Startup advice
blog.samaltman.com
blog.samaltman.com
"Make something people want."
This is a great sound bite, but it's not really actionable advice. Same as "don't die" or "get lucky." If you start a company unaware that you need to make something someone will want, you have no business doing so.
Furthermore, it's absolutely nontrivial to start out making something people want. A better piece of advice (still flawed) would be: develop a process to make sure you don't deviate from building something people want, and/or converge towards something people want. You can always rewrite history (e.g. Twitter, Google) and make it look like you always knew that you were building something people wanted. In reality, Google tried to sell itself to Yahoo for $1M in 1999 (and Yahoo said no). Those of us who were around at the time remember this.
I find it very dangerous to "tweet-wrap" advice that is the result of long slogs, and requires context. "Don't die" or "get lucky" sort of make sense if you know the back story. Otherwise you might as well say things like "grow wings" or "remember you always have a friend in the diamond business."
Soundbites are also nice because they provide quick reference points and mental footholds when you're thinking and working through problems.
I'll also refer you to #46 regarding your better piece of advice.
Therein lies the incredibly actionable advice of the 4 word statement. In fact, a lot of my technologically and ideally awesome projects falter at the step where I need to find out what type of people would want this, how they'd want it wrapped up and how the hell will I reach them.
That is a nonsensical statement. A person cannot make decisions based on something other than what they think.
Point is, you can either make decisions on what you think, or you can come up with hypotheses and find hard evidence that supports them. In fact, people often go with evidence in spite of their innate biases, thus making decisions based on what they don't think.
Much better advice in my view is this: Identify a problem and build a solution to that problem.
Payment from real live users/clients/customers is external validation too.
Which you can't fully do until you or someone else is offering it.
> Payment from real live users/clients/customers is external validation too.
Which you can't do until you have something to offer.
What you can do is this:
1) What is the problem you are trying to solve?
2) How annoying is that problem? The more annoying the better.
3) What is minimally needed to solve the problem?
I think if you focus on that process, on building minimalistic solutions and then trying to get customers who can help you take things further, that this can work. However, I don't think you can effectively do market research on something that doesn't exist yet.
Bullshit. Companies do this all the time.
Hey ABCorp, it looks like filing is a major nightmare for you, would you be interested if in working with me to find a solution for your problem? Would you be willing to commit to paying for something if I built it? What would you think is a fair price? Etc etc.
The mere act of having this conversation is the validation you are looking for. If things go really well, they may actually help you fund it prior to you being live, in exchange for a discount/preferred status etc.
Advance commitments happen all the time, you just need to take the right approach and actually listen to what people are saying.
Well, yes. That being said, as with many cliches, this one's a cliche for a reason. You'd be surprised how many entrepreneurs set out to make a cool product first, and a solution second. It's surprisingly easy to build a product -- even a beautiful, elegant, simple one -- that just doesn't solve or do anything useful for anyone in particular. It's very easy to mistake elegance, qua elegance, for utility. Elegance doesn't necessarily imply utility, but many fall into that trap.
The mantra "Make something people want" is better expressed as "Make something useful." Or, perhaps, "Make something that solves a need." Don't build a hammer and then go looking for nails.
The rest of your post ("Furthermore..." onward), I completely agree with.
True, but which of the following do you think was the problem?
1. The entrepreneurs did not know that they should make a product people want.
2. The entrepreneurs did not know that no one would want their product.
The problem with the saying is it doesn't address that obvious misunderstanding, which is why I propose changing it to "Make something useful."
No, that's not what I'm saying. What I'm saying is that "cool" is relative to the customer's needs. There is no such thing as an objectively cool or objectively interesting product. It is the user who determines whether something is cool/wanted, and the coolness/need is determined by the degree to which the product solves something.
I'm not using the word "solution" in the sense of B2B jargon. I'm using it in the sense that you have a Need, and you have a Fulfillment/Resolution of that Need. The need could be trivial, or it could be major. It could be for business utility or for personal entertainment. It could be a need you didn't know you had, or a need you had but weren't totally cognizant of until the solution articulated (and solved) it. Or it could be a need that's been plain and aching for a long time.
We are in agreement here.
The only possible disagreement is whether focusing on what people "want" is a useful metric. People may think things are cool. The better view in my sense is to focus on what people find useful.
That is extremely actionable advice. I witness lots of founders talk about all sorts of things except that and it's pretty easy to predict failure.
Wow I never knew this. Care to expand further?
Nor is it "tweet-wrapped" so much as concise. You said a better piece of advice was "develop a process..." but that's just a longer-winded way of giving the same advice. Is that better? Occam argues it's worse.
You're right that it's nontrivial to start out making something people want, but the value of the advice is to keep you focused when you're being pulled in different directions. If you're unsure how to proceed, default to the path that takes you closer to making something that people want.
It's a feature, not a defect, that translating this advice into a specific plan necessitates pondering how to get from your present position to that one.
A lot of the startups I see are just that. They're vapid and have little insight. True innovation comes from a different place than looking for acceptance.
It's easy to get people to want useful things. If you can focus on solving needs and headaches for people they will want what you offer.
I would add you could expand that to, "when evaluating a potential new product ask, 'Who will find this useful?' and 'Who needs this?'"
"A lot of times, people don't know what they want until you show it to them." — Steve Jobs
I think the start-up community sometimes feels more like a self-help community than a bunch of cool dudes (and girls, occasionally) building cool shit. There's no secret to success and these kinds of lists are both trivial and highly tautological (sometimes even wrong, but plenty of people are posting about that).
Come up with an idea you think is good. Implement it. Get people to use it. Pay attention to feedback. There's nothing mythical about being successful. Sometimes, it's nice to get lucky, and many (many) times, you will fail. Go back to Step 1.
I'm only reading the OP because I'm at work, but if I were at home working on my "next big thing," the time spent would be much better invested opening up Eclipse or Notepad++.
Go ahead, downvote me. But after that, close HN and get back to work!
Not referring to you specifically (this is likely a case of Hanlon's Razor http://en.wikipedia.org/wiki/Hanlon%27s_razor), but this is a common tactic for soliciting sympathy upvotes -- "fishing for compliments" as it goes. It's about as bad as "upvote if you like X, downvote if you like Hitler" or "Does anyone else Y" style posts.
Given the contrarian nature of HN comments, disagreeing with the OP is typically quite safe. If your comment fits the mold of "stereotypical top comment on HN" then it's hard to believe that you're bracing for downvotes.
Such writing may seem very superficial and empty, but it is in fact full of wisdom. It's just that you either have to have experienced it or go through the effort of putting yourself in the proper mindset.
"Don't pay people a fair salary; instead, offer them bespoke lunches catered by the finest local delis, a bus pass, and the promise of an amount of equity ultimately so trivial that even if the company sells for a couple hundred million, they still won't be a millionaire after all the taxing is said and done."
I'm guessing you'd need at least 2% up front to have an even chance of having 0.66% (or $1MM + taxes, in a $200MM exit) of the company at the time of sale.
"Sell before you build" - Paul Nuchheit
“Be a cockroach, impossible to kill”
“Do stuff that doesn’t scale”
“Launch early”
Ask "five why's" to get to the root of what customers want. Usually, it will reveal "they want simplicity."
"don't sacrifice the WHY in the process of speeding up the WHAT and HOW" - Hunter Walk
"To achieve great things, two things are needed - a plan and not quite enough time" - Leonard Bernstein
"The only people who see the whole picture are the ones who step out of the frame." - Salman Rushdie
10. Growth solves (nearly) all problems.
I'm amazed at the number of people who don't understand the difference between improving conversion of people already on your website and driving new users _to_ your site.
While this one is spot on for team members, and opposite when finding investors:
19. Generally, value aptitude over experience.
Aptitude is crucial for execution, but that's not the VC's job.
Notwithstanding the point that trying to predict what stock options could be worth is generally a pointless exercise at early-stage startups and may not be advisable from a legal standpoint, if you really believe the best case figures you're likely to show or want to show in such a table, you would almost always have an incentive to minimize the amount of equity doled out.
A company that prefers to treat equity as an immediate substitute for salary either a) believes that its equity is cheap, b) is cash constrained or c) has concerns about its cash position going forward. These are all red flags for someone considering joining the company as an employee.
Nothing is guaranteed at a startup and early employees should assume the risk that the money will eventually run out is high. But even so, a company that would prefer to minimize the cash part of your compensation and maximize the equity portion should be looked at with extreme skepticism.
7. In general, don’t start a startup you’re not willing to work on for ten years.
Best piece of advice he gave, and I think more important than "built something people want" (both are important).
13. Overcommunicate with your team. For some reason most founders are really bad at this one. Transparency is your friend.
Yes, but so hard to get this right. Start this early. Things you get in the habit of holding close to the m-team will probably stay with the m-team forever. This is something that's worth to spend mental energy on early, even though it won't matter until you're bigger, because it's so hard to do once you're bigger.
15. Hire slow; fire fast. Hiring is the most important thing you do; spend at least a third of your time on it.
People say this all the time and I don't believe it really works; I thought we're be a fire-fast company when we started but discovered it's hard to be fire-fast and not sociopathic. Don't count on being fire-fast unless you're sure you can do it (for instance, because you've done it before). Weight hiring decisions that much more.
16. Occasionally think about why the 20th person will join your company.
A great question, since we've been doing that for awhile, and answering it has motivated changes in how we approach the management of the company. Asking this question constantly will help your culture.
18. Hire friends and friends of friends. Go after these people like crazy to get them to join. Some other candidate sources are ok, but I always got bad results from technical recruiters.
I think this seems true because all the conventional hiring vectors suck so badly. Have a better plan for hiring than "friends of friends". There is really no reason to believe that you are so central to the talent graph that your friends of friends are going to be great. What makes this worse is that relationships will queer the readings you get during your hiring process; you'll want people to work out.
If your hiring process is fun for the interviewers, it's probably a bad hiring process.
19. Generally, value aptitude over experience.
1000x this; this is our hiring thesis.
20. Hire people that you could describe as animals.
I don't know what this means but if I look at our top performers I would not uniformly describe them as animals. Beware irrational criteria.
28. Keep salaries low and equity high.
This is probably vitally important for the kind of company YC starts but consider businesses that won't require you to do this, because retention is very very hard too.
33. Journalists like hearing directly from founders. If you hire PR people, resist their desire to control all the contact.
Be suspicious of ideas that require you to have a PR plan.
44. Get on planes in marginal situations. In-person is still better than tele-anything.
This is true and also can be usefully flipped, ie on investors or partners; someone who is serious about doing a deal with you will get on a plane, or, simpler, can be induced to take a call on a Saturday afternoon.
57. Celebrate your wins as a company. Get t-shirts for big milestones.
Something I learned at the largest company I worked for was that it's often valuable to brand milestones in advance. People can use the brand as a shorthand. Often had the effect of assumptively closing on management buy-in; once you had the manager using the branding name for the project, they were automatically sold on the project.
60. Meetups and conferences are generally a waste of time.
Yes.
63. Be suspicious of any work that is not building product or getting customers. It’s easy to get sucked into an infrastructure rewrite death spiral.
Yes.
68. Don’t have a diverse culture in the early days.
Probably no.
71. On the really bad days, remember that tomorrow will be better—it’s hard to see it being much worse!
Since I see people asking how to find cofounders so often, here I'd like to add: you know you have good cofounders when things are going wrong for you, the world seems like it's falling apart, and you suddenly realize with great relief that there are more people who will have to melt down before anything bad happens in the company. Can you imagine feeling that way about a prospective cofounder?
This is why I think the cofounder matching sites are kind of silly. But I guess if it works for dating...
94. Have a table in your offer letters that shows how much the stock you’re granting a new hire could be worth in various scenarios.
This sounds like a really up-front way to handle that. I think you should offer less equity and more salary, though.
This was a great list. Thanks, Sam Altman.
Best piece of advice he gave, and I think more important than "built something people want" (both are important).
But did he take the advise himself with Loopt..?
This is probably vitally important for the kind of company YC starts but consider
businesses that won't require you to do this, because retention is very very hard
too.
Seconded. This is exactly counter to advice I heard from PG when he was giving a talk at YC once, which was "negotiate for equity, not salary". At the time it sounded like good advice, and I still think most I've heard from him regarding starting a company is. But in this case, after having spent a year actually involved in a startup from the beginning that's been raising money, I find it hard to imagine any company who's followed a similar path to us (basically how you see most YC companies go) would be able to offer any sort of meaningful equity after even a second seed round.
It takes a lot of growth and additional options grants for single digit equity in options to amount to much at the exit. I would focus on the salary, possibly experience for your career if you need that, more than I would in getting rich off of a startup that hires you after they are funded.
When I was younger I knew a couple of people who all but flat out refused to get into business with someone they did not know socially. Their rationale was, apart from having a wide social circle, that to be a friend or a friend of a friend you had to go through fairly sophisticated and long term filtering processes.
I used to reject this in favour of "professional" approach but I am really swinging back round - I think one can be more honest and open with (almost all) social acquaintances than with people who are essentially competitors only. And that openness and honesty is going to add an extra 30 IQ points to the teams collective intelligence
So again, great list and some real nuggets in there
This is in stark contrast to the "where can I find a cofounder" and "startup dating" stuff I see these days. A company is about its people, and a startup even moreso.
I've had great luck working for awesome friends of mine and awesome friends/acquaintances/ex-coworkers of their friends. In a large software organization, so much comes down to not just the company and the problems but the exact team you're on that removing a lot of that uncertainty and working on a team with/for people close in your social graph is great.
But strongly biasing your hiring search toward people close in the social graph is a terrible idea. First of all, you're artificially limiting the pool of "ideal candidates" dramatically. Second of all, if most people you hire are friends of friends, you're likely not to build a very diverse company. Your culture will get skewed toward things that you and your friends care about and are interested in, and pretty soon there will be large groups of people who don't feel that they especially fit in at your company because eg. they're not male, they have kids, they don't drink, they don't watch sports, they read philosophy, they don't find working >40 hours/week productive or enjoyable.
This is a problem in many Silicon Valley startups that leads me, despite being demographically indistinguishable from what these companies overindex on (young, male, white, straight, elite education), not to want to work at most Valley startups. I deeply admire that Thomas cares about making a company that avoids these problems, which is a large reason why I'm working for Matasano this summer instead of one of the many Valley companies that would feed me three meals a day and in many cases would have a better-known brand.
I don't agree with that. I think for early employees, the following criteria matter most, in order:
1. Dedication and loyalty. Working in an early stage business takes dedication more than anything else. Aptitude and talent is nice, but the highly dedicated 70th percentile will be far better than the flighty 95th percentile guy.
2. Creativity. The way to solve problems early on, on a tight schedule is to be creative, repurpose what you have, and so forth. Creativity cannot be stressed enough.
3. Technical aptitude and talent. These are less important than being able to sit down, look at what you have, and shim a bunch of existing pieces together to get them to work right, or to put in long hours for deferred pay.
So with this in mind, there are two reasons to focus on people closer to you socially. First, it is far easier to inspire loyalty and dedication for those you have known for a long time or have mutual friends with than it is with people you just met during the hiring process. Secondly you can have a much better sense of the creativity involved.
What do you mean by this? Could you please expand?
Maybe he means you can describe each person as an animal according to his or her personality. People with strong personality can be easily associated with some kind of animal, so this advice basically guides you to hire an interesting group of people.
One of the best tricks I learned during our startup was a rule for deciding who to hire. Could you describe the person as an animal? It might be hard to translate that into another language, but I think everyone in the US knows what it means. It means someone who takes their work a little too seriously; someone who does what they do so well that they pass right through professional and cross over into obsessive.
One of the best tricks I learned during our startup was a rule for deciding who to hire. Could you describe the person as an animal? It might be hard to translate that into another language, but I think everyone in the US knows what it means. It means someone who takes their work a little too seriously; someone who does what they do so well that they pass right through professional and cross over into obsessive.
In other words, if you might exclaim, "That guy is an animal!", he's a potential hire. Similar terms are monster and beast, and indeed the former makes an appearance in another of pg's early startup essays, "How to Make Wealth" (http://www.paulgraham.com/wealth.html, 2004):
At Viaweb we had one programmer who was a sort of monster of productivity. I remember watching what he did one long day and estimating that he had added several hundred thousand dollars to the market value of the company.
So, a startup is a group of animals, beasts, and monsters that tries to grow fast by making something people want. To maximize its chances of success, a startup should have a CEO named "Max" who dresses in a wolf costume and sails to where the wild things are.
96. Stay out of jail.
97. You may feel you lack the time to defecate, but this is generally not something you can wait until after your big exit to do.
98. You are a special snowflake. Act like it.
I may have some bias.
"Don’t let your company be run by a sales guy. But do learn how to sell your product."
But for me it hits a little closer to home, I saw first hand how this could be a very bad situation and its a mistake I won't make again.
Founders should be product people in some way, that does not necessarily mean they need to BUILD the product, but they should definitely be more than just a "Sales" person
Don't rely on the press to sell your product; it is almost always unsustainable.
In other words, how would you advise a startup CEO/founder manage their time in a big picture sense?
2) ???
3) profit!
Following this advice only gets you dependent on that someone (if I understood the point correctly). So if they go down (read: limit/lock their API, like Twitter did), you go too.
How can complex coordination create value? An example, perhaps?
From http://www.paulgraham.com/start.html -
What do I mean by good people? One of the best tricks I learned during our startup was a rule for deciding who to hire. Could you describe the person as an animal? It might be hard to translate that into another language, but I think everyone in the US knows what it means. It means someone who takes their work a little too seriously; someone who does what they do so well that they pass right through professional and cross over into obsessive.
Something got lost in that one sentence synopsis. He wasn't suggesting that you dehumanize your employees by literally thinking of them as animals.
I really didn't understand this one - mainly the first sentence. Can anyone (sama?) elaborate?
With "friend := someone whom you personally know for sure to be a able worker"
> 57. Celebrate your wins as a company. Get t-shirts for big milestones.
If you insist on making your employees cringe ...
Examples:
1. Hire slow; fire fast.
If you screw up, you're gone. So much for job security.
17. Hire smart and effective people that are committed to what you’re doing. The last five words there are important.
You're ideas aren't important unless they echo ours, so please, don't rock the boat. kthx. See #1.
20. Hire people that you could describe as animals.
i.e, hire people that are willing to join your cult and have no outside interests. They'll wear the scars of your whipping as badges of honour.
28. Keep salaries low and equity high.
While you're at it, just post "Fuck you" on your door too... they should want to join you just based on the promise of winning the lottery and because they are generally young and ignorant. See #20.
I'm being hyperbolic and a little unfair in my criticism in an effort to make a point - working for a startup has all of the downsides and none of the upsides of founding a startup.
People quickly wise up to this, which is usually why not many people with families and such commit to the "startup lifestyle". That isn't to say that they avoid startups completely, just that they avoid the ones that treat their employees more like machines than people.
You can avoid all this by just doing the opposite of these points in many cases:
1. Hire slow, Fire slow
Let people screw up. Trust your hiring process. If they were so great on day one, why are you looking to fire them? Why don't you ask them what's changed? (Note, this doesn't mean never to part ways with anyone)
17. Hire smart and effective people that are committed to what you’re doing.
No need to emphasize the obvious. Embrace your employees, and their life changes. Embrace the fact that their thoughts, feelings, desires change as yours do. Work with them, within reason, as much as possible. After all, considering you hired friends of friends, this is your social network you're talking about.
20. Hire people that you could describe as animals.
Let people be. Your number one expense should not be high blood pressure medication. Sure, you need the tigers, but you also need people to keep you grounded. Nobody would suggest only hiring men, or only hiring people that play soccer. Don't hire people that only 100% drive. These people (myself included) also have a tendency to slip into pure assholedome. You need more than a bunch of assholes.
28. Keep salaries low and equity high.
I think this is self explanatory. Don't exploit people if you don't want to deal with "firing fast" every second month. Your employees have tremendous risk in joining a new, unproven startup with little to no income. They should receive a salary that reflects this, unless you're partnering with them and bringing them on as part of ownership.
"Yeah, this guy is totally a zebra. Oh, that programmer over there is such a squirrel".
Great advises, thanks for sharing!
<i>Pay a lot of attention to the relationship between cofounders, especially if both/all of you want to be CEO.</i>
sama can you clarify - as you could take it a couple ways
aka "Don't YOLO"