Daniel Gross: Why Energy Is the Best Predictor of Talent
safegraph.com
safegraph.com
I've worked with people who have extraordinarily high output, but slowed everyone around them down because they filled absolutely no-one in on their decisions and changes, so we were all left trying to beat the next code-base warping idea.
I've worked with people that accomplished almost nothing, but lifted morale so much that they more than made up for their lack of productivity. (I wouldn't recommend looking for this scenario though, most low-output individuals drag morale down...)
How many marriages do you think would make it 5 years down the road with only a month of knowing each other? A week? A handful of 1-hour interviews? I mean, that's the kind of time commitment you are looking at with a job.
Just filter out people who are obviously bad fits, and fire fast. The diminishing returns of trying X, Y, and Z, to me, just aren't worth it, and if you are a small company, you've got better things to dedicate brain power to.
I'd straight up not work for a place that took this approach. Seems like a great way to fuck over the folks who made a big life change to join you. Though this philosophy is probably super effective, as long as you're willing to not think about the effects it has on others. Yikes!
Personally, I would jump at the chance to work at a company with this philosophy, even if I ended up being the one getting canned. I don't think I've ever worked at a company that fired "too fast", if anything it was clear that some folks were clearly not going to succeed, but hard decisions were avoided, which took down the productivity of everyone involved, and when the folks were eventually let go or left on their own they usually ended up still being super bitter. A reluctance to fire, in my experience, just really translates into a company that has difficulty making hard decisions.
The first thing you have to remember is that Netflix pays a lot (a lot lot). The idea of expecting high performance only works when it’s accompanied with substantial benefits.
Second, Netflix does “fire fast” but not in the sense most people interpret it. New hires are given enough scaffolding to succeed. It’s not like other companies where they expect people to come onboard as revolutionaries and create impact in a month.
Third, high performance isn’t a constant like most people pretend it is. Netflix has managers who understand this fact and support their employees when things go down. They aren’t ruthless maniacs who fire you the first week you take a breather. And most managers who try to copy this elsewhere, miss that point.
But I do think it means that when it becomes clear that it's not working out, and the reason for it not working out is pretty fundamental to the employee's personality or aptitude, that you make a decision to let them go quickly and fairly.
It's really easy to have this attitude when you don't really think it will ever happen to you. I'm also guessing you've never been the target of harassment. I don't think you're going to like it as much when you're really excited about the job and on your first day you find out that the superstar developer on the team Gill just doesn't like you, never did. He was the one person who said they shouldn't hire you and is now pissed that he was overruled. Turns out Gill doesn't like not getting his way. You find yourself socially isolated because everyone knows what Gill is like and that he doesn't like you. Being seen talking to you isn't worth getting on Gill's bad side.
You get the shit assignments and code reviews are really rough. Especially the ones where Gill's around. Then you get hauled into the office about some problem with something you did. It isn't clear if it was something you did but it seems to be related to something Gill's working on. Gill's the super star and you're just the new guy so you can guess who caught the heat. It didn't go well when you suggested it might be Gill's fault. Turns out Gill was in the office right before you and he's telling a different story and they let you know what a superstar he is and everything he's done for the company over the years. It's clear that Gill's just going to grind you down as your quality of life decreases, you start losing sleep, acting out from the stress, you snap at your wife, children, and friends because you're still upset about something Gill did to you at work.
See the problem now or are you just imagining yourself as Gill?
The structurelessness & lack of a system, pressuring someone out, is no favor for Gill, but it seems like how it usually works.
i've seen 1 person in my career put on a performance improvement plan & let go. i've seen 5 fall out of general favor & get driven off. feels very lopsided & hard to see, grapple with.
dont appreciate your malice.
On the flip side, firing people around Gill, who he doesn't like, fast will make it apparent that Gill is the problem much more quickly than just keeping everyone around when things aren't working.
But I've more frequently heard stories of others who see "fire fast" solely as a way to save money and also avoid having to ever train or mentor. That's bad for the employee, and not a place I'd work, if I knew it in advance.
If you're looking to structure your employees' comp with a heavy equity component that takes a while to vest, "fire fast" becomes especially dangerous for the employee.
1. provide more value than you cost the company
2. make sure the boss knows (1) for you
Candidates should try to be aware of companies that won't give new hires much of a time window to prove their performance level or value creation. Especially if large amounts of their compensation depend on going at least 12 months without getting fired.
At Boeing, the accountants managed to pin down the cost of one pound of airplane weight. At the time, it was $250,000 per pound. This made it easy to point out to the boss how much money you'd saved the company by doing weight reductions. I.e. if you saved a pound, that was the equivalent of 6 years salary. Time to bring up that raise you deserve!
Look at Google for example. I'm quite sure that a very large percentage, if not majority, of the employees at Google produce net negative value. The problem, of course, is that it's impossible to know which ones. I know a bunch of people (brilliant people, mind you) that worked years on products at Google that never shipped, or worked on products that shipped but were discontinued long before they were ever profitable. But Google is willing to make these investments because they believe that, one day, one or two of them will pay off spectacularly.
I mean, Waymo has certainly never been profitable (heck they've barely had any revenue) - what do their employees argue?
The employee needs to be profitable, not the aggregate of all the employees.
As for speculative research projects, nothing succeeds like success (making money). Researchers who produce loser after loser projects I expect would be much more likely to be fired than researchers who had a win big enough to cover their losing projects.
> possible
The whole field of cost accounting is devoted to figuring this out.
Back in the 70s, I attended Caltech when Feynman was on the staff. It came up in conversation that Feynman was making a salary of something like $55,000 (a lot of money in those days). I said what could anyone do to justify paying him such a salary. A grad student laughed and said "you don't understand. Feynman doesn't need to do anything to be worth that salary. He just has to be on the staff, which attracts donor money and draws top talent to Caltech."
Of course, he was right.
But also, Feynman being Feynman, did do an awful lot as well.
In many universities and companies, you'll find such "prestige" staff.
Most companies are motivated by profit, and will, therefore, gladly pay someone to fill a role that earns them more money than it costs. I don't see severing a non-mutually beneficial relationship as fucking someone over.
If you are in a position that being let go would be horribly detrimental to your personal and financial goals, looking for a company that is very slow to fire may be a good way to go. However, that is for your benefit, not the company's. I was targeting my advice at employers.
The relative risk for an employee joining a new company is WAY higher than it is for that company.
For example, that may be due to any combination of age, skin colour, geography, ethnicity, visa/citizenship, neurodivergence, gender, family dependents, or other factors outside the person's control, helpfully summarised by the umbrella word "privilege".
After all, as is often said here, hiring is broken. It's not enough to be highly skilled.
I stand corrected, thank you.
I had to re-read both your comments carefully to understand them from that perspective.
Though you didn't intend to generalise, your refutation can be (and clearly was) parsed as an assertion about a whole category of people not having the high relative risk, rather than a singular counteraxample to say not everyone in the category has that risk. Your followup comment could also be parsed that way, unfortunately.
I think what motivated Bud's reply, and then mine because your followed appeared to entrench the position, is that it's common around here for people to talk as though virtually all software professionals are paid well and can easily slide into great new jobs as and when they want to, when really there's a huge "underclass" of software professionals who feel shut out of that nice world.
> I think what motivated Bud's reply, and then mine because your followed appeared to entrench the position, is that it's common around here for people to talk as though virtually all software professionals are paid well and can easily slide into great new jobs as and when they want to, when really there's a huge "underclass" of software professionals who feel shut out of that nice world.
I get it. I was once in a similar boat, so I understand the feeling.
The cost to someone in their early 20s to take a job for a month or two and then hop to another one isn't very high compared to someone far enough into their career to likely be making changes for their family and possibly face more judgement for a short stint from future interviewers.
that's "quitting fast"—quitting before you even take the job. impressive, you're hired.
What if instead of "fire fast", the idea was re-framed/restructured into the new employee initially working as some sort of consultant with a short term contract? Use telecommuting/work-from-home or other temporary workarounds to defer expensive/disruptive things like moving to live near the company until after the defined "trial" period. Instead of "firing fast", you either let the temporary consulting contract expire or you proceed with the actual hiring process and the deferred tasks.
Maybe this isn't possible in practice, but it seems like we could design a workaround to this problem if were sufficiently creative.
> think about the effects it has on others
Unfortunately, that takes proactive effort. Unless there is a mechanism to actively incentivize spending time and energy to protect others, most of the time that effort will be spent on "more important" things. ~sigh~
I've heard people claim arranged marriages last longer, on average, then non. Of course there are all kinds of confounding factors. The number one though seems to be the expectation to make it work instead of just giving up and moving on if things aren't perfect.
Cultures that arrange marriages typically do not allow divorce.
Even if we carry that comparison through, the marriages are normally arranged by family, people who have known you your whole life, and probably want you to not hate your spouse. Hiring via references from people that have worked together for a long time are much more useful than interviews and personality tests.
We could double the size of our interview panels and still get people that weren't right.
That's the whole job of seed VC investors to be better than random at selecting people who will end up creating unicorns. So I disagree it is "trying to squeeze blood from stone"
That's fair. In the context of placing bets, the dynamic is different, but VCs have the benefit of simply not investing in follow up rounds, which is as close to 'firing' as they can get.
> You could really have the worst culture and the best mismatch of talent and still have a unicorn.
True. However, I'd be willing to bet there is predictive power in how horrible a company's culture is in the early stages.
> That's the whole job of seed VC investors to be better than random at selecting people who will end up creating unicorns. So I disagree it is "trying to squeeze blood from stone"
Of the VCs I know and have met, the successful ones don't give much (almost no) weight to personality tests, outside of more obvious 'red-flags'. The checklist I've seen is:
1) History of entrepreneurial success. This is probably the best way to get in anyone's door. 2) History of domain success. They want to see that you know what you are talking about and come with the connections necessary to make the venture successful. 3) Does the founder's approach to the problem align with the firm's. Most high-quality firms look to invest in areas where their partners have deep domain knowledge, and they look for companies who they believe are making good bets.
There are very "impression" agnostic measures. I wouldn't put a lot of weight on "energy", but hey, I'm not a VC, so what do I know :P
Not only is it boring to listen to, but very much subjective.
You cannot measure quality. Its simply too difficult with all the variables in play over time.
I’m pretty convinced that most high performers follow this pattern. They’re really good at making it very visible all the stuff they’re doing, but they only focus on what’s visible, and leave all the underlying details/maintenance for the “lesser” employees to clean up after them.
Management jumps them from project to project because of how “good” they are, meanwhile every team they leave behind has to pick up the pieces. This has the perverse result of making everyone else appear to be low performers, because they’re too bogged down bailing out the leaky boat.
Well, there seems to be an awful lot of blood . . .
Oh! I bet it's from the wringing but not the stone. Nice illusion, though.
The only thing I full heartedly disagree with you is on the fire fast part. It's a lot better to move them into another department.. sometimes it's not the person it's the environment they're put in. If you're too small... this is the part of doing business is to leave them better than you found them.
My former company fired bunch of people at the beginning of COVID, because of the reaction of markets and our clients, some of whom left us. The fired employees were either new hires or bad apples. IMO it cleaned up the air a lot at the time.
On the other hand firing people is costly. For small to mid size companies, this advice might not be a as viable as it sounds.
const antilooped = [];
const looped = [];
const looper = data.map((item) = {
if (item.nonX) {looped.push(item) }
if (item.nonY) {antilooped.push(item)}
});
then would not use half of the variables he instantiated, and probably merge looped and antilooped arrays in some weird way. Invariably debugging his code led to me just rewriting it all.
And I remember just about the first thing he asked me is if I went to prison in Jylland in Denmark and a big guy named Ole wanted to make sweet love to me would I do it or would I rather die, and then he volunteered he would rather die. This turned out to be his typical lunch time small talk.
on edit: just to be clear, he didn't use the looper array for anything.
Otherwise, there is a simple rule for engineers: do not write the code that makes the next engineer fantasize about your death.
If somebody doesn't understand the fundamentals of writing code for a code-writing job, though, that's a different story. That's what self-study and the education system is for.
Training them is the lowest risk and lowest cost option of these.
As far as his being trained I mean, yes he was young, but this wasn't his first job and as I understand it had certainly talked his skills up in the interview.
I've worked with a lot of very smart people over the years who just didn't have much energy. It's not that they've all been unsuccessful, but they aren't as successful as someone who may not be quite as smart but has much more energy. I put myself in that category - I don't have the energy to be super successful and I've realized it for quite a long time (probably going back to about the time of that dinner) -I'm not going to create a startup or be a CEO/CTO. We aren't all gifted with a high-energy physiology and that's fine - we can still have a decent career. When you find the rare individual who is very smart and has a lot of energy they can go very far indeed.
Energy alone is not a good predictor.
I hope that much is obvious to everyone.
The podcast isn't literally suggesting that we discard all metrics aside from energy, or that energy automatically results in success.
They talk about a lot of different values, the difficulties in measuring them, susceptibility to interviewer bias, and so on.
Spotting talent is hard.
I would say that energy is an X factor. If you have two candidates and both demonstrate similar system comprehension, go with the one with more energy.
Energy is great for short term success, contemplation is great for long term progress. Maybe hire a mix of tactical and strategic thinkers? And then get them to work together somehow? It's easy to say and hard to do!
Writing about failure is easy, because you get to do it after the fact when everything is obvious.
Writing about your own successes is fun, because you get to talk about something that turned out well.
Trying to tell someone else how to be successful by following what you did is often a fool's errand, because you can never convey all the things you did to be successful; and in the case that you were VERY successful, you have changed the marketplace of products and ideas!
When people want to get it done right now, they don't tend to make good decisions.
Maybe you're thinking hyper activity?
Not my personal philosophy but I've seen it used.
But that doesn’t mean that it isn’t a top predictor for talent. Can you think of a better proxy that this might have been about?
They talk about many different dimensions of individuals and how feasible they are to evaluate in the context of an interview.
There are also great tidbits about techniques to help get honest answers from candidates, such as structuring the interview format in ways that leave less leeway for candidates to try to game up the answer they think the interviewer wants to hear.
It's an hour long podcast. I remember it being interesting at the time. Worth a listen.
1. Smart but lazy. 2. Smart and energetic 3. Dumb and lazy 4. Dumb and energetic
Type #4 is dangerous. Those are the ones who energetically do the wrong thing and get people killed. The most famous losing generals (Custer, Cardigan, Burnside, McClellan, Westmoreland) were type 4.
— General Kurt von Hammerstein-Equord, 1933
Heard this before, always fun, but random observation here... major nod to the Gervais principle? The slackers as the ones with the true model of reality, with a genuine assessment.
The most successful person I know has boundless energy at work, and is often used as a resource for many other people, helping them get their work done. She's on the national board of the professional association that oversees her field. She takes exceptional care of her extended family, and somehow manages to squeeze in vacations, nights out dancing, and time with her friends watching football on the weekend. Furthermore, she volunteers for an organization that helps new immigrants feel welcome, get settled, and find employment. Also, she decided to teach at the local college two nights a week for the last few years and gets glowing reviews from her students.
Makes me tired just typing it all out.
One of the authors is known for being a YCombinator partner. The other is known for responding to every email. I'm not sure which is more impressive.
Another extremely good signal are people that won’t give up easily.
Tenacity, grit, whatever you want to call it.
All that trumps degrees and knowledge, because all you have to do is spark the interest of this kind of person and they will find a way because they just won’t stop until they do.
Sometimes students just need a lift up and then they find that ability along the way, but it just doesn't seem to ever surface in others.
Maybe it's just a case of "I did things the hard way so everyone else should too" -- my path was certainly not efficient! But I have to wonder if they lack that "grit", and if it can be taught or if is something innate.
Here's another example which might illustrate the same mindset: I go to IRC as a last resort, because I don't want to ask anything I could have solved in a few minutes by myself (or worse, that I could have found on Google!)
But when I do go to IRC, I formulate my question very precisely, explain the facts, the steps I took, my hypothesis... and 9 times out of 10, just the act of writing the question out causes the solution to become obvious! (Or reveals something I forgot to try, which then solves the problem...)
I'm currently taking a class, which is not programming related, and it's maddening because the teacher and course material do a very poor job of covering the basics, so from my perspective as an outsider it is near impossible to figure out how everything relates. It's possible that your current students need a general overview of how all the pieces interrelate before they can internalize and build their own knowledge.
More than being some kind of super genius, to be a successful engineer you need to have the stones to face your stupidity every single day. Most jobs have clear a to b to c to d steps. You know when you're done.
With software dev you faceplant almost every couple of hours until you finally succeed and get that high. Then you start all over again with a new ticket.
Most people have no desire to live that every day. They are happier in excel massaging spreadsheets.
The ability to sit with hard problems and continue to be curious in the face of pain and adversity is a very important skill.
Regardless, seems like a bad idea as blanket advice for all roles.
High energy people are also in my opinion the most draining to work with and the most likely to be discounted as juvenile or inexperienced.
In my opinion, the best indicators of talent are being humble, metered, incredibly well spoken / articulate and willing to help someone out of a situation they'd otherwise be afraid to go through with a superior. Handling stress without being a adderral ridden monkey while maintaining resolve is a trait that the most talented brilliant people I've ever worked with have had.
Given this quote, I feel like the title needs to be qualified really heavily, it's looking at a very specific type of talent for a very specific type of situation.
Instructions were to forward his resume if he was passionate about it. Didn’t care what side of issue the candidate was on.
Just that they had an opinion and showed enthusiasm about why their “side” was best.
It worked very well.
Depth first learning meaning that I go deep a specific topic before learning adjacent topics (e.g. if building a database access layer + cache, I'll go deep on Redis documentation, which cause me to go deep on specific redis commands, etc. then after redis I move on to cache rehydration strategies, etc).
This is reflected in my enthusiasm in interviews and ability to "talk their ear off" if they are interested in the topic. Maybe I have it backwards, my enthusiasm for learning technologies is why I go deep in the first place.
I read this more as "in defensive of high energy people" which then attracts the sort of vitriol you see in the comments section here, e.g. accusations of sloppiness, short attention spans, etc.
Obviously, quality is required to the extent customers judge products in that category. Tesla had/has astoundingly poor quality, and Henry Ford is famous for optimizing-down quality as well.
Why? He's a 31 year old from a startup which left zilch to the world.
A good predictor is something that is highly correlated with the thing you care about and not anything else.
Also, I've worked with a super energetic person who couldn't stand in his place, talked way too much, was rude, and, in general, unbearable. He was fired later.
So... maybe to soon generalize.
For (1) the VC, my summary view is that the VC criteria, what VCs pay attention to, is growth in traction -- they want the graph of traction over time to be "up and to the right". For more, they look at the size of the market. For one more, maybe consider if there is a "Buffett moat", that is, a barrier to entry for would be serious competitors. Maybe they look if the startup is in a field, space, they are interested in. Finally, if all these look good, they look to see if anything is seriously wrong.
At times I have suspected that some VCs write essays on their investment criteria that, really, are intended broadly to get attention, have their name better known, and in particular to get applications from founders, that is, "deal flow".
Then I would insert into this thread that, once a startup looks good in the above criteria, "energy" will not receive serious evaluation. Or, the above criteria are the real ones and will NOT be ignored or overruled, and if a startup has done well on those then other criteria, e.g., "energy", must be close to irrelevant. Or, if a startup looks good on the above criteria, then passing up the opportunity for a unicorn for relatively subtle reasons will be regarded, by the VCs partners, as foolish.
For (2) the manager, I would hope that he (or her, here and elsewhere) has a situation, that is, some work to be done, has that work fairly well defined, regards currently his team as too busy to do that work, so wants to hire. In that, the main criteria should be, can the candidate do the work the manager has in mind?
I would insert into this thread: That situation should be rather clear and definite and not subtle or complicated.
As someone in my late 20s, I was honestly concerned about working for a teenager; but he left such a mark that every few years I Google him and Greplin. Interesting to see where he goes.
Sometimes when I’m not actively working, I’m thinking and planning stuff in my head while seemingly loafing around.
Other times when I’m energetic I don’t know what to work on.
It’s a cycle. With humans nothing is constant.
Ick. It’s a really unfortunate bias that quiet people can’t be viewed as good leaders.
Personality types is an area that has been researched, and I balk at the idea that only a small subset of personality types can be leaders.
I have witnessed energetic optimistic leaders before, and they can people crazy, because they never actually address problems in concrete ways. People experience this all the time in companies: managers getting promoted up the ladder of management, leaving unsolved problems in their wake. The Enneagram personality system address things like this.
I stopped listening after the first 5 minutes but I believe he literally says that he has "1% of the appropriate knowledge to answer this question." I continued to listen anyways. 1% is almost the definition of beginner for Dunning-Kruger[1].
"Spy the Lie" or "How Emotions are Made" are both good books that go into depth about why it's near impossible to accurately judge anything about what or how someone is thinking and that there are no* accurate universal theories about human thinking or the link between human behaviour and human thought. (Feldman Barrett extensively footnotes her research in HEaM and it's really impressive.)
So, in short, this is an armchair psychologist who's gotten lucky in his particular area of the world (probably California?).
[1] If you read carefully, you'll realize that I don't believe in Dunning-Kruger either, given what I say. Sorry!
That is up to leadership.
What's your opinion of Elon Musk?
What's an Elon? I assume not a river.
How did I do?