Why candor is hard at the top
Most CEOs believe they have a candid team. Most teams are candid about many things and careful about a few, and the few usually matter most. Understanding why helps you fix it without blaming anyone.
There are structural reasons. Inside a company, there are incentives to frame things a certain way: a leader who owns a result has a stake in how it is described, and a leader who reports to you knows you influence their future. Nobody wants to look bad, so problems arrive with a plan attached, a reassurance, or a tone that signals they are under control. Busy leaders rarely sit down to understand each other's perspectives, so the disagreements that exist between them are never discussed with you, and often not with each other. And teams sometimes stay quiet when that feels more productive. Raising a concern can cost an hour of debate and a reputation for being difficult, while saying nothing costs nothing in the short term.
The result is that you get a version of reality that is accurate in each detail and misleading as a whole. It is not dishonesty. It is the ordinary behavior of capable people in a hierarchy.
The methods and their trade-offs
No method is the answer on its own. The right mix depends on how much trust you have, how large the team is, and what you are trying to learn.
- One-on-ones: the richest source, because people can explain and you can follow up. They depend on trust, and they carry the reporting-line problem. Leaders tell you what they think you can act on.
- Skip-level meetings: they bring you closer to the field and show you how leaders' messages land. They can unsettle leaders if they feel bypassed, so explain the purpose first.
- Anonymous surveys: cheap and broad, and people say things they would not say aloud. They give you ratings, not reasons, and small teams are rarely truly anonymous. They are good for spotting where to look, not for understanding why.
- Offsites and group workshops: they build shared language and momentum. They reward confident speakers, and people agree in the room more readily than in private.
- Outside facilitators and consultants: they bring structure and a neutral voice. They are slower and costlier, and their findings can arrive after the moment has passed.
- A neutral third party holding confidential conversations: it removes the reporting-line incentive and lets people explain themselves at length. It works best with a clear question and a commitment from you to act.
Question design: ask so the real answer is the easy one
The same person will give you a different answer to a different question. Most unhelpful answers come from questions that invite a performance, such as how is the team doing or are we aligned. Better questions make it easier to say something true.
- Ask for the why. Not whether the integration is on track, but what is making it harder than expected, and why that is.
- Ask each person to define the key words. What does ready mean for this launch? What counts as a customer in this plan? The answers often differ between leaders, and the difference is the finding.
- Ask what they would change first. It moves people from description to judgment and shows what they care about.
- Ask what they think others believe. Where does the sales leader think operations disagrees? People often say through others what they hesitate to say for themselves.
- Ask what they have stopped raising, and why. It finds the topics the team has quietly decided are not worth the effort.
- Ask for an example. A specific recent decision tells you more than a general opinion.
Match the method to the question
It helps to decide what you want to learn before choosing how to ask. If you want to know how widespread a concern is, a survey is efficient. If you want to understand why two leaders see a project differently, only a conversation will tell you. If you want to check whether a message you sent has landed, a skip-level meeting is the most direct route. If the issue involves you or your relationship with the team, a neutral party is often the only way to get a straight answer.
Many CEOs combine methods. A short survey finds the topics, one-on-ones explore the reasons, and a group session turns the findings into decisions. The order matters. Ask privately first, so that the group discussion starts from what people actually think rather than from whoever speaks first. Keep each round small and tied to a decision you are about to make, because open-ended listening exercises with no decision attached tend to produce long lists and little change.
Finally, be realistic about the effort. A good round of conversations takes more of your time than expected, and the follow-through takes more still. It is better to run one focused round well than three broad ones badly.
Build the conditions before you ask
Candor is a response to conditions. Before you ask for it, make three things clear. Say exactly who will see what people say, and keep that promise without exception, because one breach cancels many assurances. Say what you will do with the input and by when. And model the behavior by sharing something you are uncertain about, so asking questions does not look like weakness.
Timing matters too. People are more candid before a decision than after, and they are more candid when the stakes are shared than when they sit on one person's desk. A short, specific question asked early works better than a long review near a deadline.
Close the loop visibly, so people keep talking
Many leaders ask well and lose the benefit by saying nothing afterward. People watch what happens to what they said. If nothing changes and nobody explains why, they learn that candor is optional and quietly stop.
Closing the loop does not mean doing everything people ask. It means telling them what you heard in aggregate, what you are changing, what you are not changing and why, and when you will check again. Do this in specifics. A line like we heard that the handoff between sales and operations is unclear, so Priya and Marcus will own a single process by the end of the month, does more than a general statement of thanks. Names in that example are illustrative.
When you decide against a suggestion, say so directly. People can accept a reasoned no. They find it hard to accept silence.
Pitfalls
A few habits quietly defeat the effort, even when intentions are good.
Most of these come down to one thing: people are always checking whether it is safe to be honest with you, and they update on every signal. A single careless reaction can undo months of patient listening, so treat the first uncomfortable answer as the most important test of the process.
- Reacting to the first piece of bad news with visible displeasure. People will remember, and the next piece will not arrive.
- Treating a single comment as the view of the whole team. Look for patterns across several conversations.
- Asking for input on a decision that has already been made.
- Promising confidentiality you cannot keep, for example by repeating a distinctive phrase in a meeting.
- Using an anonymous survey on a very small team, where people can be identified by role or tone.
- Interpreting disagreement between leaders as a personality clash rather than as different definitions or incentives.
- Collecting input once and treating it as permanent. Views shift as events happen.
Where Halo fits
Halo, HaloVision's AI, gets you what a leadership team doesn't say to its CEO: where leaders define the same words differently and what they would change first. It does this as a neutral third party. Every conversation is confidential, and nobody sees anyone else's answers, so people explain their reasoning at length, and its follow-up questions get sharper after each conversation. It is a conversation with each person, not a survey. Each takes about 30 minutes, and the report lands within 48 hours of the last conversation. It supplements your own listening and does not replace it.