(1) Do incentives actually work for non-geeks? For anyone? (2) Are incentives intended to motivate or control? Does it matter? (3) Are incentives an attempt to overcome perceived lack of alignment between geeks and others? Is that real? Could it be overcome another way? (4) Do we fall back on incentives because that's the way the rest of our society seems to work? (5) Are incentives a way to avoid investing time and energy in understanding people? (6) Why do we think that "they" need an incentive, but "we" are motivated by pride in our work? (7) Do incentives work better when the lower levels of Maslow's hierarchy are unsatisfied? Probably. Does that correlate incentive success with environments of fear? Discuss.
As always, Steve, you cut to the heart of the matter. No two people will have perfectly aligned incentives. Can we get our incentives aligned enough so we can work together? Empirically, it's easy to get our incentives misaligned enough so that we can't work together, even when we're trying to align them.
Agreed. I'm always curious about how much misalignment is real, and how much is perceived, based on stereotypes, or miscommunication. I'm sure both elements are often in play. I don't see many honest conversations about this though - I see incentives (or blatant control) deployed to "fix" lack of alignment without actual investigation and discussion. Those incentives are about control (but we call them motivation).
That's the situation I'm trying to address. Hey our incentives aren't aligned but can we make them aligned enough to work together? Punished By Rewards and Drive are both early influences to this line of thought.
Maybe we're saying the same thing, but I'd use slightly different words. I'd say our goals aren't aligned (initially) and we're using incentives to align our goals. I don't think of it as alignment of incentives.
I understand that you need to have a think, and I don't want to subvert that, but I think there's potential value in sharing my thoughts, since I was falling into a trap of vagueness as well.
When I hear "incentives" I think of what Kohn terms rewards (with punishments being the other side of the same coin). And I believe that *most* rewards are attempts at control. If we all share the same goal, then our actions/behaviors are aligned in pursuit of that goal, and I won't feel the need to control you. If your behaviours aren't what I expect, I'm likely to believe you have some other goal in mind, and I'll feel the need to change (control) your behaviours. By offering a contingent reward (if you do X, then you get Y) I'm trying to make the your goal the achievement of the reward - I haven't aligned our goals, I've tried to align your behaviour to my goal by goal replacement. I guess you could call that alignment of goals, but it's really the alignment of behaviour I want, the reward goal is just a proxy.
I'm sure that's still not clear, but it's different to what I had in mind yesterday.
Daniel Pink's book "Drive" is good reading. It states that "real" motivation should be intrinsic (part of the work) and not extrinsic (external to the work). Three main motivation factors are: Autonomy, Mastery and Purpose. Which means, the more you give your team these three things, the more motivated they are. So, incentives should surround these factors.
What can we give to Geeks to make them more autonomous?
What can we give to Geeks to make them master on something?
What can we give to Geeks to make them feel acting with purpose?
Imagine... you ask geeks what they really want. Imagine... they are able to express their feelings in words. Imagine... you can offer them all those perks and benefits. Will it be enough?
Often in this topic of incentives, I seldom see something that IMHO makes sense: "ask people what they want". Unfortunately, I also believe that making that question is not efficient and that it would be better to request feedback using something similar to: "describe what you are feeling right now (both for up and down moments)". My feeling is that, even when you get all the incentives you want, everything you "demand", both from organizations - bosses (the non-flat-hierarchy) or even professors - while geeks are still in school/academia, it is not enough. It is not enough to keep you in the job, position, motivated to learn. My point being, people don't know: what they really want - the incentives that keep them engaged - what makes them happy. On the other hand, a person knows that she "feels" engaged, happy, satisfied with what surrounds her in the work environment (or bad and want to leave). There is a certain amount of benefits in the midst of a mystique surrounding (that was not asked for, but was attained usually by something/someone external to the geek).
What are folks thoughts on the paradigm of promotion as an incentive? Are we giving individuals control over others as a reward or are we expecting managers to innately know how to bring out the best in others?
At my previous and current employers the distinction between manager and individual contributor was separate from levels and promotion. We were careful to call moving from IC to manager a "transition" and not a "promotion". The two never happened at the same time.
I look forward to hearing shortly what options people are exploring for incentives in this type of organization. This is a many to many relationship as different people are motivated by different incentives. Maybe the path is letting the individual choose the package that best inspires them?
I believe much of our behavior is driven by our old lizard brain pattern-matching according to millions of years of evolution. Then our neocortex just tries to rationalize it all. So as it applies to incentives, I think part of why they can "miss" is because they only ever "hit" at a logical level and not one that pattern matches right against the older, more instinctual part of the brain.
So for example, one answer to the question of "Incentives seem inherently misaligned. How do we make any progress together ever?" might be along the lines of "because we are genetically programmed to work together in small groups in order to survive".
Is the problem embedded in how we structure and govern corporations? Are the prevailing incorporation strategies incentivizing failed "incentive systems?" And, until we change how corporations are formed, will people return again and again to the same dysfunctional operating modes?
I also think that there may be some promise in non-profits, employee owned businesses, B corps, and other innovative alternatives to the common "grow at all cost", "winner take all", "take the money and run" corporate strategies.
This is related to the Goodhart's Law mystery presented above: How can management ever be convinced that there simply are no useful proxy metrics for developer productivity? No numbers to measure and optimize. When their default operating mode is to endlessly hunt for the number that will finally work, can they be taught to move towards qualitative ( non-tayloristic ) assessment? Or is management theory forever mired in the "if you can't measure it, you can't manage it" mentality?
Maybe it's related to the size of the population that the information needs to be communicated to? I've never seen a 20-person shop that bothered with developer productivity metrics, and I've never seen a 1000-person shop that didn't at least try.
Well, from what I see in my issues with motivation:
- There may be $1 000 000 prize for your work but you will rather do something totally different for free just because you are passionate about it right now
- Incentives is an approach to make people want to do the work but there is always something more about really good and "right" involvement. It is like the issue that we are trying to fix is on the other layer of abstraction.
- When you are getting deep into the work and do not notice the time... Man, that's like the other level. You are not doing that for money or appreciation. You are even not spending a split second to think about it. You are doing it because you and the work are whole thing. You just cannot imagine living without this job done.
There's some compelling work done here around safety critical systems; the angle there being that we have incentives to GetShitDone™, but occasionally we push the boundary there so far that we end up endangering either lives out outcomes.
competency-based hierarchies seem to be deeply ingrained in human society and biology. Software engineering competency is difficult to define, but I know it when I see it. A highly competent software engineer brings more than technical competence. A highly competent software engineering manager often is not the most technically competent member of a group.
On a high-functioning team, every member believes they are being materially compensated fairly and adequately for the value they contribute in their respective roles and areas of responsibility. Above and beyond material compensation, they feel a sense of purpose and fulfillment, and experience joy in their work and associations.
I've never seen a hierarchy based on "competency". There are many ways to contribute to a team's accomplishments. Hierarchies typically reinforce whatever power differentials already exist. Those don't align with any dimension of competence.
I agree with your summary of incentives. We collectively seem to be experts at creating incentive systems that induce the exact opposite of the feelings you describe. Geek Incentives is intentionally creating and maintaining aligned-enough incentives to encourage good geeky work.
Several examples of what I consider to be competency-based hierarchies: professional athletes--the best earn the highest compensation and are esteemed by their peers and fans; medical professional--the best command the highest prices, are esteemed by their colleagues, are in high demand and have their choice of where to practice; and even software engineers--the best can pick what they work on, where they work, command their price, and have the esteem of their colleagues and customers.
(1) Do incentives actually work for non-geeks? For anyone? (2) Are incentives intended to motivate or control? Does it matter? (3) Are incentives an attempt to overcome perceived lack of alignment between geeks and others? Is that real? Could it be overcome another way? (4) Do we fall back on incentives because that's the way the rest of our society seems to work? (5) Are incentives a way to avoid investing time and energy in understanding people? (6) Why do we think that "they" need an incentive, but "we" are motivated by pride in our work? (7) Do incentives work better when the lower levels of Maslow's hierarchy are unsatisfied? Probably. Does that correlate incentive success with environments of fear? Discuss.
As always, Steve, you cut to the heart of the matter. No two people will have perfectly aligned incentives. Can we get our incentives aligned enough so we can work together? Empirically, it's easy to get our incentives misaligned enough so that we can't work together, even when we're trying to align them.
Agreed. I'm always curious about how much misalignment is real, and how much is perceived, based on stereotypes, or miscommunication. I'm sure both elements are often in play. I don't see many honest conversations about this though - I see incentives (or blatant control) deployed to "fix" lack of alignment without actual investigation and discussion. Those incentives are about control (but we call them motivation).
That's the situation I'm trying to address. Hey our incentives aren't aligned but can we make them aligned enough to work together? Punished By Rewards and Drive are both early influences to this line of thought.
Maybe we're saying the same thing, but I'd use slightly different words. I'd say our goals aren't aligned (initially) and we're using incentives to align our goals. I don't think of it as alignment of incentives.
I need to think about the distinction between goals and incentives. Thank you.
I understand that you need to have a think, and I don't want to subvert that, but I think there's potential value in sharing my thoughts, since I was falling into a trap of vagueness as well.
When I hear "incentives" I think of what Kohn terms rewards (with punishments being the other side of the same coin). And I believe that *most* rewards are attempts at control. If we all share the same goal, then our actions/behaviors are aligned in pursuit of that goal, and I won't feel the need to control you. If your behaviours aren't what I expect, I'm likely to believe you have some other goal in mind, and I'll feel the need to change (control) your behaviours. By offering a contingent reward (if you do X, then you get Y) I'm trying to make the your goal the achievement of the reward - I haven't aligned our goals, I've tried to align your behaviour to my goal by goal replacement. I guess you could call that alignment of goals, but it's really the alignment of behaviour I want, the reward goal is just a proxy.
I'm sure that's still not clear, but it's different to what I had in mind yesterday.
Daniel Pink's book "Drive" is good reading. It states that "real" motivation should be intrinsic (part of the work) and not extrinsic (external to the work). Three main motivation factors are: Autonomy, Mastery and Purpose. Which means, the more you give your team these three things, the more motivated they are. So, incentives should surround these factors.
What can we give to Geeks to make them more autonomous?
What can we give to Geeks to make them master on something?
What can we give to Geeks to make them feel acting with purpose?
Pink's book is very accessible. If you liked that, I strongly recommend the books by Alfie Kohn that contain a lot of the work Pink is citing.
Imagine... you ask geeks what they really want. Imagine... they are able to express their feelings in words. Imagine... you can offer them all those perks and benefits. Will it be enough?
I'm not sure where you're going with this. I'm listening.
Often in this topic of incentives, I seldom see something that IMHO makes sense: "ask people what they want". Unfortunately, I also believe that making that question is not efficient and that it would be better to request feedback using something similar to: "describe what you are feeling right now (both for up and down moments)". My feeling is that, even when you get all the incentives you want, everything you "demand", both from organizations - bosses (the non-flat-hierarchy) or even professors - while geeks are still in school/academia, it is not enough. It is not enough to keep you in the job, position, motivated to learn. My point being, people don't know: what they really want - the incentives that keep them engaged - what makes them happy. On the other hand, a person knows that she "feels" engaged, happy, satisfied with what surrounds her in the work environment (or bad and want to leave). There is a certain amount of benefits in the midst of a mystique surrounding (that was not asked for, but was attained usually by something/someone external to the geek).
Interesting topic which I'm sure has much debate
I try to go with the simplest (not always correct, but sometimes) explanation
1. Humans are intrinsically good
2. How do you reduce the friction to them doing the right / good thing
3. Can you reduce dissatisifers — admin, compensation, friction
3. Can you increase satisfiers — learning, mastery, autonomy, focus
> People who create incentives are often oblivious to creating incentives. How can they wake up?
Not only that, they are actively hostile to being told that they're creating unintended incentives. WIthout fail.
What are folks thoughts on the paradigm of promotion as an incentive? Are we giving individuals control over others as a reward or are we expecting managers to innately know how to bring out the best in others?
At my previous and current employers the distinction between manager and individual contributor was separate from levels and promotion. We were careful to call moving from IC to manager a "transition" and not a "promotion". The two never happened at the same time.
I look forward to hearing shortly what options people are exploring for incentives in this type of organization. This is a many to many relationship as different people are motivated by different incentives. Maybe the path is letting the individual choose the package that best inspires them?
I believe much of our behavior is driven by our old lizard brain pattern-matching according to millions of years of evolution. Then our neocortex just tries to rationalize it all. So as it applies to incentives, I think part of why they can "miss" is because they only ever "hit" at a logical level and not one that pattern matches right against the older, more instinctual part of the brain.
So for example, one answer to the question of "Incentives seem inherently misaligned. How do we make any progress together ever?" might be along the lines of "because we are genetically programmed to work together in small groups in order to survive".
Is the problem embedded in how we structure and govern corporations? Are the prevailing incorporation strategies incentivizing failed "incentive systems?" And, until we change how corporations are formed, will people return again and again to the same dysfunctional operating modes?
I have enjoyed Gary Hamel and Michele Zanini's recent take on this: https://fortune.com/2020/11/21/capitalism-entrepreneurship-economic-inequality-us-bureaucracy/
I also think that there may be some promise in non-profits, employee owned businesses, B corps, and other innovative alternatives to the common "grow at all cost", "winner take all", "take the money and run" corporate strategies.
This is related to the Goodhart's Law mystery presented above: How can management ever be convinced that there simply are no useful proxy metrics for developer productivity? No numbers to measure and optimize. When their default operating mode is to endlessly hunt for the number that will finally work, can they be taught to move towards qualitative ( non-tayloristic ) assessment? Or is management theory forever mired in the "if you can't measure it, you can't manage it" mentality?
Maybe it's related to the size of the population that the information needs to be communicated to? I've never seen a 20-person shop that bothered with developer productivity metrics, and I've never seen a 1000-person shop that didn't at least try.
Well, from what I see in my issues with motivation:
- There may be $1 000 000 prize for your work but you will rather do something totally different for free just because you are passionate about it right now
- Incentives is an approach to make people want to do the work but there is always something more about really good and "right" involvement. It is like the issue that we are trying to fix is on the other layer of abstraction.
- When you are getting deep into the work and do not notice the time... Man, that's like the other level. You are not doing that for money or appreciation. You are even not spending a split second to think about it. You are doing it because you and the work are whole thing. You just cannot imagine living without this job done.
On the last point:
`What incentives do managers experience that they come to believe that they themselves and their work is “the point”?`
Maybe they experience the incentive of authority, of exercing authority over others?
There's some compelling work done here around safety critical systems; the angle there being that we have incentives to GetShitDone™, but occasionally we push the boundary there so far that we end up endangering either lives out outcomes.
Interesting references:
— https://safetydifferently.com/why-do-things-go-right/
— https://youtu.be/pmZ6wtOmTZU (Dekker talking about IT disasters)
— Foone chatting about a notable analysis with the Challenger spacecraft: https://mobile.twitter.com/foone/status/1095550629919961088?lang=en
competency-based hierarchies seem to be deeply ingrained in human society and biology. Software engineering competency is difficult to define, but I know it when I see it. A highly competent software engineer brings more than technical competence. A highly competent software engineering manager often is not the most technically competent member of a group.
On a high-functioning team, every member believes they are being materially compensated fairly and adequately for the value they contribute in their respective roles and areas of responsibility. Above and beyond material compensation, they feel a sense of purpose and fulfillment, and experience joy in their work and associations.
I've never seen a hierarchy based on "competency". There are many ways to contribute to a team's accomplishments. Hierarchies typically reinforce whatever power differentials already exist. Those don't align with any dimension of competence.
I agree with your summary of incentives. We collectively seem to be experts at creating incentive systems that induce the exact opposite of the feelings you describe. Geek Incentives is intentionally creating and maintaining aligned-enough incentives to encourage good geeky work.
Perhaps it depends on how you define competency. A competency-based hierarchy, based around who is most competent at climbing the hierarchy?
Several examples of what I consider to be competency-based hierarchies: professional athletes--the best earn the highest compensation and are esteemed by their peers and fans; medical professional--the best command the highest prices, are esteemed by their colleagues, are in high demand and have their choice of where to practice; and even software engineers--the best can pick what they work on, where they work, command their price, and have the esteem of their colleagues and customers.