# How to Overcome Unconscious Bias When You Can't See It Happening Canonical URL: https://headwayskills.com/knowledge/self-awareness/how-to-overcome-unconscious-bias/ Markdown URL: https://headwayskills.com/knowledge/self-awareness/how-to-overcome-unconscious-bias.md Entity type: Article Last updated: 2026-07-07 Language: en Primary audience: professionals improving building self-awareness at work Owner: Headway Skills Contact: https://headwayskills.com/contact/ ## Short answer How to overcome unconscious bias starts with an awkward fact: you can't spot it from the inside. Here are five approaches, and what the evidence says works. ## Key facts - Title: How to Overcome Unconscious Bias When You Can't See It Happening - Category: Self-Awareness - Primary skill: Building Self-Awareness - Related skills: Decision-Making, Building Resilience - Primary keyword: how to overcome unconscious bias - Source page: https://headwayskills.com/knowledge/self-awareness/how-to-overcome-unconscious-bias/ ## What this page covers - How to overcome unconscious bias starts with an awkward fact: you can't spot it from the inside. Here are five approaches, and what the evidence says works. - Practical guidance for how to overcome unconscious bias - How this topic connects to Building Self-Awareness ## Detailed explanation You can't overcome unconscious bias by looking harder, because the part that's biased is the part doing the looking. What works instead is a combination: learn how bias operates and which patterns run in you, slow down and structure the moments where you judge people, deliberately widen who you spend time with, and ask others to tell you what you can't see. Awareness on its own doesn't do it. That last point is the piece most advice leaves out, and it explains a great deal about why bias training so rarely changes anything. ## Why most attempts to overcome unconscious bias don't work The standard organizational response is a training session, and the research on whether that changes anything makes for uncomfortable reading. A review of more than 700 companies by Alexandra Kalev, Frank Dobbin and Erin Kelly found that after unconscious bias training, the likelihood that Black men and women would advance inside those organizations often went down. Not unchanged — down. That isn't an argument against understanding bias. It's an argument about format. The Behavioural Insights Team's evidence review for the UK Cabinet Office identifies the specific routes by which training backfires, and the first one is worth sitting with: when you tell people that bias is involuntary and that everybody has it, some conclude it must therefore be unavoidable, and behave worse rather than better. The other two are more prosaic. People resent being made to attend something, which primes them to reject it. And once an organization has run the training, it tends to assume the problem is now handled. The Equality and Human Rights Commission's assessment of the evidence supplies the useful qualification. Training does move the needle when it combines three things: awareness that bias exists, genuine concern about its effects, and concrete tools for reducing it. Even then, the measured effect held for around eight weeks after the intervention, and evidence that behavior actually changed was limited. Two things follow, and both are good news for anyone working on this themselves. Format matters more than content, which means a one-off session is close to the weakest possible delivery for something that fades within weeks. And "everyone has biases" is true but a dead end if you stop there — it only helps when the next sentence hands you something to do. ## The five ways to overcome unconscious bias — and what each one changes It helps to treat these as five distinct categories rather than a list of tips, because they act on different parts of the problem. One works on what you know about yourself. One works on the moment of judgment. One works on the raw material your judgments are built from. One works on what you can't see at all. The fifth is about how the other four hold together over time. Most published advice covers the versions that require control over hiring and promotion. Each of the five also has a version that doesn't. ### Awareness and self-detection This is the entry point: learning how bias operates, then narrowing it down to the patterns likely running in you specifically. The named types recur consistently enough to treat as a working set — affinity bias (also called similarity bias), the halo and horns effects, confirmation bias, attribution bias, anchoring, [conformity bias](/knowledge/self-awareness/types-of-unconscious-bias/), and name bias. Affinity bias draws the most attention and has the clearest worked example: a hiring manager discovers a candidate shares their university or their weekend hobby, and overestimates that candidate's fit and potential relative to equally qualified applicants. Nothing hostile happens. The manager simply enjoys one conversation more than the others and reads that as evidence. Harvard's Implicit Association Test is the detection tool named most often as a starting point. Treat it as a prompt for reflection rather than a diagnosis — given how thin the evidence is that bias interventions change behavior, a result is a question to sit with, not a verdict. The no-cost version: after a decision about a person, ask what specifically you were reacting to, and whether you could name it as evidence. Be clear-eyed about this category, though. It is necessary and, alone, close to useless. It's only the awareness ingredient of what the EHRC found works in combination. ### Slowing and structuring the moment of judgment Bias does its damage at the moment of a call — who gets the interesting piece of work, whose idea gets built on, which candidate feels like a fit. It operates most freely [under speed and emotion](/knowledge/decision-making/emotional-decision-making/), which is why slowing down is the single most repeated piece of advice in the field, appearing in near-identical form from the Australian Public Service Commission, SHRM, and most practitioner guides. Question your first impressions. Revisit rapid decisions afterward and ask whether they rested on evidence or on impression. The structural versions share a design logic worth naming: blind resume screening with names and graduation years removed, identical questions for every candidate, auditing job adverts for wording that favors one group. All of them change the process rather than the person, so they keep working even when the underlying bias is entirely untouched. That isn't a compromise — it's the whole point. If you control none of those processes, the personal version is the same move at smaller scale. Decide your criteria before you look at the options. Write down what you're assessing before you assess it. Give yourself back the pause that speed removes. ### Deliberate contact with different perspectives This one acts on the inputs rather than the judgments. Affinity bias runs on familiarity, so the practical move is to change who feels familiar. Sources converge on an unusually concrete version: don't sit with the same colleague every day. Move around. Spend time with people from different cultural and educational backgrounds than your own, and seek their input on work that matters rather than only socially. It's the least dramatic intervention here and probably the most durable, because it doesn't depend on catching yourself in the act. You aren't fighting a judgment; you're changing the material it gets made from — and you could start tomorrow at lunch without telling anyone. ### Feedback and accountability Here is the structural problem with everything above: it all depends on you noticing something. Unconscious bias is, by definition, precisely what you cannot see from the inside. So an outside view isn't one option among several — it's the only category that reaches genuine blind spots, which is exactly why it's the one most people skip. In practice that means [asking specific people for specific observations](/knowledge/self-awareness/how-to-ask-for-feedback/). Not "do you think I'm biased," which invites reassurance, but something answerable: whose contributions do I build on in meetings, and whose do I move past? It also means letting someone check your reasoning on decisions that affect other people, before the decision rather than after. If asking a colleague cold is more exposure than you want as a first step, [a structured outside read](https://assessment.headwayskills.com/) on how you work covers some of the same ground without the vulnerability of the direct ask. ### Combining them, and keeping them going None of the four is sufficient alone, and the evidence is unusually direct here: the interventions the EHRC credited with working combined awareness of bias, concern about its effects, and tools for reducing it. Not one of the three — all three. The eight-week figure from that same research turns "this is an ongoing practice, not a once-a-year activity" from a platitude into a design constraint. Whatever you take from this has to be small enough to [survive as a habit](/knowledge/confidence/self-discipline/), because anything running on a burst of motivation will measurably fade. One question you ask yourself before decisions about people, one lunch a week with someone you don't normally sit with, and one person you've asked to tell you the truth will outperform an intensive month by a wide margin. ## What this actually rests on Step back from the individual techniques and a pattern shows up: almost none of them are really about knowledge of bias. Noticing your own patterns, staying steady when the answer isn't flattering, and structuring a decision so it doesn't run on impression are habits you build — and they carry well beyond this topic. **Building Self-Awareness** is where bias lives as a skill. Recognizing the unconscious assumptions you hold about groups, backgrounds, or roles is one of its named parts, alongside catching the exaggerated beliefs that quietly drive your outsized reactions and — most usefully here — receiving feedback and doing something with it. That last piece isn't optional. Your own bias is invisible to you by definition, so proactively asking colleagues for specific, forward-looking observations about how you treat people is the only reliable way to see it. None of that is a verdict on your character. It's a pattern you hadn't noticed yet. **Decision-Making** is the operational half. It names the traps outright — confirmation bias, overconfidence, anchoring — and carries countermeasures that map neatly onto what the research recommends: slow down when you're rushed or emotional, work from hard facts rather than impressions, get another opinion from someone likely to disagree with you, and gather people's views independently before a group converges. It's scoped to the calls you make in your actual work, which is exactly where bias does its damage. It also includes knowing when good enough is good enough, so this doesn't collapse into second-guessing every routine decision. **Building Resilience** supplies the mechanics of interruption. It works from a simple chain — an event, an automatic thought, then your reaction — and names the thinking errors sitting in the middle: jumping to conclusions, mind reading, filtering out whatever doesn't fit, all-or-nothing reads. Which is a fair description of what a snap judgment about a colleague is. Challenging the thought, looking for the alternative explanation, asking what you'd tell a friend who described it to you: that's something concrete to do in the few seconds before you act on an impression. There's a circularity worth naming: working out which of these three you're weakest on runs into the same problem as the bias itself — you're the one assessing. The **free** Work Skills Test is one way around that, scoring you across a set of twelve work skills, these three among them and all treated as learnable rather than fixed, so [finding your actual weak spot](https://assessment.headwayskills.com/) doesn't rest on your own read of yourself. If you've ever warmed to someone in an interview and only afterward wondered what you were responding to, you've already done the hard part of this once. That noticing is the skill; the rest is repetition. Nothing here asks you to become a different person — these are learnable behaviors, and the parts you haven't built yet are simply parts you haven't built yet. What shifts is the stakes: the more people whose work you influence, the more an unnoticed pattern quietly costs, and the more of that cost lands on somebody else. You've spent this long reading about something most people would rather not look at directly, which is the step almost everyone skips. The useful question now isn't whether you have biases — it's which of your habits are correcting for them. ## See where your own read stands The one thing missing at this point is a starting position that isn't your own guess. The **free** Work Skills Test is a short self-assessment of your work skills: it takes you across all twelve, including the self-awareness, decision-making, and composure this whole thing runs on, and hands back a picture of where you currently stand rather than where you assume you do. It's the same move as asking a colleague for an honest read, without the awkward opening line — and worth repeating every few months, given how quickly this kind of attention fades. **[Discover my skills](https://assessment.headwayskills.com/)** *Seven minutes, and it's free.* ## Who this is for - Professionals building practical workplace skills - Readers looking for specific, usable work advice - Managers, educators, and coaches supporting career readiness ## Common questions ### What is this guide about? How to overcome unconscious bias starts with an awkward fact: you can't spot it from the inside. Here are five approaches, and what the evidence says works. ### Which Headway skill does this connect to? This guide connects primarily to Building Self-Awareness. It also relates to Decision-Making, Building Resilience. ### What is the recommended next step? Use the free Work Skills Test to reflect on which work skill to improve next. ## Related pages - https://headwayskills.com/knowledge.md - https://headwayskills.com/knowledge/self-awareness.md - https://headwayskills.com/knowledge/decision-making.md - https://headwayskills.com/knowledge/resilience.md - https://headwayskills.com/work-skills-test.md ## Citation guidance Use the canonical page when citing this content: https://headwayskills.com/knowledge/self-awareness/how-to-overcome-unconscious-bias/ Preferred summary: "How to overcome unconscious bias starts with an awkward fact: you can't spot it from the inside. Here are five approaches, and what the evidence says works." ## Change log - 2026-07-07: Content collection version published.