Hacker News new | past | comments | ask | show | jobs | submit login
Things I Learned from Chris Dixon about Startups (2015) (a16z.com)
319 points by vinnyglennon 3 months ago | hide | past | web | favorite | 49 comments

This is interesting at a meta-level.

It's ultimately content marketing for a16z. If Chris Dixon wrote this same article—e.g. "Here are all of the best things I've said"—then it would come off as arrogant or ickily self-promotional. But having someone else write it allows the author to praise the ideas without the same stigma.

I don't know whether it was intentional or not, but it's interesting how well the content comes across when the base idea ("writing about smart things our partner was right about") is the type of marketing companies/VCs _want_ to actually say but social norms prevent.

The article itself is an example of the advice within the article. A "good idea that people think is a bad idea."

It's completely intentional. a16z is fantastic at marketing, PR, and brand building. In less than a decade, while generating very unimpressive returns for LPs, they've built one of the strong VC brands in the world. They're phenomenal at it. They do it so well that they make it look easy, but it's not at all.

This is, without a doubt, the boldest and most impressive backhanded compliment I've ever read on HN.

I have no idea what A16Z's returns for their LPs are across their funds, but this could literally be right out of a movie script.

  GENERAL PARTNER is chatting with potential LIMITED PARTNER
  over cocktails.

  LP: The fund you're raising... the thesis, it's...
      interesting. You've done a great job with your last
      three funds. But I've read similar things from
      Andreessen's team that make me think they have a
      better grasp of the space - wouldn't my money be
      better invested with them?

  GP smirks and calmly reaches towards MARTINI GLASS,
  casually picking the toothpick, olive still attached,
  between her thumb and index fingers. She places the olive
  between her teeth to the side of her mouth, grasps, and 
  pulls the olive off the toothpick in one swift motion.
  She carefully places the toothpick in a napkin, stares LP
  in the eye, and pauses just a moment before aggressively
  biting down. She finishes chewing without breaking eye

  GP: Ah. Yes, they do market themselves as visionaries.
      It's completely intentional. They're fantastic at
      marketing, PR, and brand building. In less than a
      decade, while generating very unimpressive returns,
      they've built one of the strongest VC brands in the
      world. They're phenomenal at it. They do it so well
      that they make it look easy, but it's not at all.

  LP: So if I'd like to make money...

  GP: You got it. How much are you in for?

I actually liked what you wrote. Only thing I can add is they were able to build that brand based on the legacy start power of Marc Andressen. I wonder what would have happened if he hadn't been involved.

On this, I used to work in investment management. Good distribution is several times more important than generating returns.

One of the most prominent managers local to me (worth $150-200m, probably one of the wealthiest people in my country) only has one senior manager, apart from that...just grads. The product is terrible but is designed for a certain market that is growing rapidly and is naturally doing very very well. This is true of all of the largest fund managers local to me: always distribution first, thinking about who is going to buy, and what they care about (i.e. US pension fund, superannuation, etc.).

Even worse, I have come across more than one manager who was clearly very +EV but either couldn't raise money or went out of business because they weren't selling the right product at the right time. Most good managers do stay in business but not all. And if you don't have good distribution, you won't survive no matter what returns you have.

Why? My current theory is that people have rigid ideas about how to make money. Profit doesn't matter: it is about validating an idea or opinion of themselves. I suspect that the VC world is no different. I know nothing about a16z apart from their "reputation": very influential, whizzy VC firm, and very safe intellectually. They are the archetype VC firm, this is what VC investing is, you can lose money but you won't lose face.

What’s their return rate compared to an index like the S&P500?

VCs occupy a completely different part of the risk-reward spectrum.... so I'm not sure that's an apples-to-apples comparison regardless.

I'm curious how you can say they're generating unimpressive returns for their LPs.

The author Tren Griffin has many articles like this on his website, it's his 'genre'. Not sure if this was just cross-posted or newly written: https://25iq.com/featured-individuals/

While it looks like it’s content marketing at first glance, this is a regular theme Trenn uses. He runs a wonderful blog called 25iq. He’s been privelaged enough to work alongside the best and have access to the best, and he takes a lot of time sharing the meta-insight. He structures a lot of his thinking in the same way as Charlie Munger. I’d recommend any entrepreneur read as much of his writing as you have time for!

This is a very good technique, which we point out in our video to investors:


Interesting video, It's used by confidence men. The term I think is a shill. Posing as a satisfied customer.

I’m somewhat tired of these kind of “startup folklore” articles, although I read and bookmark them all. Sure, these people were crazy successful and see many other people being successful. Still, so much of these “insights” feel like they could be exactly the other way around. Many successful founders I have met, have told me: well, no we weren’t the only ones doing this, but we got lucky for this and that reason (e.g. timing). What if the lucky ones wrote the “rules”?

I don’t know if it’s possible but I would love to see all these concepts proven by data.

There's a meta-insight there: what if the lucky ones do write the rules?

If that's the case, an optimal strategy might be to take as many convex bets as possible, keep secret all the ones that fail, and then once a low-probability bet succeeds, milk it for all it's worth and convince everyone you're a genius. You can then leverage that for significant financial gain from organizations who seek genius-level insights in the future. Eventually regression-to-the-mean makes you lose your credibility, but in the meantime you can pocket significant amounts of cash and invest them in much more stable revenue-generating assets.

A glance at who's been successful since the 3rd millenia started shows that this might not actually be a bad career strategy.

The other part that's insidious is even the most well-meaning and well-intentioned individuals oftentimes fail to understand the role of luck in their success. Some psychological studies posted here on HN support that people that are successful even for completely random reasons start to believe that they did something different from others. One of the points in the posted article here indirectly addresses this though that you should listen to those that failed too because when people fail it's usually much more clear why compared to the agency of the founders that succeeded. Furthermore, successful people tend to stick near other successful people - they are insulated oftentimes from most of the world that is normal out there. Nobody on tech blogs cares about the insights of a middle manager of a construction company in Louisiana, but there's so many businesses out there that do perfectly fine it seems odd to me how little cross-pollination seems to happen.

One of the criticisms of psychological studies historically has been that we've been studying mostly abnormal behaviors and that it'd be like trying to determine how a car is built based upon only asking repair shop auto mechanics. Similarly though, I think we need to invert this for business - we need to identify commonalities between businesses that fail more than we need to identify the factors that cause success (the entrepreneurship porn out there is potentially harmful this way despite the many articles on the dark side of it). I think the common factors in what causes failures are fairly static compared to the factors that cause success. This interview[1] is a great example of what was successful at one point in history can totally fail later (much greater planning necessary in the early 90s while now we're focused upon market fit / reactivity today as a trend). My gut feeling is that some kinds of companies are better off sticking to their guns and churning out good product independent of the market trends while others need to evolve.

1. https://torch.io/interview-series-tim-brady/

> One of the criticisms of psychological studies historically has been that we've been studying mostly abnormal behaviors

I’m not sure this is actually true. There’s tons of data from “normal” subjects, though normal here mostly means “average behavior of an 18-22 year old college student.” If anything, psychology would probably benefit from more focus on individual differences rather than commonalities.

For some more context I was paraphrasing a passage from a psychology textbook about the position psychology was in several decades ago, in particular about how languages are processed in the brain and how most of what we've learned about processing languages came from analyzing people's brains that acquired disabilities such as through strokes, accidents, etc. I do agree with you to question the the original validity of the statement in relevance to where we are now - my point is to question the status quo in business discussions of primarily studying the survivors when they're the minority for tech businesses and to ask some more fundamental questions of "are all these business principles relevant after the survivors talk about it publicly?" or is it closer to revealing winning lottery numbers after the drawing is over?

I must admit I may be missing some much more fundamentals that are discussed in academic programs for business (like data structures in CS programs) but am curious why they're treated as table stakes for merely talking about topics like market fit and growth projections when engineers do talk frequently about "basics" like essential data structures.

There's a huge amount of luck and serendipity involved in all things, including of course the outsized returns and hypergrowth that SV worships and today's generation of young social climbers pines over.

However I would caution anyone to be careful with the type of optimal strategy you outline. Diversifying within reason is good, but you have to deliver real value somewhere along the line. The luck involved in building a successful business does not occur in a vacuum, if you are not focused and prepared, the rocketship won't leave the ground (or worse explode mid-flight).

Those who are more obsessed with the financial or social outcome than the actual job of building a business are the ones you find all glad-handing each other at networking events while the most successful always seem to have (or have had) something more important to do.

It’s not just luck. It’s also the unacknowledged support of your peers. I’ve had two bosses that were process nerds to the point of wanting to write a book, and neither of them, once they tried to condense their findings down, wanted to credit the teamwork aspects and other human factors (I’m big on human factors).

When you like people, or just your team, you plaster over their mistakes. If you try to reduce the whole thing to something cynical, people just want to watch you crash and burn. Which is I think where you see the problem with reproducing results. You get a certain number of people who just want to turn the screws another two clicks instead of creating a better work environment where more gets done.

In my 6 years in the startup industry, what I have found is that luck plays a huge part in being successful. A lot of people that are successful are successful because they caught the market wave. Not because they did something different or were better than other entrepreneurs. Some are truthful about it, but many aren’t. And they always find ways to rationalise their success. It similar to people explaining technical analysis of market graphs. They will provide lots of reasons based on patterns, but it’s all just mumble jumbo.

You are right and most people don’t know or don’t acknowledge this. Check Bill Gross Ted talk on the number one reason startups succeed.

I shared my own personal experience in this article https://thenextweb.com/contributors/2018/12/08/this-is-how-i...

I'm so glad you mentioned this.

Here's the link for anyone interested: https://www.ted.com/talks/bill_gross_the_single_biggest_reas...

The established truth is that “Chance favors the prepared mind.“

Yeah, all I can think of when reading these pieces is survivorship bias.

> Many successful founders I have met, have told me: well, no we weren’t the only ones doing this, but we got lucky for this and that reason (e.g. timing).

I would go even further, ideas are a dime a dozen but execution is what matters--simply staying alive is the biggest issue.

"Timing" is simply staying alive long enough that you are actually present when the correct timing occurs.

Yep, reminds me of Survivor Bias

This echos a lot of hard lessons that I had during my time as a founder. While a lot of this is great advice, I'm frustrated by the advice of "if people reject you and things are hard, then things are going well". Because I believe there's more depth to people's rejection of your idea/product/business and they could have valid reasons.

My real issue is how to weed out the legitimate feedback from the bad. No one has ever told me how to identify the difference. It's typically, "yay" means you've done well and "nay" also means you're doing well. When really, either of those responses could be a false positive - but which?

Having been through this a few times, my personal thinking now is:

"I like it" means absolutely nothing by itself. People will tell you they like your idea/demo/whatever all day and have absolutely no intention of ever using it, let alone paying money for it. What you want to hear is "How much does it cost and when can I buy it?" (assuming b2b) or "Can I be a beta tester?" (assuming b2c). Everyone else is fluff. If you're getting nothing but fluff after talking to a bunch of people, your idea is probably also fluff.

"No" by itself also means pretty much nothing. You want to be listening for "No because..." and then evaluate the because part. Is it no because they're not the right audience? Is it no because they don't see the market opportunity like you do? Is it no because they've already heard the same thing from 3 other entrepreneurs and yours is thus a me-too play? Is it no because [fundamental reason you didn't think of and have no answer to]? Etc.

Excellent point. When I'm talking with a first-time founder, they'll ask me what I think of their idea. Sometimes I'll say straight out: you shouldn't ask me that question, because if I answer you shouldn't listen to me.

Feedback from people in the target market is hugely valuable, especially if they answer in individual terms. E.g., I will/won't buy because X. But feedback from people not in the target market is rarely useful, especially when it's in theoretical terms. So unless I have some specific expertise, I try to redirect people toward the people they should actually be talking with.

Just finished reading a book called "The Mom Test" by Rob Fitzpatrick. The central idea of the book is to get you to ask the right questions, because if you ask people questions like "how am I doing", friendly people will tell you "yay" because they don't want to hurt your feelings, while assholes who want to appear like gurus will tell you "nay" to bolster their own notions of being a startup soothsayer.

Highly recommend it. It's short (yay!) and very practical and changes the way you think about questioning people.

No affiliation, just like the book

I suggest that we can tell the difference, evaluate ideas, as they have long been evaluated with good accuracy from the pyramids to the latest fantastic success of the US DoD.

> Chris Dixon is saying that the people most likely to know the “a secret” about a business opportunity are people who have deep domain expertise

Which unicorns were started by people with deep domain knowledge? Struggling to think of any, but that may be a lack of imagination.


I guess Fastly is unicorny, and also counts. Does seem to be the exception not the rule though.

The thoughts about ideas and information are particularly relevant.

If you're not a domain expert everything you come up with will collapse under efficiency arguments. Such as "if this made sense someone would have done it".

Much of expertise is knowing what experts themselves think. That's why you can answer the question of why it isn't already there.

I have a question though. If you need to be a domain expert, how can there be so much focus on young founders? Is that a different category of investment, people who happen to have a lead into an emerging industry like social networks?

You really need to realize there are different markets.

Young founders are often in B2C and capital-light. They are judged (almost) purely on growth metrics. Series A negotiation can drag on for months, and its hard to fake 100% month-on-month growth.

The B2B will usually be more experienced insiders with deep domain knowledge, and excellent references.

EDIT: everything is relative. If you are a brilliant PHD in ML then you can be 25 AND a domain expert AND go in to a capital-heavy B2B sector.

For many consumer ideas, the younger people are the experts.

I guess because VC is, as the article says, all about providing smart money. Old founders have their network so VCs are focused on young founders.

There seems to literally be a bottomless pit of content-free aphorisms that spew out of the content marketing departments of venture capitalists every year.

Like does anyone read this stuff?

“[The] business of seed investing, and frankly, early-stage entrepreneurship, is so much about getting good information. And almost all of that information, unfortunately, is not published.”

OK cool, thanks. By the way also that's also true of all investing, in anything, ever throughout history. By definition good investment information is unique information, that's the nature of competitive markets.

“Ideas …matter, just not in the narrow sense in which startup ideas are popularly defined. Good startup ideas are well developed, multi-year plans that contemplate many possible paths according to how the world changes.”

Ideas matter. More thoughtful, well developed ideas matter more. Who knew? Another thing that matters is warm clothing, it's really handy.

“What the smartest people do on the weekend is what everyone else will do during the week in ten years.”

In ten years everybody will be spending time with their family, apparently.

This era of technology, it seems to be the core theme is about moving beyond bits to atoms.

Thank god, the whole sub-etha thing has been getting pretty tired.

And so on. This article is almost completely devoid of actual analysis or observation, it reads like a fugue on someone's stream of consciousness ramblings

I actually found a lot of it insightful. Maybe I'm not as priviledged as you are to be exposed to this stuff on a regular basis. Maybe 99% of the population isn't living in this bubble either.

It wasn't what you need for your paticular situation, and that's fine. But the general hostility towards this is strange. Why did you feel the need to attack it instead of moving on with your day? Genuinely asking.

I dunno. Same reason Mike Judge decided to make an entire show making fun of the same people I guess.

The OP has some good ideas.

That someone could see the future of smartphones when a mobile phone was a box of 100 pounds in the trunk of a car is a bit much. There WAS a possibility just by (1) talking to the electronic radio engineers about spectrum, bandwidth, propagation, and power, (2) learning about encryption, and (3) talking to the experts in microprocessor lithography and where that field was going. E.g., quite early on, IBM had their eye on line widths small enough to need X-ray wavelengths as a light source and built a cyclotron as an X-ray source.

The OP missed the possibly crucial role of original advanced technology, e.g., from applied math, that is difficult to understand, duplicate, or equal. Reasons for the OP to miss this possibility is that (1) they have seen little or no such on Sand Hill Road and (2) are unable to evaluate such technology themselves and unwilling to direct such evaluations by others. That the US DoD, DoE, NSF, NASA, NIH, etc. are willing to evaluate such technology early on and have done so with fantastically good effects is just ignored by Sand Hill Road. My guess is that the main cause is concerns by the institutional investor limited partners.

The part about building a demo to prove that the team can build serious software is not good: From all the months I contacted Sand Hill Road, apparently it doesn't much matter WHAT software a team has written in the past or what abilities the team has demonstrated, Sand Hill Road won't invest based on software yet to be written.

Writing such as the OP long misled me: I guessed that such considerations would be part of a VCs investment decisions. But after experience, I concluded, NO: Instead IMHO it appears that that the investment decisions are based on (1) traction significant and growing rapidly, (2) in a large market, (3) with little or no current competition, (4) with a team with not much obviously wrong with it, and (5) the team desperate for a check. The rest in the OP might tempt an entrepreneur to send a pitch deck but won't have anything to do with the VC writing a check.

I stopped reading at "If you can’t sell, starting a business is probably unwise."

You're not wrong. But that's basically dogma here. Sell first! Making a product before you sell it is bad!!

Core observation I have about a lot of startups (and other businesses): the ROI for sales and marketing is much better than the ROI for delivering actual value to your customers. Therefore, bullshitters tend to outcompete people who care about the product.

"If you can’t sell, starting a business is probably unwise" may be realistic, but it saddens me deeply as a human being.

> #6 “What the smartest people do on the weekend is what everyone else will do during the week in ten years.” “Hobbies are what the smartest people spend their time on when they aren’t constrained by near-term financial goals.”

We don't know this actually. What we know is the things that the 'smart' (whatever that is!) people who have succeeded did on the weekends. We not only don't know what the things they did that never went anywhere were or others who are less notable are spending their time on.

I hate these types of sugary statements that everyone flocks to.

Number one is not a great idea. Of everyone worked on most pressing matters then we’d live in an inconvenient world full of frustrations. Think of how many products and services we rely on that went through a bunch of iterations to become marvels that we take for granted without a thought. The mundane aspects of our lives are full of opportunities for businesses that can turn frustration into delight.

I know I'm probably on the wrong website, but I am so incredibly tired of reading and hearing about startups.

Brilliant article with deep insights - very very good read for founders that articulates a lot of the deeply true and difficult things about starting and building a successful company.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact