The Likability Tax on Technical Leadership
Why women who protect teams get labeled “obstructive,” and how that bias becomes an execution problem
I’ve watched a particular pattern repeat across companies, decades, org charts, and “culture decks.” A woman in a senior IT role pushes back on an unsafe launch, blocks a scope creep ambush, or tells a stakeholder “no” with a full sentence. The room quietly files her under obstructive. Difficult. Not collaborative. Meanwhile a male peer makes the same call, with the same level of force, and gets filed under decisive. Strong. Protecting the team.
If you are a C-level executive in a male-dominant environment, you don’t need a sermon about bias. You need an accurate model of what your organization is doing, because it is doing it whether you intended it or not. And like most systems, it produces the outcomes it is designed to produce.
Research on gender and leadership is clear about the pattern: the same assertive behavior is more likely to be read as “abrasive” in women and “decisive” in men, especially in male-dominant environments. The result is backlash for doing the job the way the job requires.
If you want the short version, it is this: many workplaces evaluate women leaders with a narrower range of “acceptable” power than men leaders. That range is easy to violate simply by doing the actual job.
What “protecting the team” looks like from different seats
In tech leadership, protecting your team is often conflict-shaped work: refusing timelines that push risk into production, preventing on-call from becoming support, and saying “no” to priorities that would trade reliability for short-term wins.
Those moves create friction because they are supposed to. They are boundary-setting in a system that will otherwise default to “yes” until the only remaining lever is burnout.
In many male-coded cultures, a man creating that friction is read as strength in service of outcomes. A woman creating that friction is more likely to be read as friction for its own sake. Same behavior, different story.
You can see this dynamic in miniature in the Heidi/Howard case, where identical accomplishments can produce different “likability” judgments depending on whether the leader is presented as a woman or a man.
That “likability tax” is not a soft concern. It becomes a hiring concern, a promotion concern, a succession concern, and eventually a retention concern. When a leader learns that protecting the team reliably generates social penalty, the organization is training the leader out of the behavior it claims to want.
The executive trap: confusing discomfort with dysfunction
Here is where I see smart executives get fooled.
A woman leader advocates forcefully, the room feels more uncomfortable than it expected, and executives conclude something must be wrong with her approach. But the discomfort is often just the sensation of power being used by someone the culture has unconsciously trained itself not to expect to hold it.
When men set hard boundaries, the discomfort gets attributed to the boundary itself: “That was a tough call.” When women set hard boundaries, the discomfort gets attributed to the woman: “She’s hard to work with.” That attribution shift is the mechanism: the same boundary becomes “a tough call” in one case and “a difficult person” in the other.
You can tell it is happening when feedback gets vague. “Tone.” “Presence.” “Not a team player.” “Could be more collaborative.” These phrases can occasionally point to a real interpersonal problem. More often, they function as an organizational solvent: they dissolve a concrete business disagreement into a personality critique.
A useful forcing function is simple: when you hear critique of “style,” ask for the transcript version. What was actually said. What decision was on the table. What risk was being named. What alternative was offered. If you cannot get to concrete content, you are not doing performance management. You are doing culture reinforcement.
When women undermine each other
Sometimes women participate in the enforcement. That tends to surprise men and it often shames women, neither of which helps. The more useful question is what incentives the environment is creating.
In male-dominant cultures, women can end up choosing between visibility and safety. Some claim power openly and pay a social cost. Others learn to deliver influence indirectly: soften the edges, let someone else “own” the idea, stay legible as agreeable. Both approaches can be rational. Both can work.
The friction starts when a woman who has adapted to indirect influence meets another woman who is direct on purpose. The directness doesn’t just register as a style difference. It raises the temperature of the room, and it can put the “indirect” woman’s hard-won safety at risk. In that context, undermining can become a form of self-protection: align with the dominant expectations, position yourself as the “reasonable” translator, and gain credibility by policing what the system already punishes.
This dynamic shows up in research on what’s sometimes called the “queen bee” phenomenon. The point is not to moralize it. The point is to notice the design: when visible power is penalized, the system creates rewards for people who enforce that penalty.
If you want less of this, the lever is not “women should support women.” The lever is making direct boundary-setting safe and legible, regardless of who does it.
What you can do as a C-level leader without turning this into theater
If you want this to change, you need to change what your organization rewards, and you need to change what it calls “a problem.”
Start by auditing your own interpretations. When a woman leader blocks something, do you treat it as obstruction, or as risk management doing its job? When a man does it, do you call it “protecting the team?” That gap is often the first leak.
Then make evaluation criteria more concrete than vibes. The role’s job is outcomes plus stewardship. Reliability is an outcome. Burnout avoidance is stewardship. “Saying no” is frequently the mechanism.
Finally, treat “protects the team” as an explicit leadership competency, not a personality trait you tolerate when it arrives in the right wrapper. In other words: if you want people to hold boundaries, stop penalizing them for boundary-shaped behavior. If you want an organization that ships without eating itself, you do not need less advocacy. You need more consistent interpretation of it.
The goal is not ideological correctness. It’s accuracy: you want risk decisions evaluated on content, not on who delivered them.
A practical operating practice: the “Decision Transcript” check
Here’s an easy mechanism that prevents a lot of this from hiding in the walls. When feedback uses labels like “bossy,” “abrasive,” “uncooperative,” or “difficult,” require a transcript before it can influence performance narratives:
Decision: What decision was being made
Recommendation: What did they recommend, and why
Risk: What risk were they naming
Alternative: What path did they offer instead
Outcome: What did the business choose, and what happened next
If you can’t produce the transcript, you don’t have feedback. You have discomfort. This single move forces the organization to evaluate leadership where leadership actually lives: judgment under constraints, not social comfort.
It also makes the double standard easier to see. When you compare transcripts across leaders, you will notice how often the “problem” was not the content. It was who delivered it.