A Silent Cause of Bad Business Decisions | Hacker News Reader

A Silent Cause of Bad Business Decisions

medium.dave-bailey.com

95 pointsdavesuperman55 comments

Open articleView on HN

When my manager comes up with a bad manager-idea, I ask questions like:

() what concrete problem are you trying to solve?

() what do you think is the root cause of that problem?

() how, specifically, is this idea going to address that cause?

() what might be other possible solutions?

This doesn't always work right away, but at least it points the conversation in the right direction.

Sounds like you're managing your manager.

Seriously: Why wouldn't everyone manage their manager?

We could make a new word for people who manage each other mutually, coexisteling in their operations without blind hierarchy...we could call it mumanagement

>mumanagement

Not 'synergistic' enough to be adopted.

Could use "μmanagement," which includes side-irony.

Ψdirony?

How about 'teamwork'?

Because I won't take on the responsibility of someone making twice as much as I do. ;)

It's very hard to succeed if your manager fails.

What do you care if they get paid more than you? Do you care that they can out bowl you? Do they care that you can out chess them? Life is a marathon, you don't win a marathon by coming first in the race. You win a marathon by beating your last time.

Don't misunderstand, I'm always competing with my last paycheck. I always want to be getting paid more. I just don't care about other people's salary and if our responsibilities are lesser/equal/greater.

For me its all about optimizing the things I can control. I can control my future paycheck, and I can control how much effort I need to be putting in from week to week. I can't control someone else's salary, and it can't negatively effect me, so why expend the energy thinking about it?

Meanwhile: Responsibility != Effort.

For me its all about correcting mistakes with the least effort possible (I'm lazy you see). Also, if I'm paid the same, and I input less effort, then I'm technically being paid more :)

I would way rather put in the minimal effort required to question and correct (i.e. "manage" as you put it) an error upfront while its an idea, whoever's idea that might have been, rather than to put in the effort to correct the error while shit is hitting the fan.

This is just my personal preference. Your milage may vary.

I love your response, and would only want to add: I only care about how much other people make because it helps inform me in terms of what I can work towards for my own paycheck, and it shows me what the organization values.

As an ex-manager at a very large gaming company, I assure you, your manager doesn't typically make twice what you do unless you're at the very highest levels (the CTO is, of course, going to make quite a bit more than a mid-level engineering manager, esp. if they've gotten a lot of stock and have been around a while) or at a company with some seriously skewed engineering pay grades.

I regularly managed people that made 20-30% more than I did. That was usually corrected within a raise-cycle or two, but it was by no means unusual. I don't think I ever had a direct report that was paid less than 75% of what I got, let alone 50%.

I'm still not doing your job for any less than your pay.

For many it's because their manager is a bit power mad and would rather burn the whole world down than accept they are wrong.

There actually are managers out there capable of managing themselves. They're not as common as we might like, but trust me, they're out there.

Being good at management, is knowing when you don't need to manage at all.

To paraphrase the late Andy Grove, it's mostly about understanding your subordinates and selecting the optimal approach.

The term I've heard is "upwards management." It is (unfortunately?) a valuable skill.

In my experience, there's a lot of two-way communication and decision-making with good managers and good employees (and perhaps even more is required with bad managers and good employees...).

The higher up the org chart someone is, the more high-level their view is, not from ignorance but from necessity. They can easily have ideas that sound great on paper but might not be so great once you account for the nuts and bolts of whatever subsystem(s) would have to change to implement that idea. So if a manager proposes an idea that only sounds good in theory but would be a terrible idea in practice, management needs to rely on their engineers to do their part and explain when and why that is.

It's kind of a dance with a lot of gray areas, because engineers/managers/designers all have their own biases and blind spots and gaps in knowledge, and prioritize things differently. But as long as everyone knows what they're doing and can communicate effectively to people with different backgrounds, it's ideal.

Of course there are plenty of terrible managers who don't know what they don't know and don't take useful feedback. And there are terrible engineers who don't know how to provide useful feedback. But with ideal personnel and a good power dynamic, everyone works together in such a way as to cancel out each other's blind spots and add to each other's expertise.

I think that in a highly skilled workplace, the hierarchical aspect of management accounts for a relatively small fraction of what managers do. For the most part, I treat managers like any other workers. I'm not afraid to have substantive discussions with them, or to delegate tasks to them.

For roughly a year, I actually outranked my boss. I did the engineering, and he dealt with schedules, priorities, communicating upward, etc. It was a great relationship, because we both had jobs to do, and we did them.

In the 21st century, we're getting beyond the point where the hierarchy of authority is accompanied by a hierarchy of knowledge or expertise. In the 19th century business of my imagination, the head clerk knew more about clerking than the junior clerk. A manager was expected to know more about management and business.

Today, people change jobs and fields throughout their careers. It's not unusual for a manager to have subordinates who have been managers themselves, or who have even run entire businesses. People know more about the behavior of the businesses that they work for, than they can learn from management pronouncements, because we can read about the behavior of similar businesses. Managers have diverse staffs. I don't expect my boss to be knowledgeable about my field, because he's a chemist and I'm a physicist.

>For roughly a year, I actually outranked my boss. I did the engineering, and he dealt with schedules, priorities, communicating upward, etc.

That doesn't sound like you outranked him. He scheduled your time, set your priorities, and was gatekeeper to the broader organization.

That's because you're defining 'rank' as 'setting schedules'.

You could just as easily view the situation as the manager providing an interface to the rest of the company, and helping with organisation - somewhat secretarial role.

Yes it does. It sounds like his boss was actually his administrative assistant.

To be fair, more like a project manager, and a damn good one. He could keep things straight in complex projects, and was deft with politics, but without sacrificing his integrity or engineering cred.

That would drive me out of corporate life in a hurry, especially the gatekeeper part.

It was a different kind of relationship. My job comes with a fair amount of autonomy and authority. I'm trusted to communicate with people directly, and consider that to be a vital part of how I do my job.

And the department is "matrix" organized, so there's a lot of cross communication.

Anyway, the rank inversion was temporary, and was just a matter of having different numerical rankings for "technical track" and "management track" employees.

Sounds like you're communicating with your manager.

  > A leader gets an idea. He finds logical, data-backed
  > reasons why the idea is the right course of action. The
  > team scrunch up their faces and ask innocently ‘is this
  > really the simplest way to address this particular
  > problem?’. The founder insists, explaining how it will
  > ‘kill several birds with one stone’. The team nods
  > their heads in agreement and execute the idea. And the
  > problem survives.

I've seen this so many times, As a "leader", I'm guilty of it too. I make sure to preface my suggestions to the team with, "Now, this is probably another terrible manager-idea, but how about we…".

> I make sure to preface my suggestions to the team with, "Now, this is probably another terrible manager-idea, but how about we…".

Does this work for you? I mean, do you get more constructive criticism using this? I tend to think in real life we are programmed to agree with our managers most of the time. Have you received more sincere feedback when you start saying this?

Just indicating that you are okay with being incorrect is enough to garner at least SOME contrary opinions.

Yeah, it makes it easier for people to push back, disagree, or propose alternatives. Part of it is that, as a manager, I really shouldn't be making suggestions for engineering approaches, and I try to defuse the feeling of "my boss told me to do this" by explicitly pointing out that I'm usually wrong.

Probably if is a trend, instead of a single remark here and there..

From a manipulative boss, that sounds like "convince me I'm right."

I try to flip it around and say "here are my assumptions, what information or understanding am I missing? why is this is a bad idea?"

We all miss things due to our own biases, lack of understanding or information, experiences, or even just not being on that email thread. I think my questions have worked because it plays towards people who want to demonstrate how smart or well-informed we are... and we're all there at some point. :)

I just straight up say, here's a bad naive way to do this. How can we improve on it or do it differently? It's unlikely I've come up with the best solution in the world and the people on my team, who are probably smarter than I am, together will be able to get a lot closer

The thing is, when the team ask "is this really the simplest way...?" what they're really saying is "we're not convinced, can we talk about this?" They're trying to open a dialogue. Replying with a stock, amorphous phrase like "kill several birds with one stone" is a conversation stopper.

It's not about whether the idea is really a good or bad idea. It's that the team feel that the proposed course of action doesn't make sense, or that a better one is apparent, but that the leader ignores their concerns. That's going to destroy morale. If, as the leader, you can't even convince your own team that your idea is good, it's probably not good.

To go one step more the team is also acting on their intuition -- they haven't fully used deductive reasoning to come up with their own solution to the problem, it just "smells bad", and now they're coming up with rationalizations (weird that OP calls it strategic reasoning) why it's not a good way.

This is correct.

I was at a particularly dumb company in the early 2000s, when outsourcing was all the rage. It was completely wrong for us, as we were still doing R&D. Yet the folks inhabiting Mahogany Row read Fortune and Forbes and listened to consultants who all said that outsourcing was the new hot thing to do.

And when I kept asking why, the reason kept changing. First it was cost. Then, when it was clear that it would not be cheaper, it was some bullshit about freeing us up to work on tougher problems. And then it changed again.

So of course they did it. And of course it was a disaster.

It looks a bit like impact mapping. One of the techniques I learned from that is to say the opposite of your intuition, and then try to prove it wrong. Obviously, it doesn't prove that your original intuition was right, but it's a useful exercise in critical thought.

For example:

* We'll get more signups by making this button bigger

* Opposite - we'll get more signups by making it smaller

* That's not true because fewer people will see it. But will they? What's if it's just a bit smaller? What if it's a different colour or in a different place or has more whitespace around it?

The 'obvious' original intuition starts to look less obvious.

anecdotal evidence: many of the bad decisions we've made have been data-driven. in other words, we should have listened to our guts more. i think the challenge is knowing when to use data, and when to use gut.

either way, making decisions is very hard. it's basically predicting the future based on flawed information about the past.

You're still making a couple gut decisions when you use data:

- This data was collected in a representative way.

- The problem we are addressing is the same as the one this data investigates.

The first one assessment is a pretty big one, given the possibility of survivorship bias and the silence of users who are only mildly satisfied/annoyed.

Is like:

https://en.wikipedia.org/wiki/Hacker_koan

So Sussman began working on a program. Not long after, this odd-looking bald guy came over. Sussman figured the guy was going to boot him out, but instead the man sat down, asking, "Hey, what are you doing?" Sussman talked over his program with the man, Marvin Minsky. At one point in the discussion, Sussman told Minsky that he was using a certain randomizing technique in his program because he didn't want the machine to have any preconceived notions. Minsky said, "Well, it has them, it's just that you don't know what they are."

> many of the bad decisions we've made have been data-driven

The author of the article would probably say that people's preconceptions influence their data collection in these cases.

yeah, sure. i mean, how meta do you want to get?

at some point you have to draw a line between data and feelings. you can't say that data is really just feelings, so everything is feelings.

A different anecdotal take - from my experience, the problem isn't in knowing when to use data (the answer is IMO "always, if you have it and can afford it") - it's in how to use data. The way I see it typically used is only as an argument to convince yourself you're right all along. You know, people proclaiming "the A/B test shows it works!", without any understanding of what "A/B test", "showing" and "working" even means.

The article wasn't arguing against being data driven, it was warning how you can fool yourself into thinking you are being data driven when in fact you are simply following intuition.

Actually, intuition is actually the cause of pretty much every decision. Think about how people are deciding to vote in the upcoming US presidential election, for example.

Ironically, you just asked us to use intuition in order to support a baseless claim that voters are using only intuition.

No, you intuited that that's what the poster meant. :) It's not like this isn't a well-documented phenomenon.

I can't even tell you how many times this or that politician, analyst, blogger, or forum poster has defended some statement by Trump by saying, "he didn't really mean that", seemingly based on nothing more than assuming he must be way more reasonable than his words suggest, while complaining about people taking him too literally. It's become a cliche in this election. Selectively ignoring someone's words based on an assumption that those words don't reflect their beliefs or character or competence is absolutely relying on intuition.

Selectively misinterpreting them or removing them from context has similar flaws.

I think the solution is even simpler. I'll call it "Market Driven Decision Making"

- Define the problem statement with your customer

- Brain storm solutions with your team and advisors

- Stack rank solutions by time to implement and completeness

- Present the top three solutions to the customer

- Execute on the one the customer chooses

Managers / Leaders should do this continuously in my opinion. Either you are solving a problem for a customer, or your ego. One pays, the other doesn't.

I've always thought that choosing good experiences is always important even if passive, because our intuitions are always undergoing tuning and calibration.

Can you expand on that? I feel like you're onto something, but I don't want to put words in your mouth.

So what is the story with calibration?

I see intuition as something which integrates over hundreds of experiences, weighted by emotion. Arguably, our intuition adapts to our experience to improve our prediction of outcomes, so that, given some situation, we can quickly understand what's going on, and situate ourselves. For the sake of example, if you binge watch TV series about terrorist situations (e.g. 24), you may be calibrating your intuition in terms of survival, bombs, torture etc. As a result, you may be less inclined to reflect thoughtfully, e.g. on whether to buy food which is organic or not, rather buy bas ed on whether it's produced in a country associated with terrorism or whatever. That could be a totes bulls hit example, but my point and theory is that our intuitions are built from millions of tiny moments where we observe cause-effect, hypothesis-experiment-result etc.

The conséquence is that we can strive to improve or cultivate our intuition, by avoiding passive consumption of prepackaged reality like TV and actively interact with real world (however you determine what that is).

I suspect there are a number of ways to support this view, from psychological, behavioral and neuroscience literature, but then I'd be guilty of exactly the thought process the posted article seeks to deride. Nevertheless my degree was in these subjects, so I doubt these ideas are original or unfounded.

aka "use of Effectual Reasoning where Causal Reasoning is more appropriate".

> I built and sold tech businesses in the UK, US and Brazil.

It's always intriguing when people write that without mentioning the names of any businesses.

Reply on news.ycombinator.com