Holistic Compensation [Barry]

Source: Nathan Barry

Stumbling upon Nathan Barry’s piece, which was referenced in one of the various newsletters that I’m subscribed to, was such a pleasant surprise!

Why I changed my mind on team stock options

 Note that I’ve picked a very different title to my own post, because I’m taking his content in a different direction. But first, a bit of context. 

Compensation matters

Compensation in my mind represents a core attribute of the professional collaboration effort that we call work. The fact that we’re not only trying to do something together, but we’re also trying to distribute the benefits of doing so in a fair way adds a massive amount of complexity to the effort. The thought exercise I always like to go through when thinking about large-scale collaboration efforts (aka work) is “would this still be a problem on an open-source project?”. Open-source projects are massive collaboration efforts in which the primary value that contributors expect out of the effort is the work product itself, which in economic speak is “non-rivalrous”: its use by one contributor does not prevent its simultaneous use by another contributor. Revenue, on the other hand, is a perfect example of a rivalrous good: if I take a dollar out of the pile, it’s one less dollar that can be distributed to anyone else. And that makes things, well… complex. 

Which is why I’ve spent several posts writing about it covering aspects such as the foundations of a good compensation philosophy, the implications of pay-for-performance, the tension between internal and external fairness, the varying levels of transparency around comp, the challenges with equity, and several others. The latter in particular is a good jumping off point to Nathan’s post. 

A holistic approach 

Recognizing similar challenges in the use of equity to the ones I (and Henry Ward of Carta) called out, Nathan initially decided to avoid using equity grants altogether in his startup, ConvertKit, and instead implemented a revenue/profit-sharing system, which we’ll dive deeper into in a moment. Nathan’s post, however, covers his decision to supplement the profit-sharing program with some old skool equity grants. His decision was primarily driven by fairness, recognizing that company value appreciates faster than profits and therefore withholding equity grants leaves his team significantly worse off in the long-term. Nathan’s new and holistic approach, beautifully captured in the 2×2 at the top of this post is in my mind the biggest generalizable lesson from his experience: recognizing the fair compensation needs to span to distinct dimensions, the long-term/short-term and the guaranteed/success-based leads to the conclusion that a different compensation instrument is needed in each quadrant. 

I believe this 2×2 is a great blueprint for other organizations as well, with a couple of tweaks, one mechanical and one philosophical. 

The mechanical tweak has to do with venture-backed companies, who tend to not turn a profit for quite some time. Without fully opening pandora’s box on that matter, profit in Nathan’s framework is simply a proxy for the company’s success. So in situations where it’s not a good proxy, a portion of revenues can be allocated proportionally to a more adequate success metric (revenue target, user growth, margin improvement, etc.). 

The more philosophical tweak has to do with “performance-based” which I’ve replaced with “success-based” since performance is also one of the key challenges in my mind, with ConvertKit’s specific profit-sharing implementation. 


In its latest iteration ConvertKit’s profit-sharing system works as follows: 

A fixed % of profits is distributed to the team every 6 months (the remainder goes to taxes and reinvested back in the business). 

That pool is distributed across 3 categories: 

  • 52% — Team profit sharing
  • 8% — Leadership bonuses
  • 40% —Ownership distributions

Team profit sharing is further allocated as follows: 

  • 25% (13% of total) is allocated based on tenure: your share = your tenure/sum of all tenures
  • 75% (39% of total) is allocated based on individual 0–4 performance rating: your share = your score/sum of all scores

The payout for new hires who started in the last 6 month is pro-rated to the portion of the period they’ve been with the company (3 months = 50% of your share). 

There are no details on the “ownership distributions” so I won’t engage on that. 

But both the “leadership bonuses” and the performance component of the team-profit sharing don’t sit well with me as I strongly believe that rewarding short-term performance does more harm than good (more on that here and here). Furthermore, I don’t think that “leadership”/execs are special snowflakes that require special treatment more than engineers are. They just have a proportionally higher impact on the success of the business, just like a tech lead will have proportionally more impact than an entry-level engineer, and that needs to be baked into the allocation. This is also where the logic behind the decision to ignore salaries, because “salary is a reflection of your market value and not exactly your value to the company”, breaks. While potentially correct looking cross-functionally, there is proportionality between salary and value to the company within a given function. 

Better aligning the allocation with my own compensation philosophy is pretty straight forward. First, eliminate “leadership bonuses”. Second, replace the performance component with a factor that’s proportional to salary. A more egalitarian alternative would be a factor proportional to level, and somewhere in between is applying role-specific factor (a-la Buffer) on top of the component proportional to salary. Further carve-outs to either company value or personal needs drivers can be done similar to the way tenure is handled. 

In Sum 

ConvertKit’s holistic compensation approach represents an important step-function improvement to the default compensation schemes that are out there, and with a few tweaks to the profit-sharing mechanics can be aligned with everything science tells us about the connections between compensation, fairness, and motivation. 

Holistic Compensation [Barry]

The Feedback ↔ Self-Reflection Polarity

Feedback has been a recurring theme in this publication (“Affirmative feedback” and “staying on your side of the net” are good examples) and following the debate I briefly mentioned in “Wise Interventions” I serendipitously came across and authored posts around self-reflection (“Care Pods” and “Cognitive Journaling”).

To catch everyone up to speed, I was having a conversation with a group of colleagues on “the limits of feedback” which really got me thinking. We spend a great deal of time building “cultures of feedback” in our organizations. But can these cultures have a downside?

The case for feedback

A quick glance at the Johari window makes the case for feedback quite clear: 

We all have blind-spots, insights about ourselves that are known to others but are not known to us. Feedback is the mechanism to disclosing those blind spots, which in turn allow us to build a more accurate picture of ourselves to drive our growth and development. 

Furthermore, as Adam Grant pointed out, our own capacity for self-reflection is bounded and imperfect:

Sixteen rigorous studies of thousands of people at work have shown that people’s coworkers are better than they are at recognizing how their personality will affect their job performance.

Lastly, since we’re talking about feedback in a professional context, with the intent of collaborating better together, how others perceive us and react to us matters just as much (if not more) as we perceive ourselves. This in and of itself is a good enough reason to pay attention to feedback and to take it into account. 

The case for self-reflection

Despite the compelling case for feedback, it’s not without its shortcomings. 

To reference the Johari window again, some of our knowledge of ourselves is only known to us (“hidden area”) so any feedback will be based on partial information and will, therefore, be incomplete and inaccurate.  

Furthermore, delivering good feedback requires a high level of mastery, from avoiding projecting our own values and beliefs on the person receiving the feedback to only addressing the things that we can credibly observe and know (staying on “our side of the net”). 

Lastly, while self-reflection and self-knowledge will never be perfect, our ability to more accurately know ourselves and how we impact others is a learnable skill, a muscle that we can build. An environment abundant with feedback or a culture focused only on feedback will crowd-out any motivation to strengthen that muscle and will leave it weakened and atrophied. 

So which one is it?

Astute observers will recognize the patterns of a polarity in the tension between self-reflection and feedback. Therefore, the answer is not an either/or one but a both/and one. We need to create cultures that help our teams build both their feedback muscles and their self-reflection muscles. Favoring one at the expense of the other will lead to a sub-optimal outcome. 

Side note: for those of you looking for a deeper scientific analysis of this tension, Timothy Wilson and Elizabeth Dunn’s Self-Knowledge: Its Limits, Value, and Potential for Improvement may be a good place to start. 

The Feedback ↔ Self-Reflection Polarity

Cognitive Journaling [Ragnarson]

Continuing our recent arc on feedback/self-reflection, this piece by Richard Ragnarson does a great job introducing in detail one highly-effective self-reflection practice: 

Cognitive Journaling  

Journaling is making a comeback these days alongside specific journaling techniques and (obviously) customized journaling products. While some journaling techniques aim to be more forward-looking (aka “planning”) others are more reflective. The good news is that the hype is leading to more innovation in the space and making effective techniques more accessible. While the obvious downside is the increased difficulties separating signal from noise: techniques that are truly effective from ones that are merely popular. 

Ragnarson’s technique sits on a very solid evidence-based basis in the form of Cognitive Behavioral Therapy (or CBT for short). His article is an extensive primer on the technique, walking the reader through the motivation behind the technique, the model of the mind on which it is based, key constructs, high-level principles, a step-by-step guide to the process, a practice program for developing habit and mastery, ways to measure progress and last but not least — an FAQ and troubleshooting guide. Incredible work just putting all of this together. 

In this post, I will only cover the high-level principles and the technique itself. If it resonates, reading the whole post by Ragnarson is highly encouraged. 


  • Falsifiability — describe internal and external facts. Facts can be falsifiable with a yes/no question on whether it happened or not. Falsifiable: “I only have two hours per day to work on my project”. Not-falsifiable: “I have no time to work on my project.”
  • Nonjudgment — describe events, thoughts and feelings, avoiding inferences/deductions regarding their possible causes. Nonjudgment: “I feel demotivated”. Judgment: “Feeling demotivated is bad.”
  • Detail — describe contexts, events, thoughts, emotions, and behaviors with as much detail as possible, while being mindful of not violating the first two principles. 

Putting it all together — journaling while following the principles: 

I went to the supermarket. I met my boss Chris by chance. We spoke and he brought up my work. I thought, “Why can’t he leave me alone even when I am not at work?” I felt annoyed. I thought, “I don’t like feeling like this.” I felt angry. I thought, “I can’t stand getting annoyed anymore,” and then I thought, “I need to change jobs.”

Journaling while not following the principles: 

I was out and met Chris; he’s such a jerk. I can’t stand dealing with him. I need to quit this job.”

The ABC Process

ABC refers to a model of cognition based on the view that any life experience is constituted of a series of activating events, beliefs, and consequences (ABCs): Activating event → Beliefs → Consequences (emotions + behaviors).

The journaling process, however, follows a different sequence: 

  1. Start with the C (consequences): emotions and behaviors: writing down the emotion or behavior that you want to reflect upon, in the form of “I felt [insert emotion]” or “I did/behaved [insert behavior], applying the three principles (falsifiability, nonjudgment, and detail).
  2. Describe the A (activating event): Describe the situation you were in when you experienced the consequence from before, in the form of “This [insert event] happened” or “The situation was [insert situation or place], applying the three principles.
  3. Find out the Bs (beliefs): With the consequence and activating event at hand, try to remember the thought that you entertained in your reaction. Express it in the form of “I thought that [insert belief]”, applying the three principles. 
  4. Challenge the Bs (beliefs): You challenge a belief by evaluating its validity, doubting it, and finding a better alternative. Consider its flexibility, logic, congruence, and usefulness. 
  5. Write down good alternative Bs (beliefs): Ask yourself: Which alternative thought can I think? Which alternative thought is logical, reality-based, flexible, and useful in pursuing my goals and feeling good? The following table, also created by Ragnarson, illustrates this distinction well: 
Source: Ragnarson

In sum

Ragnarson did an incredible job putting together a comprehensive and detailed guide for cognitive journaling, addressing many of the nuanced points needed to start building this powerful self-reflection having and strengthening our self-reflection muscle. 

Well worth a read!

Cognitive Journaling [Ragnarson]

Fostering responsible action by peers and bystanders [Rowe]

I was hoping to write the post about the feedback ↔ self-reflection polarity this week, but upon attempting to do so, realized that it needs a bit more percolation time. So instead, I’m picking something slightly less cognitively taxing (for me). 

Still connected to the meta-theme of the previous post around diffusing the monolithic single-hierarchy org structure, I strongly believe that key behaviors that are typically attributed to leaders (at the top-tiers of the monolithic hierarchy) are in fact, basic acts of “good corporate citizenship” and we’ll be better seeing these behaviors democratized/spread out throughout our professional working community. While specialization/division-of-labor is essential to any large-scale collaboration effort, the purist form in which it is typically practiced has some painful drawbacks (more on that here). 

This rather abstract preamble is just meant to set the context that existed in my head when I encountered Mary Rowe’s work, and specifically: 

Fostering Responsible Action with Respect to Unacceptable Behavior: Systemic Options to Assist Peers and Bystanders 

Because it’s a concrete example of the more abstract point I was making above. Conflict resolution and dealing with violation of the group’s laws, norms and code of conduct is often viewed as the job of HR and managers. And it is. BUT. That does not mean that anybody else in the org, namely, peers and bystanders, don’t have a role to play as well. But as we all know too well, whether peers and bystanders will act is in many ways a byproduct of the context or system in which they operate. Under one set of circumstances, they tend to act. Under a different set, they won’t. Rowe set out to identify the attributes of the system that will increase the likelihood of peers and bystanders taking responsible action. In her own words: 

Peers and bystanders are important in organizations and communities. Peers and bystanders can help to discourage and deal with unacceptable behavior. They often have information and opportunities that could help to identify, assess and even manage a range of serious concerns. Their actions (and inactions) can “swing” a situation for good (or for ill)… [bystanders] often have multiple, idiosyncratic, and conflicting interests — and many feel very vulnerable. As a result, many potentially responsible bystanders do not take effective action when they perceive unacceptable behavior. Bystanders are often equated with “do-nothings.” However, many bystanders report thinking about responsible action, and say they have actually tried various responsible interventions… Many peers and bystanders might do better if they had a conflict management system that takes their needs into account. A central issue is that peers and bystanders — and their contexts — often differ greatly from each other. As unique individuals, they often need safe, accessible and customized support to take responsible action, in part because of their own conflicting motivations. They often need a trusted, confidential resource. They frequently seek options for action beyond reporting to authorities.

She first defined or decomposed taking action as a 4-step process: 

  1. Perceiving behavior that may be unacceptable
  2. Assessing the behavior
  3. Judging whether action is required
  4. Deciding whether and how to make a particular personal response (or responses.)

Which in turn allowed her to distill the key reasons for why bystanders do not act or come forward: 

  • The bystander does not “see” the unacceptable behavior
  • The bystander cannot or does not judge the behavior
  • The bystander cannot or does not decide if action should be taken
  • The bystander cannot or does not take personal action

Conversely, bystanders do take responsible action if: 

  • They see or hear of behavior they believe to be dangerous, especially if it seems like an emergency, and especially if they think that they or significant others are in immediate danger
  • They perceive that an apparent perpetrator intends harm, especially if that person is seen to have hurt or humiliated family members or people like themselves
  • They wish to protect a potential perpetrator from serious harm or blame 
  • They are angry, vengeful or desperate enough to ignore the “barriers to action”
  • They are certain about what is happening, and they believe they have enough evidence to be believed by the authorities

With the spectrum of drivers that discourage and encourage actions more clearly mapped out, she was able to identify and prescribe 8 systemic leverage points that are likely to create the context that will encourage bystander action: 

  1. Provide training and discussions sponsored and exemplified by senior leaders
  2. Build on safety and harassment as issues of special importance
  3. Share frequent and varied success stories
  4. Appeal to a variety of socially positive motives
  5. Discuss the potential importance of imperfect “evidence”
  6. Provide accessible, trusted resources for confidential consultation
  7. Provide safe, accessible and credible options for action
  8. Improve the credibility of formal options

If this is a topic that’s particularly relevant in your own organization, the full paper is well worth the read. 

Fostering responsible action by peers and bystanders [Rowe]

Care Pods [Enspiral]

Source: Boyatzis (2006)

A big hurdle in adopting alternative organizational “operating systems” (roles, responsibilities, etc.) instead of the traditional, single-hierarchy, authority-driven system has been the incomprehensiveness of those alternative systems. 

Some core elements of the collaborative efforts are simply left unaddressed by the alternative systems. Oddly enough, those gaps often have to do with the human aspects of the collaborative effort: compensation, performance management, hiring/firing, professional development, etc. — you know, the “easy” stuff…

So when I stumble upon a practice that seems to be filling some of these gaps, without relying on the traditional structures — it is certainly worth sharing.  

And such is the case with Enspiral’s Care Pods which offer a very compelling alternative for driving personal and professional development in a way that is not dependent on a manager (as-a-coach) role, or cumbersome feedback cycles. 

At their core, Care Pods aim to “operationalize”, or implement, Richard BoyatzisIntentional Change Theory (ICT) through a series of 8 sessions (that can then be run iteratively, in perpetuity) carried out by a small group of 4–6 people (aka the Core Pod). The slightly more detailed session plan with high-level agendas for each session and supporting exercises, is available in the original doc but here’s the summary of the summary: 

  • Session 1: Overview
  • Session 2: Getting started
  • Session 3: Ideal self
  • Session 4: Real self 
  • Session 5: Developing a learning plan for change
  • Session 6: Implementing the learning plan
  • Session 7: Care pod retrospective
  • Session 8: Iteration

To me, this seems implementation-ready in its current form and already a massive step up compared to the way 99% of organizations are handling personal and professional development today. It’d also offer a few additional tweaks to make it even better, in my opinion at least: 

  • I do think that this approach skews a bit too heavy towards the “self-reflection” pole of the “feedback”<->”self-reflection” polarity (more on this next week). What this means in practice, is that a thoughtful, well designed peer-feedback exercise that extends beyond the members of the Care Pod and carried out between sessions 3 and 4 can provide fantastic fodder for the formulation of a more accurate “real self” picture in session 4, leading to a more effective learning plan in session 5.
  • I would also either extend or split session 5 to create the space to introduce Immunity to Change as a core framework for understanding our “default” behavior, and using it to design behavioral experiments that are more likely to yield the change that we seek.
  • Lastly, I’d sprinkle in 1–2 purely “social” sessions, to strengthen connections between the Care Pod members in a more informal setting (drinks, dinner, some other outside-the-office activity). 

Net-net this is a fantastic practice that I’d be eager to implement in either the future org that I’ll join or the organizations that I’ll be consulting with. 

Care Pods [Enspiral]

Inclusive hiring: a short primer

As the debate about diversity metrics and quotas rages on, I’d like to share my attempt to find common ground and a path forward. 

To do that, let’s start by defining our “north star” first: 

A fully inclusive hiring effort is an effort in which we engage and attract all relevant candidates for the role, evaluate them fairly for exactly what the role requires (nothing more, nothing less) and give them a clear picture of what working at our company is like, so they can evaluate the opportunity fairly.

Note that it doesn’t include any references to diversity, identity, minority, etc. 

Now we can ask: what gets in the way of this ideal end-state? And the answer: our own “humanity”. Our susceptibility to certain biases in our thinking and actions which eventually manifest themselves as selection bias: either we end up selecting/rejecting candidates, or candidates selecting/rejecting us based on attributes, knowledge or actions that have no impact on their ability to do well in the role that we’re hiring for. 

Selection bias tends to creep up across 4 different dimensions of the hiring process. While they may have some overlap between them and are not fully mutually exclusive, discerning between them helps move us forward: 

  1. The way we attract/reach out to candidates
  2. The way we define what success in the role requires (and doesn’t require)
  3. The way we conduct the assessment of the candidate’s performance
  4. The way we evaluate the candidate’s performance in the assessment

With these dimensions in mind, we can now consider specific hiring practices and articulate their impact on helping us create a more inclusive hiring process. While these practices have a compounding impact when used together, they have little dependencies between them and can certainly be used in a more piecemeal or a-la-carte way.  

The list below is not comprehensive and I’m continuously adding to it, but I believe it to be a good start: 

Inclusive hiring: a short primer

If we know in which direction we want to go, does it matter where we are?

And for that matter, to where exactly we’re trying to get? 

One use of numerical measurement: to describe direction

Continuing to reflect on some of the topics that I’ve covered in the last couple of posts, I want to spend more time today on numerical measurement and its alternatives. 

While the answer to the question posed in the title may seem like a resounding “yes”, I’d like to suggest that in some circumstances, it may actually be “no”. 

To be clear, I’m not a measurement or metrics hater. And this is not a rant about metrics. But given that I’ve spent quite a bit of time calling out their deficiencies, I owe it to myself and others to offer alternatives that are not meant to completely replace them, but to expand the toolbox so we can use the more effective tool to the problem at hand. 

I won’t repeat my whole case on the challenges with numerical measurement but will just briefly mention that they typically elicit some non-trivial “operational” challenges in both the collection and interpretation of the data. And perhaps more importantly, they also pose some more “strategic” challenges — they reduce a highly complex reality into a very simplified representation. Sometimes that’s incredibly helpful — separating signal from noise and creating clarity on what’s truly important. But often times over-simplifaction leads to solutions with both cognitive and behavioral flaws when applied back in the complex reality. 

The alternatives to this approach depend on the purpose we were trying to accomplish with numerical measurement begin with, something that I’ve noticed I haven’t given enough attention to in the past. Introducing some distinctions there helps to identify viable alternatives, at least in the two cases outlined below. 

Measurement to articulate direction

One common use case of numerical measurement is to articulate direction. By describing where we are right now and where we want to get to, we implicitly define the direction in which we want to go: 

  • We want to get from point A to point B = we need to drive in direction C
  • We want to improve margins from 13% to 15% = we need to improve margins/become more efficient
  • *nerd alert* describing a vector using the coordinates of its start and end points

Often times the start and end points are rather meaningless in and of themselves. It’s the direction or delta between them that matters. 

Yet describing the start and end point are not required to describe the direction. I can still “drive south” without saying “get from SF to LA”. Not the perfect example, I know, but hopefully still gets the point across. 

So alternatively, we can use either “even over” statements or slightly more detailed polarity maps to describe the direction we want to go. 

Measurement to choose between options

Another common use case of numerical measurement is to choose between options. 

For example, choosing what driver to focus on next in order to improve employee engagement. This is often done by having employees rate each one of the drivers using a Likert Scale, translating each rung in the scale to a numerical score and sorting the drivers from the lowest rated to the highest rated or from the one that worsened the most to the one that improved the most. 

The distinction between cardinal and ordinal utility can help us find an alternative. We don’t even have to go into the debate about the feasibility of truly measuring the cardinal utility of each one of the drivers and simply say that since we’re only using the measurement to choose between the options, the ordinal utility is sufficient. And in that case, there’s a simpler alternative to numerically rating each one of the drivers in isolation: asking employees to stack-rank the drivers from the one we should focus on the most, to the one we should focus on the least. 

If we know in which direction we want to go, does it matter where we are?