominate the niche.
  2. Is this niche more important than other people realize? Maybe most people think the niche caps out at $50M/year in revenues but you believe it is five times bigger. Sometimes the niche gives the business a jumping off point to other niches because of its centrality. In general, niches that are more central (have more adjacent niches) are more valuable than niches that are less connected.
  3. Is the team capable enough to move to adjacent niches once it dominates its first niche? Some teams find themselves in a good position but cannot take advantage of their position. This is actually why most VCs pass on companies. Of course, they cannot tell the founders that they passed because they do not think the founders are smart enough. So they make up another reason (the “market is not big enough”) which is just code for “we do not think you have an excellent team.”
  4. Is the price of the investment reasonable? This one is hard to understand but if the first three are yes and only a few investors think they are all yeses, then the price is probably reasonable.

LiveRamp’s niche dynamics: dominating onboarding

One interesting example is LiveRamp (NYSE:RAMP). (note: I was the founder and CEO of LiveRamp for its first 9 years … so I am incredibly biased). LiveRamp launched its initial product at the end of 2010 going after the “onbooarding” niche. At the time, the market was less than $3 million worldwide! (Now that is a really small niche).

LiveRamp’s first year revenues from on boarding was $1 million and we ended the year with about 25% market share. But there were a few things that made the niche interesting:

  1. We believed the niche was a total of $50M year. (Turned out we underestimate the niche by 4-6 times). So there was room to grow.
  2. We thought there were network effects in the business — it made sense (for a bunch of reasons we will not go into now) for one company to be the winner — essentially it was a winner-take-most market. Ultimately we were proven correct as LiveRamp quickly got to over 70% market share.
  3. We understand the capabilities of all the competitors and figured that they would not invest appropriately to dominate the onboarding market. Each competitor was already in many other markets and it did not make sense for them to continue their investment.
  4. We believed that onboarding, while a small niche, had significant centrality to other markets in the marketing ecosystem. We assumed we could use our position to move into those other niches. This ultimately turned out to be true in some cases and more difficult than we hoped in our cases.
  5. We had a lot of confidence in our team. Even today, almost nine years later, LiveRamp is known for having an extraodinarily talented team. Of course, most start-ups think they have a great team (and many overvalue their talent). But in 2010 our team was extremely young and inexperienced — so one could forgive an outside investor for undervaluing it.

Carta’s niche dynamics: dominating cap table management for start-ups

Carta, formerly known as eShares, is a great company. (another disclaimer: I am an investor in Carta and also a customer across many businesses).

Carta helps companies manage their capitalization table. If you have invested in a bunch of start-ups, you almost surely have gotten some of your stock certificates via Carta. In fact, of the 130+ start-ups I have invested in, Carta is the ONLY forward-facing cap table management system that I have ever interacted with (except for mergers and acquisitions where I have seen many different systems).

Carta, even from its early days, dominated the cap-table management for start-ups. And yes, it was a small niche (one that many VCs underestimated). But even today, most start-ups run their cap table on Excel — so there is still a lot of growth in the niche.

If you were an investor when Carta was starting, the first thing to understand was do you think Carta could dominate its niche. Surprisingly, many investors that passed on investing actually thought Carta WOULD dominate its niche. Given Carta’s huge current success, the investors either made one of two errors:

  1. They underestimated the power of owning the niche of cap table management in start-ups.
  2. They underestimated the talent of Carta’s team and its CEO (Henry Ward).

My belief is that any VC that passed for Error #1 should stop being a professional investor. That is not a good mistake to make.

However, my guess is that the vast majority of VCs made Error #2. That error is much easier to make as it is extremely difficult to evaluate people (especially after just spending a few hours with someone). Henry Ward has turned out to be an excellent CEO. But everyone has vastly underestimated people before. And everyone has vastly overestimated people before.

What you should do when a venture capitalist tells you that your TAM isn’t big enough

Obviously you should spend time evaluating the TAM. But you should also take solace that many, many great businesses (from AirBNB to Zoom) were passed by talented VCs who underestimated the team.

Summation: When VCs tell you “your TAM is not big enough” what they are really saying is “I don’t think your team is smart enough to move to an adjacent market once you dominate your initial niche.”

What would happen if all job offers had to be quoted post-tax and in PPP-adjusted dollars? (Thought Experiment on Compensation)

Today when you get a job offer from a company (in the U.S.), your salary is quoted in dollars. If you get a job offer for $100,000 and one for $120,000, you can easily compare the compensation levels.

But can you?

It is true that comparing the salary for two jobs in San Francisco is relatively straight-forward. But what if you had to compare a job offer in San Francisco with one in Plano, TX?

A $100,000 salary in Plano goes a lot further than a $120,000 salary in San Francisco. In fact, it probably goes further than a $200,000 salary in San Francisco.

Even within countries (like the U.S.), different regions are higher-cost and others are lower cost.

The Purchasing Power Parity (PPP) between cities can be drastically different. One simple way to see the difference is the price per square foot to purchase a home. The median price per square foot for a home in the U.S. is $123 but in San Francisco it is $810 (and Detroit is just $24). So it takes a lot more dough to live in San Francisco (or Manhattan) than most places.

Also, the tax rates can change substantially between regions and cities. The top income tax rate in California is 13.3%. The top income tax rate in the State of Washington is zero (the seven states with zero income tax are Alaska, Florida, Nevada, South Dakota, Texas, Washington, and Wyoming). So even though Seattle is getting really expensive, you can save a lot of money by taking a job there instead of San Francisco.

So what would happen in there was a law that stated that all salaries need to be quoted in post-tax PPP-adjusted dollars?

Imagine that there was a law that forced every employer to quote both the absolute salary (like $120,000 in SF) and the after-tax PPP-adjusted salary (would transform to probably $50,000 in SF).

What would happen?

First thing that would happen is that many fewer people would want to work in places like San Francisco and New York City. While everyone intuitive knows that these places are high-tax and high-price, seeing the stark different on the job offer would make a significant number of people pause before taking a job.

The second think thing that would happen is that many employers would need to react to this. One way to react is to increase salaries. But the salaries in places like San Francisco are already much higher than most places and would likely need to go up another 50%+ to compensate. The more likely reaction is for employers to hire more people outside high-tax and high-PPP areas.

Long term, more people are going to start thinking about their income in “real” dollars — which means the dollars they have left over after living their life.

Photo by Pixabay on

the power of even a slightly better product. (or why do people use Zoom instead of Google Hangouts?)

Old lessons die hard.Everyone of a certain age has heard the VHS verses betamax tale.

VHS was an inferior technology to betamax but it won out due to marketing, etc. After hearing enough of these tales, one starts to wonder how important a better product actually is. Is it all about marketing? That was the moral of the VHS story.

Turns out a better product … even a slightly better product … is REALLY important.

One interesting case study is Zoom — the videoconferencing solution. Now let me put my cards out there: I use Zoom at least once a day. SafeGraph uses Zoom (and Zoom rooms). I like Zoom and would recommend Zoom. And we pay for Zoom (it isn’t free).

Why does one pay for Zoom?

Well, you might say that you need a videoconferencing solution, you evaluated the market, and choose Zoom. Maybe Zoom is more expensive than its competitors but it is the best so it is worth paying for.

The problem with that logic is that one of Zoom’s most feature-filled competitors is Google Hangouts. And Google Hangouts is “free” if you are already a Google Apps customer (which 99% of technology start-ups are).

So there is a choice to be made.
Google Hangouts which is a very good product and is effectively free.
Zoom which is a better product (but not massively better) and is also pretty expensive.

Tons of companies need to make this choice. A lot of them have chosen to go with Zoom (as evidenced that Zoom is one of the fastest growing B2B companies). Why is this?

Of course, from a customer’s perspective, free is much preferred than paid. My company chose to use Google Drive rather than Box or Dropbox because we thought Google Drive was pretty good and did not think Dropbox or Box was enough of an improvement to justify their very high enterprise cost.

So for video conferencing, why don’t people choose Google Hangouts over Zoom?

First off, to state the obvious, Zoom is actually better than Google Hangouts on almost every dimension (the one dimension that Hangouts is superior is that it has a better integration with Google apps: no surprise there).

So if you are choosing to go with Google Hangouts verses Skype or verses GoToMeeting or verses Webex or verses one of the other dozen video conferencing systems, choosing Hangouts (because it is free and it is very good) is a no-brainer decision.

But Zoom is just better enough that people are happy to pay for it. Well, they might not bee “happy to pay” exactly. No one loves spending money. But companies are certainly willing to pay for Zoom. Zoom Rooms is an amazing product and they have really focused on a great user experience. The Zoom video quality is really strong. The mobile experience isn’t wonderful but seems to work better than most of the competition.

One of the things that Zoom proves is that you can be extremely successful even when you have a crowded category, lots of great competition, and when even your strongest competitor is giving away the service for free.

Twenty years ago no one would think that a company like Zoom would thrive.

One of the biggest trends that is driving Zoom’s success is that companies are forgoing the full stack and buying the best-of-breed. The number of vendors the average company is buying from has increased almost 10x in the last 12 years. Companies are happy to buy from many different places … they are even happy to buy from new start-ups.

In fact, it has never been easier to sell to large companies. Large companies are open for business. They want to be sold to. They are sick of having a third-rate solution. They want to use the best product. If you can show them your product is superior, they are excited to buy.

The best product is actually starting to win. Sales and marketing and partnerships are really important (as is brand), but it is so much easier to market a great product than one that is fifth-best. Even amazing companies like Google, Microsoft, Oracle, SAP, Salesforce, etc. are struggling to get their clients to use products or features if they are deemed sub-par by the customer (even when they bundle it in for “free”).

That wasn’t true 20 years ago. In the 1990s, it was really hard to sell software to a big company for less than $3 million. You had to hire Anderson Consulting (now Accenture) to integrate the software. So big companies spent most of their money buying from a very small number of big trusted vendors. And they mostly had a fourth-best solution across their stack.

Today it is much easier to buy. The SaaS revolution has changed everything. Big companies can dip their toe in the water and start for $10,000 per yer in many cases. So even if it doesn’t work out, no one gets fired. It is a low cost option to try out the later and greatest technology.

Even the most crowded markets and even those markets dominated by amazing companies are open to new ideas, new products, and new companies.

Having the right vendors is as crucial to one’s success as having the right employees … and in the case of large companies potentially even more crucial (because it might be impossible for a large boring company to hire the best people in the world but it is still possible to get the best vendors … because a software vendor will sell to everyone).

In fact, one of the best ways to evaluate a company is looking at what vendors it has. You should have a really good idea about the sophistication of the talent, the ability to move quickly, and how fast the company can respond just by knowing which vendors it uses.

Before I invest in a large public company I personally like to review what vendors it employs (you can get the data for free on a site like Siftery). The list of vendors is essentially like a DNA snapshot — no two companies are alike … and like DNA, there are some genes that are just better than others and some genes that work with each other better.

Summation: we need to take new learnings from the old lesson that superior products lose to superior marketing. While both are important, the quality of the product ultimately trumps the quality of the marketing.

Enough better than hangouts that they are doing really well.

Truth verses Religion — four quadrants of data companies

Data companies fall in four quadrants: Truth verses Religion and Data verses Application

If you are thinking of starting a data company, you have to make a very important choice: what kind of company will you be? There are four basic types of data companies and all can be very successful … but the biggest mistake data companies make is that they try to do more than one at a time.

First let’s define the x and y axis…

Truth verses Religion

Truth companies are backward looking. They tell you what happened or when something happened or something about a person, product, or thing. The main objective of these companies is to have true data. Good examples of truth companies are a credit bureau (like Experian, Equifax, and Transunion), middleware (like LiveRamp, Segment, Improvado, and mParticle), and financial services data (like large parts of Bloomberg). These companies are usually very long on data engineers.

Religion companies predict the future. They tell you what will happen based on a set of data. The main objective of these companies is to accurately predict the future. Good examples of religion companies are credit scores (like FICO), fraud prevention (like ThreatMetrix), and measurement (like Nielsen, Market Track). These companies are usually long data scientists (and sometimes machine learning engineers).

Religion companies often purchase data from truth companies. For instance, FICO uses the data from the credit agencies as the core ingredient for its credit score.

Data verses Application

Once you have a valuable set of proprietary data, you have to choose if you will be a pure data company or if you will build an application on top of your data.

Data companies just sell data. The best way to know if you are a data company is if you have no UI or a very limited UI. Data companies sometimes sell direct to end buyers but often also sell to applications (which is why it is so important they do not become applications as you do not want to compete with your customers). Good examples of data companies are in financial services (like Yodlee, Vantiv), a pure data co-op (like Clearbit), location (like SafeGraph), wealth predictions (like Windfall Data), and others.

Applications make data sing. To really get benefit out of data, you need an application. These companies will have nice UI and more front-end engineers. Good examples are query-layers (like SecondMeasure), refined datas co-op (like Verisk and Abacus), integration layers (like Vantiv, Plaid), B2B product usage (like G2Crowd) and others.

Winners and lowers and winner-take-most markets

For a “truth” company to dominate its field, it has to be clearly better than everyone else. And “better” means its data needs to be the most true AND the market needs to believe it is the most true. In addition to truth, breadth and price are very important to dominate.

For “religion” companies, the most important factor is brand. When predicting the future, ideally you want to believe that the Nostradamus within the religion company is making accurate predictions. And while some people may dive into the Bayesian logic, most will trust the market perception. That’s why there are so many poor predictive analytics companies, because one can buy brand with money.

Series beats parallel

The biggest mistake data companies make is that they attack more than one quadrant at once. For the first $100 million in revenue, you should be focused on just one type of business.

Note: the original version of this was posted as Truth Vs. Religion: What Kind Of Data Company Are You? in AdExchanger in 2017.