6 ms·
A former Google exec on how to make tough decisions quickly
- juddlyon 11y agoSide note: I've never read it before, but Quartz is a supremely well-designed website.
- puranjay 11y agoIt's funny because it doesn't have any of the usual gimmicks lay people associate with "well-designed". No parallax, GIFs or fancy effects. Just plain good design that focuses on usability. I honestly love it.
- vinceguidry 11y agoWhat's really funny is that two years ago, every single qz article posted to HN came accompanied by a horde of comments on how awful the site UX was. Site redesigns occasionally make the front page, I don't remember Quartz's ever doing so.
- dredmorbius 11y agoFixed headers and footers aside, it's tolerable on old Android mobile as well. Fairly effective design though, yes.
- kzhahou 11y agoTLDR: Make decisions quickly.
- dingaling 11y agoOne of my best managers had the default response of 'no' to any unexpected questions with unrealistic deadlines. Nothing flowery, just "no my team is not available to do that." Usually worked well ( except when C-levels were invoked to over-ride ) and saved us from a lot of half-arsed unplanned work at silly hours. And all the post facto refactoring that would involve. Her mindset was that positive answers could only be made when there had been sufficient internal discussion with all the facts present.
- GauntletWizard 11y agoThe phrase I heard tossed around is "Your poor planning is not my emergency". It could probably be more nicely worded.
- RBerenguel 11y agoIt's worth re-reading Hamming's "You and your research" to find where it appears (I just did and thought it was worth mentioning.)
- snorkel 11y agoThat stance can also backfire in the form of "OK you're too busy to help me so I'll hire outside contractors to do it instead." which leads to a bigger mess to sort out later. I try to at least offer a partial solution.
- vinceguidry 11y agoThe more unrealistic their demands, the less understanding they have over their situation. I found the best approach is to spend 10-20 minutes to help them think things through, so they understand what they're asking for and what they should be asking for and what a realistic timeframe is. Then put that on the roadmap.
- shostack 11y agoNot just that, but it creates a true "us vs. them" mentality in an organization where you're all on the same team. I've found in my career that there are very few people who want things done on extremely short notice with short deadlines for shits and giggles. Most of the time they are responding to some external (or higher level internal) pressure. Even if it is due to less than ideal planning, they are typically aware that they are asking a lot. Of course, that is in reasonable work environments.
- joe_the_user 11y agoI'd agree that's the right answer. The problem is that once other actors realize that this is how you operate, the question immediately becomes "how do we override this person's decision". Aside from just accumulating enough influence to make your positions stick, I'm not sure what the best approach - maybe it's necessary to give the impression you'll agree to some of those unrealistic questions merely to stay in the decision making loop.
- logicallee 11y agoThis is how you make decisions quickly: "Fuckit, youtube is the biggest video site, gmail is the biggest mail server, and people have accounts on there! What's Facebook got that we don't! From now on errybody gets a Google+ account. Want to post on YouTube - bam, there's your real name. Do you have a gmail? Now you have Google+. Let's see Facebook compete with that! That's it, decision made. Go!" I think on the surface this kind of a decision is a fantastic quick, touch decision, and one that fast-moving, risk-taking, companies can quickly try and implement. Unfortunately for Google, sometimes making tough decisions quickly is the wrong kind of quick.
- smacktoward 11y agoIt does however make the person making those decisions look "dynamic" and "decisive", which means by the time the damage they have caused starts getting noticed they'll already have been promoted out of the blast zone.
- dageshi 11y agoIn the end it didn't cost them anything though did it? They gave it a decent shot and ultimately perhaps the only other decision to take was not to try and make a social network at all.
- logicallee 11y agoI for one don't think they gave it a decent shot. What I mean is that I, and others, were annoyed by the heavy-handed attempt, and didn't see its value. This is despite the fact that Facebook leaves loads to be desired - even Facebook internal propaganda says "People don't use facebook because they like facebook - they use facebook because they like their friends." Of course, all of my friends are on YouTube or Gmail. But I'm not social networking with them on Google+. Something was missing.
- Karunamon 11y agoOpportunity cost and a lot of good will, but it's hard to measure both of those things.
- 11y ago
- trhway 11y agothe key isn't "quickly", the key is to have knowledgeable and smart decision makers who do in 1 second the amount of mental work that would take other people orders of magnitude longer.
- ljk 11y agoso.. it takes experience
- 001sky 11y agoIn most (large) firms it takes a week to have/schedule a meeting with those people. Most small firms don't have that amount of experience, or if they do its not accross a very broad base. So part of the story is about access to people, there is an extra layer of gatekeeping in there, by necessity needed to make sure those people are being used effectively.
- ape4 11y agoRight. Quickly making incorrect decisions probably isn't so good.
- rbobby 11y agoSee: lemmings
- dandrews 11y agoOriginally published here: http://firstround.com/review/speed-as-a-habit/ http://firstround.com/review/speed-as-a-habit/
- 0xdeadbeefbabe 11y agoI quickly decided not to read this as it would just disrupt my philosophy of not painting myself into a corner.
- aaronbrethorst 11y agoSuddenly Tim looked back at Sabih and asked, ‘Why are you still here?’ Sabih left the meeting immediately, drove directly to San Francisco Airport, got on the next flight to China without even a change of clothes. But you can bet that problem was resolved fast. What about Sabih's passport? Did he just carry it around at all times in case Tim Cook wanted him to fly to China?
- chkuendig 11y agoalso, china doesnt do visa on arrival...
- stephencanon 11y agoIf you travel to China for work, you already have a multiple-entry visa.
- drzaiusapelord 11y agoNot to mention his Visa. U.S. citizens must get visas before visiting. They can't just buy one after landing. So much of this sounds like managerial bullshit, to be honest. "He got on the plane straight to China and made it work like the Ayn Rand superman he truly is! You skill workers can learn a lot from this!"
- mikeyouse 11y agoI don't disagree but you can now get a 10-year 'multiple entry' visa for China if you're a US business person, so as long as you have a passport, you're free to visit whenever you're needed. http://www.usccc.org/#!china-visa/c1nj9 http://www.usccc.org/#!china-visa/c1nj9
- kbenson 11y agoMaybe you can get a visa that covers a time period instead of a trip? If he's the guy that regularly goes out there once every week or two, and that's a possibe to get a visa like that, I imagine they would. Otherwise, maybe his assistant was frantically greasing the wheels for his arrival while he was on the plan. Beyond that, the story read to me less like "a funny story about my old pal Sabih Khan" and more like "here's how much of a dick Tim Cook is". If you think saying “This is bad. Someone ought to get over there.” in the middle of a meeting and then later in that same meeting, feel compelled to single out a specific person and ask "Why are you still here?", then you are a dick. Being purposefully vague on the who and the when to then thirty minutes later single someone out for not doing something is a dick move.
- mrslx 11y agoDifferent google than the one i worked in, it took managers many quarters or even years to make big decisions. they needed tons of data and trends - by nature they are reactive to a fault. yet they run around talking about quick decisions making and being agile.. strange world it was.
- BurningFrog 11y agoThere are many different Googles.
- toephu2 11y agoYMMV. One google manager cannot represent all of the thousands of managers at google.
- EliRivers 11y agoFormer discussion here https://news.ycombinator.com/item?id=9953526 https://news.ycombinator.com/item?id=9953526 and https://news.ycombinator.com/item?id=9923239 https://news.ycombinator.com/item?id=9923239 The general HN feeling on those occasions, broadly speaking, was that this is ridiculous.
- dataker 11y ago> Google is fast. General Motors is slow. Startups are fast. Big companies are slow
- rokhayakebe 11y agoAll else being equal, XYZ makes a difference. is the weirdest logical statement.
- incepted 11y ago> A good plan violently executed now is better than a perfect plan next week. Er... what? In what world is this a universal truth?
- exelius 11y agoIn a world where situations are constantly changing. A good plan now can help make progress, while next week you'd still be debating what changes to make to the plan in light of new developments. Another version of this quote often attributed to Eisenhower is "Plans are worthless, but planning is essential."
- shas3 11y agoSeems like a re-phrasing of truisms cautioning against 'perfectionism' like, "perfect is the enemy of good." The idea behind these things is that perfectionists tend to overestimate how much actually needs to be done to achieve a goal. It is a form of the planning fallacy.
- cek 11y agoI loved this post. It makes a ton of great points, that resonate based on my experience. One of my favorite sayings is "90% of the decisions you make don't matter. For those, just pick. For the 10% use diligence."
- danielam 11y agoAnd then there's festina lente...
- basseq 11y agoFrankly, I think the tactical HOW part of this article is very light. There's much content on WHY it's a good idea and apocryphal anecdotes. Here's some actual advice I gleaned from the article: - Say, "We’re going to make this decision before we leave the room." (And do it.) - Begin every decision-making process by considering how much time and effort that decision is worth, who needs to have input, and when you’ll have an answer. - Internalize how irreversible, fatal, or non-fatal a decision may be. (And get comfortable with it.) - Give important decisions 24 hours, even if you think you know the answer. - Know when to end debate and make a decision. Use your "CEO prerogative" sparingly but decisively. - Gauge comfort to get to the right speed: low-level discomfort (stretching) is good. Then execute on those decisions. (This is the second half of the article.)
- ilaksh 11y agoSo basically the executive's job is to make snap decisions and crack the whip.
- Ologn 11y ago> Challenge the when...for items on your critical path, it’s always useful to challenge the due date. All it takes is asking the simplest question: “Why can’t this be done sooner?” Asking it methodically, reliably and habitually can have a profound impact on the speed of your organization. Sounds like a great organization - the person implementing something, who has the most knowledge of how long it will take to implement, needs to be "challenged" by someone who has no clue about how to implement what they are asking for or how long it will take. What they are describing is a broken organization, or the beginnings of behavior that will break an organization. I mean, it's fine to say we need to prioritize implementing certain elements of business logic, so we'll shrink the project scope so that those elements will be implemented faster. But to advise people to "habitually" "challenge" every schedule given by implementers is pathological behavior. It can work for a few months, but then the people doing the work get burned out and leave. I have seen shops with confident IT managers who are not afraid to say "no" to unreasonable requests and deadlines, with a solid team of programmers and admins who have worked together for a while and get along, who have stayed at the company for a while and who have executed well on many projects together, with clean code bases and solid infrastructure, and who generally work forty hour work weeks, with the occasional marathon before a big release, or if things are breaking. I have also seen shops with browbeaten IT managers, often new to the job, who are overrun by their bosses and business unit managers, with an IT team with a lot of turnover (except maybe 1 or 2 embittered people who have been there longer than the others), where people and departments are engaged in office politics, where projects have vague and ever-changing requirements, unrealistic deadlines, death march coding marathons by overworked coders, which are interrupted by putting out fires due to the code base with massive technical debt and broken infrastructure. These are the kinds of companies where the executive "habitually challenges" deadlines the weak IT manager gives him, who gives in and dumps the new unrealistic deadline on his team. This is the kind of company where a programmer is thrown into an existing project at the last minute, because the programmer who was working on it quit, and at your first meeting a Microsoft Project slide is shown and you're told that you're already three weeks behind schedule on your contribution. Which of these two companies wind up succeeding, and which end up floundering or even failing? (The only caveat I give to my own scenario, is that in companies where IT is not central to the business, they can often survive a broken IT department, when their core non-IT business is doing very well. Their company would work even better if they followed the rational scenario, but their broken IT department is not always fatal to the company when they're doing well in their core non-IT business.)
- decisiveness 11y agoA lot of the general ideas here seem helpful. I just have some slight issues with the section on rallying support. >Maybe you tell them that you used to work with a competitor who was quite speedy >I highly recommend this over a brute force method of escalating things to the person’s manager or throwing competition in their face. Seems contradictory. I'd also point out that questions and comments like: >Can you help me understand why something would take so long? >Hey we’re really betting heavily on this, and we really need you guys to deliver. >Are we working as smartly as we can? when directed at someone, can be vague and off-putting without some valid specificity behind them.