Bounded Specialization (and the limits of human collaboration)

Differences are at the core of any synergetic collaboration effort. The whole ends up being larger than the sum of its part, only when the collaboration leverages the ways in which the collaborators are different, rather than the way in which they are the same.

Consider the difference between two people carrying together a rock that is too heavy for one person to carry on his own, and two people brainstorming a solution to an abstract problem that one of them was not able to solve on his own. The former collaboration utilizes their similarities, while the latter utilizes their differences.

While variation and differences naturally exist in nature, recent studies suggest that specialization, which I’ll define here as the deliberate creation of differences for the purpose of more effective collaboration, is “a human innovation, drawing on our ability to learn and improve by practice, and to trade goods and services”. Ironically, this innovation was inspired by misinterpretation of the ways other animals collaborate.

This innovation has enabled humans to undertake collaboration efforts at unprecedented levels of scope and complexity such as nation states and multinational corporations.

One particular form of specialization seems to be playing a rather pivotal role in such massive collaboration efforts — multi-tiered authority structures, or hierarchies for short.

To avoid a common confusion with “dominance”, Ronald Heifetz, defines authority as “conferred power to perform a service”. Viewed through a specialization-in-support-of-collaboration lens, both parties voluntarily agree to assume “leader” and “follower” roles: “Given your know-how, I give you power to make decisions to accomplish a service, and I’ll follow these decisions as long as it appears to me that they serve my purposes.”

The critical role that authority and hierarchy play in large-scale collaboration efforts is as a mechanism to encapsulate the massive complexity of such endeavors.

But like almost everything that’s taken to its extreme form, specialization, especially of the authority type, has a dark side, which stems from the dichotomous leader/follower distinction.

The dark aspects on the leader side are often summarized as “power corrupts”, but the follower side is not free of sin either. When we fully integrate the follower role into our identity, we are more prone to counterproductive behaviors such as developing “learned helpless”, viewing ourselves as victims, not acknowledging our role in both creating the problem and finding and implementing the solution, and rebelling for the sake of rebellion.

In one of my favorite HBR pieces of all times, Chris Argyris described this dynamic beautifully in an organizational context:

There seem to be two rules. Rule number one is that employees are to be truthful and forthcoming about the world they work in, about norms, procedures, and the strengths and weaknesses of their superiors… Rule number two is that top-level managers, who play an intensely scrutinized role in the life of the company, are to assume virtually all responsibility for employee well-being and organizational success. Employees must tell the truth as they see it; leaders must modify their own and the company’s behavior. In other words, employees educate, and managers act.

How do we manage the dark side of specialization and authority? Well, I don’t think that it’s by getting rid of authority altogether, and “going flat”. That would be throwing the baby out with the bathwater. Until I come across an alternative solution that’s capable of addressing the complexity encapsulation challenge, paraphrasing Winston Churchill’s famous quote about democracy seems appropriate: “Hierarchy is the worst form of organizing, except for all those other forms that have been tried from time to time”.

I can think of three things that we can do instead, all acting to constrain the dynamic in various ways so I’d like to group them under the banner of “bounded specialization”.

1. Bounded by duration

Duration-based elections are a core component of modern democracies and help ensure that authority relationships are not indefinite. Progressive organizations are starting to follow suit as well and elect all members in authority positions, not just their board, to predefined terms.

2. Bounded by context

Distributing authority by narrowing its context can limit its hold on our identity. In a sense, this is an extension of the “separation of powers” concept in democracies where no single body or person has both legislative, executive and judiciary authorities. But it goes beyond that to ensure that we’re followers in some contexts and leaders in others. For example: I trust my Doctor with my medical decisions, and my Financial Advisor with my financial decisions, but I am myself an Educator, so others trust me with the educational decisions for their kids. To bring this back to organizations, this can be done by unbundling the monolithic authority often given to managers (the executive trinity) into separate roles.

3. Bounded by focus

I find this one to be the most challenging to explain, but the most critical to understand, so here goes. The idea here is to blur the dichotomous distinction of roles and responsibilities between leader and follower, manager and individual contributor, or those who hold public office and those who don’t. It’s a little easier to explain in the civil context: consider the difference between a citizen of a country and a resident of a country. They both authorized the government of that country to make decisions on their behalf. They pay their taxes, and they abide by the law. But a citizen also has a right to vote, which requires her to play a more active role in the relationship as both Heifetz and Ret. General Stanley McChrystal point out:

“The concept of social contract may be at the cornerstone of democracy, yet democracy is not so easily achieved in light of our inclination to look to authority with overly expectant eyes. In part, democracy requires that average citizens become aware that they are indeed the principals and that those upon whom they confer power are their agents. They have also to bear the risks, the costs, and the fruits of shared responsibility and civic participation. “ — Ronald Heifetz

“This critical caveat to Tocqueville’s predictions of American democratic success cuts to the heart of what makes democracy tick: a political structure in which decision making authority is — in some ways — decentralized to the voters, rather than concentrated in a monarchic or oligarchic core, requires a high level of political awareness among the public in order to function. If people are not educated enough to make informed decisions at the polls, the feedback system on which democracy is premised will not work. “ — Gen Stanley McChrystal

Now ask yourself this: “how much of our time do we invest in doing our civic duty?” not just paying our taxes and abiding by the law, but playing an active role in the civic process.

For many of us, myself included, the answer is typically “very little, to not at all”. I believe that changing that extreme dynamic in which civil servants are spending 100% of their time on civil matters and many of us spend none of our time on civil matters is critical to managing the dark side of specialization. I’m not advocating for 50–50, that would be missing the point and eliminating all the benefits of specialization. But how about 95–5? Or 99–1? The pattern is applicable to the dynamic within the organization where the distinction is between people who are working in the org (individual contributors, doing the actual work) and people who are working on the org (managers, making the organization better).

I’ll conclude with these unanswered questions: What if every individual contributor spent 5% of his time working on the org? What if every manager spent 5% of her time working in the org?

Bounded Specialization (and the limits of human collaboration)

Goodbye AltSchool, Hello Grammarly!

A few weeks ago, I bid farewell to the friends and colleagues I’ve made over the last couple of years at AltSchool. Today, is my first day with Grammarly and I could not be more excited.

Periods of transition are natural opportunities for reflection, and I did my best to use this one as such.

My tenure at AltSchool, while far from easy, has been a tremendous period of personal and professional growth. For the time being I’ll keep some of the more personal insights private, but on the professional front, upon reflection, a handful of insights bubble up to the top:

  • To create a transformational solution, you have to get the people who experience the pain of the problem, the experts on how to solve it, and the people who can build it, to actively participate in the problem-solving process.
  • Most of our collaboration challenges as adults are rooted in gaps in our own socio-emotional learning. Progressive classrooms are ahead of progressive workplaces in solving that challenge, and a great source for good solutions. Many things that work well with kids, with minor adaptations, also work well with adults.
  • Strong relationships, social connectedness and a sense of belonging are explicit goals in and of themselves.
  • Curiosity and wonder drive self-awareness and personal growth. But sometimes they are dormant and hard to be awakened.

On a slightly more personal note, the past 2 years have been a tremendous validation that continuing to focus my career on the “people” track allows me to bring much more of my unique abilities to my professional life in a way that’s both highly valuable to the organization and highly rewarding to me. Which is part of why I’m so excited to join Grammarly and lead their People function.

But that’s just part of the story. The mission and the team played a key part in this decision. I guess sometimes startup clichés turn out to be true 🙂

First, the mission. Effective interpersonal communication is at the heart of every human collaboration effort. Tying it with one of the lessons-learned above, it’s a good example of a skill that many of us didn’t get to hone and develop fully while at school. The explosion of asynchronous and narrow modes of communication over the last decades, from SMS to Slack, seem to be making matters worse, increasing the likelihood of miscommunication and misunderstanding. Grammarly is taking big strides in growing from its modest-yet-impressive stronghold of Grammatical Error Correction to actualizing this mission in broader and deeper ways, operationalizing cutting edge Machine Learning and Natural Language Processing research in an unprecedented scale.

Second, the team. Grammarly’s EAGER values are deeply ingrained in the organizational DNA and came to life in every human interaction I had with a team member. In addition to making it an amazing group of people to work with, it’s also a testament to the disciplined way in which the current organizational practices were implemented and the importance that the leadership team attributes to organizational health. Spending a few weeks “on the market” in this transition period made me really appreciate how rare this attitude is among the prospective companies that I’ve interacted with during this period, and how essential an ingredient it is to my ability to contribute and add value in the organization that I’m going to be part of.

I am off to starting a new chapter in my professional (and personal) adventure. Can’t wait to see what new experiences and learning opportunities it will unlock!

Goodbye AltSchool, Hello Grammarly!

“Breaking them in” or “Revealing their best” [Cable, Gino & Staats]

Every company reaches a stage, usually very early on in its life as an organization, when it realizes that setting up a new employee up for success doesn’t stop with hiring the right people and then just letting them “figure it out.”

Usually, the next employee touch-point that gets thoroughly thought out and designed is the employee’s first few weeks on the job, often referred to as “onboarding.”

Typical onboarding aims to ramp up new employees as quickly as possible on “how we do things around here” and making them feel “part of the team,” so they have a solid operational and relational understanding of how to get work done in the new company.

An interesting paper by Daniel Cable, Francesca Gino, and Brad Staats suggests that often a powerful positive component is being left out of that experience:

Breaking them in or revealing their best? Reframing socialization around a newcomer self-expression.

In their paper, they argue that “helping employees frame their new role and its necessary tasks as opportunities to use their signature strengths and unique perspectives at work, thereby bringing more of their authentic best selves to the job.” leads to a more positive attitude, improved performance and stronger retention.

They present a field study and a lab study to support their conclusions. The field study in particular is quite impressive.

The authors partnered with a large Indian call center company to run the study and measure its result on the participants’ on-the-job performance and duration of retention. Different onboarding cohorts were given one of three treatments:

  1. Control group — standard onboarding focused on skills training and general company awareness.
  2. Individual-identity group
  3. Organization-identity group

Both individual and organization identity treatments consisted of augmenting with standard onboarding by adding a 1-hour presentation during the first day of onboarding, giving participants two fleece sweatshirts and a badge.

In the individual-identity group, participants were given sweatshirts and badges with their names on them, and the presentation consisted of:

  • 15 min discussion on how working at the company would provide
    each new agent the opportunity to express himself or herself and generate individual opportunities.
  • 15 min to individually complete an exercise that would permit self-reflection in the next part of the orientation session.
  • 15 min thinking about how the decisions they had made in the exercise may have compared with other people‘s responses and answering a few reflection questions.
  • 15 min introducing their best selves to their future work group.
    and discussing their answers and the approach they took to solving the exercise.

In the organization-identity group, participants were given sweatshirts and badges with the company logo on them, and the presentation consisted of:

  •  15 min discussing the company’s values and why it is an outstanding
  • 15 min discussing the company values and why it is an outstanding
  • 15 min individually answering reflection questions on the information they just heard about the company
  • 15 min discussing their answers as a group.

The statistically significant results were quite impressive: participants in both the organization-identity and control groups were more likely to leave the company, compared to those in the individual-identity condition. The odds of turnover increased x2.5 in the organizational-identify and x1.5 in the control group, compared to the individual-identity group.

The work done by the authors suggests that augmenting/tweaking “traditional” onboarding programs, by adding a relatively small “best self” component, can have an outsized positive impact on employees’ tenure at the company.

“Breaking them in” or “Revealing their best” [Cable, Gino & Staats]

Making working relationships work [Mind Gym]

I recently discovered some of the interesting work that the folks at the Mind Gym are doing, through a tip from a colleague. In a domain that’s rife with dogma and (not-really-)”best practices”, I found their approach, which is deeply rooted in behavior science and real data, to be a breath of fresh air.

The content I’m about to cover in today’s post comes from a piece I’ve spent quite a bit of time noodling on:

Do Managers need to be liked? [Webinar]

The webinar covers the “7 talents of effective managers” and spends the majority of the time on one talent that matters more than the rest — “Relate: the ability to create, strengthen and manage relationships of all types”.

Once I got over my general excitement with the research-based approach and the beauty of the framework, I found myself getting a bit stuck on what seemed like a critical assumption that is implied by it: that “management” is a role, and that the abilities and responsibilities covered by this framework are therefore the responsibilities of people in a “manager” position. While this may seem like a trivial/obvious assumption to most people, many organizations these days are experimenting with operating models in which the traditional managerial responsibilities are handled in a more distributed fashion. If this framework requires those responsibilities to be grouped under a single role (embodied by a single person), then its applicability is limited — which makes it significantly less interesting.

But then I decided to test that assumption. I asked myself: what portion of the content will stop making sense if rather than associating these abilities with the manager role, we’ll generalize further and make these the required abilities from everyone in the organization?

To my pleasant surprise, most of it still made a lot of sense. The guidance they offer on how to make working relationship work, for the most part, does not rely on an asymmetrical manager/employee dynamic. Most of it is just as applicable in a peer/peer dynamic.

Even with this big roadblock out of the way, there are still aspects of this approach that I’m still working through. But nonetheless, it’s a great reference point to triangulate one’s point-of-view to and is, as always, best read with a critical eye:

1. Remember why you’re here — we are here to accomplish something together, the relationship is not a means to address our egoic needs (be liked, be the hero, etc.) or an end in and of itself.

2. Keep emotional control — self-regulation trumps all. Identify your emotional triggers and the “alarm calls” that can help you tell that you’re in a triggered state (fight/flight/freeze).

3. Set the boundaries — formally and informally contract to create the right level of intimacy, involvement, and control. Those of you who’ve been following this publication for a while would identify this as a very good example of using polarity management to avoid being too distant or too close:

4. Reinforce and repair the boundaries — we’re all humans. And we all have psychological needs. Sometimes, as a result of that, or other reasons, either we or the other person will overstep the boundary we’ve set up. It’s our responsibility to proactively repair those ruptures.

5. Build trust over the long haul — adopt a long-term attitude. Focus on connectedness, credibility, and consistency.

Making working relationships work [Mind Gym]

The Big List of Class Discussion Strategies [Gonzales]

The Big List of Class Discussion Strategies by Jennifer Gonzales

One of the most humbling lessons from my tenure at AltSchool is that it is often the case that progressive education has leapfrogged progressive workplaces, in accomplishing some shared goals they both have in common.

My approach to tackling a new organizational challenge I just uncovered always starts with some research, to see whether other practitioners in my domain (people) have come across the same challenge and how they went about solving it. Often times they have, and building on their solution leads me to a better outcome than reinventing the wheel, making the same mistakes they have made, and learning from them the hard way.

Nowadays, I try to supplement the search in my own domain, with a search in the school/education domain. Educators are often times trying to tackle similar organizational challenges in their classroom. And what’s working well with kids, tends, with some mild modifications, to work well with adults.

Jennifer’s list is a good example. Large group discussions (often referred to as “meetings”) are an integral part of corporate life. Creating a meeting experience in which participants feel fully engaged is a challenge, regardless of whether the participants are kids or adults. Jennifer offers some good strategies that are worth considering for your next meeting or workshop.

  • Gallery Walk
  • Philosophical Chairs
  • Pinwheel Discussion
  • Socratic Seminar
  • Affinity Mapping
  • Concentric Circles
  • Conver-Stations
  • Fishbowl
  • Hot Seat
  • Snowball Discussion
  • Asynchronous Voice
  • Backchannel Discussions
  • Talk Moves
  • Teach — OK
  • Think-Pair-Share

Give ’em a try!

The Big List of Class Discussion Strategies [Gonzales]

A Compensation Polarity

One of the most fun aspects of curating of and writing in this publication is when a new challenge reveals hidden connections between seemingly unrelated pieces. This story is an example of a recent occurrence tying these three stories that were published more than a year ago:

About a year ago, I took a stab at synthesizing my thinking on compensation at the time:

I’ve argued that compensation is not a hidden intrinsic attribute of an employee, that should be revealed through market mechanisms, but an attribute of the company and the way it values an addition to the team, based on its compensation philosophy (and therefore, it makes total sense that different companies would value a certain individual very differently). I also suggested that a compensation philosophy is evaluated mostly through a “fairness” lens, in comparison to 4 key drivers: market, personal needs and preferences, contribution/value to the company and company success.

Recently, I ran into a compensation question that I wasn’t able to fully reconcile with the framework: Thinking about fairness in external and internal contexts. Let me try to explain. Companies often times hire employees in roles that have very different market rates. For example, the market rate for a junior software engineer can, in certain markets, be 2–3x the market rate for a junior ops person. From an external context perspective, paying these rates seems fair. But from an internal context perspective, comparing the value the software engineer adds to the company and the value the ops person adds to the company, some companies may find it difficult to reconcile the compensation discrepancy. Some companies may not see any issue here, which is totally ok, but is the less interesting use case to explore.

How can a company that experiences a tension with its values and compensation philosophy in the above situation, reconcile what seems to be at first glance like an irreconcilable paradox between “external fairness” and “internal fairness”?

To our rescue comes Polarity Management:

Polarity management helps us see that this tension between “external fairness” and “internal fairness” is not an either/or problem that needs to be solved, but a both/and polarity that needs to be managed.

Going through a polarity mapping exercise can help a company determine where it currently falls on the polarity spectrum, in which direction it wants to move, and what may be the warning signs that it’s getting too close to one pole or the other.

If that sounds a bit too abstract — fear not! Unbeknownst to me at the time, the Buffer salary formula, is a perfect example of the way one company chose to manage that polarity:

A key element of Buffer’s salary formula is the “role” component, which is the calculated as follows:

Role: (overall base + location base + cost of living)* role value

The first three components: “overall base”, “location base” and “cost of living” are, using the distinction I introduced earlier, “external fairness” components — they are set by the external market rate for the role and other factors driven by the location of the position. However, “Role value” is an “internal fairness” component. The Buffer team explains that perfectly:

We don’t agree with the market salary data all the time (for customer service roles, for example) and so we create our own “role value adjustment” based off what we feel is fair.

So here you go: a paradox emerges in a framework, a new tool helps reconcile it, and a real-world example grounds the whole debate in reality.

Reading through old posts is a humbling experience. It’s hard not to adopt the omniscient hindsight perspective and feel a bit embarrassed by the simplicity (and sometimes dead-wrongness) of some of the posts. But I hope to experience the same feeling as I read this post a year from now. Because it’ll mean that my understanding of the world around me has continued to grow, develop and evolve.

A Compensation Polarity

Good Holacracy, Bad Holacracy [Nixon]

I’m enjoying following Tom Nixon on Medium. While I don’t always agree with his point of view, of with the sometimes-controversial way that he chooses to present it, his writing, and further conversations that sometimes ensue, are always a learning opportunity.

Good Holacracy, Bad Holacracy is no different.

In this piece, Tom unpacks his general criticism of Holacracy into a more nuanced point of view that addresses both the good and the bad. Having personally gone through a similar process myself, I’ve landed in a similar position.

Tom describes “Bad Holacracy” as:

when the whole effort to adopt it becomes about Holacracy. Leaders are sold on the Holacracy concept and then a project begins to ‘install’ it… In the worst cases it’s used forcefully, even when the people doing the work don’t believe it will help them. The people are considered the problem. They’re labeled ‘change resistors.’ … In Bad Holacracy, the rules are tightly held… Bad Holacracy seems to value its own processes and rules above all else, and it’s all or nothing… If adherence to Holacracy dogma is valued more highly than doing great work, living the values, and contributing to the vision, you’ve got problems.

And “Good Holacracy” as:

Good Holacracy always works in context. That means solving real issues that are getting in the way of the work and overall progress. It offers process and structure, not to constrain or for its own sake, but to open up creativity and autonomy for people doing the work.

The focus isn’t on selling people on Holacracy and coaching them to use it, but on the underlying principle of self-management. Above all, Good Holacracy doesn’t start with process… The most important thing is the inner work for people collaborating together to better understand themselves and connect to each other. To become clearer in our intentions. To uncover and work through the unconscious biases and stories from our past which shape our behaviour yet do not always serve us… If you add some strategies for org process and structure to an inner journey like this, the possibilities for people to work together to realise worthwhile ideas in the world are enormous.

As a nice bonus, Tom also shares his self-management principles (the Three A’s):

  • Autonomy (for people to take the initiative and make decisions)
  • Accountability (to keep on top of how commitments are being met)
  • Alignment (to ensure everything’s contributing towards an overall vision)

At a high level, Tom touches on two core principles/failure modes in adopting new ways for people to work together (aka “organizational operating systems”). You can easily replace Holacracy with Team, Agile, Lean, TQM or your own pet system:

  1. Treating the system as dogma — viewing the practices as rules that must be followed, rather than suggestions that should be considered and adapted to the organization’s unique context.
  2. Wrong motivation — trying to adopt the practices in order to accomplish some external business outcome (faster time-to-market, higher quality, etc.) rather than in order to better reflect our current or desired shared core assumptions about people and the way they work together.

These two failure modes will lead to the “theater phenomena” (practices adopted with no/marginal benefit) at best or to a complete “organ rejection” at worst.

Adopting a new way of working together has to start with articulating a set of assumptions about people and the way they can work together. If those assumptions are significantly different than the ones that are widely accepted in the organization, the adoption effort must be accompanied by work that exposes those assumptions and helps people grow and develop to be able to adopt those core assumptions. The specific practices that reflect these beliefs need to be viewed as means to an end, and adjusted to the unique context and the progress that the organization has made on that journey.

One of my all-time favorite quotes is a Franz de Waal quote which seems highly relevant to the topic of this post:

The enemy of science is not religion. Religion comes in endless shapes and forms… The true enemy is the substitution of thought, reflection, and curiosity with dogma.

Good Holacracy, Bad Holacracy [Nixon]