Your First Question Is Already an Answer

The question is ready before the meeting

Imagine a finance analyst whose weekly forecast has gone in late three weeks running. Her manager has a one-to-one tomorrow, and the first question is ready: "Why have you stopped prioritising the forecast?"

Nothing has been asked yet. But the question has already decided that she stopped, that it was a choice, and that the forecast was hers to put first.

This article is about the question you ask first, and how it shapes everything that follows.

The short version: A first question can carry a conclusion inside it. When it does, the person answers the conclusion, not the situation. Describe what you've seen, find the judgement in your question, and rewrite it to leave room for an answer you haven't predicted. Then listen, especially when you're surprised.

What one word can do

The wording of a question can change the answer. The classic evidence comes from a setting far from work, so we'll say what it does and doesn't show.

Loftus and Palmer showed 45 students seven short films of traffic accidents. Afterwards, each student was asked how fast the cars were going. Only one word changed. Nine students were asked about cars that "hit" each other. Nine each were asked about cars that "smashed", "collided", "bumped" or "contacted".

The mean estimates differed by word: 40.8 mph for smashed, 39.3 for collided, 38.1 for bumped, 34.0 for hit and 31.8 for contacted. The films were the same. The authors concluded that the form of a question can "markedly and systematically affect a witness's answer". In their case, a single word did it (Loftus & Palmer, 1974).

Be careful with it. The students were estimating speed from films, not answering a manager. The authors also offered two readings: the word may have nudged an uncertain guess, or it may have changed what people remembered. It doesn't show how employees answer managers. It shows how little it takes for wording to move an answer, even when the person watched the event themselves.

Three ways a question carries a judgement

At work, the wording that steers is rarely a single verb. It's the assumption underneath.

It assumes a cause. "Why have you stopped prioritising the forecast?" The words "stopped" and "prioritising" are the manager's explanation, not an observation.

It offers a short menu. "Is it the workload, or is the work no longer a priority for you?" Two options, neither tested. A third, such as a change nobody told her about, isn't on the menu.

It signals the answer you expect. "You're still happy in the role, aren't you?" The easy reply is yes.

Asking like this isn't unkind. It can be an attempt to get to the point, or to help. The question still does the steering.

Here's a test. Could the person answer without first correcting your question? If they'd have to push back on the wording before they can say what happened, the question is steering.

Why it matters

In our view, a question with a judgement inside asks the person to answer a charge first. They defend, or correct you, or agree so the exchange can end. Whichever happens, you learn how they respond to your question, not what happened.

Agreement carries its own risk. "Yes, I've been distracted" can look like confirmation of your theory. It may just be the quickest way out of the room.

In Performance Is a Signal. Not a Diagnosis., we suggest naming the signal and saying you don't yet know why. The first question is where that admission becomes real. Our guide to Behaviour Intelligence puts conversation between understanding the context and making the decision, so the first question is worth planning.

Prepare a question that leaves room

Five minutes is enough.

Write the question you're tempted to ask. Exactly as it would come out of your mouth.

Mark the judgement. Underline every word that explains, labels or assumes. In our example, "stopped" and "prioritising". What's left is the observation: the forecast has gone in late three weeks running.

Ask for an account of one recent instance. "Talk me through last Thursday's forecast, from the first request to sending it." A sequence of events gives you facts to work with. A why-question invites a justification.

Ask one question you can't predict. If you know what she'll say, you're checking, not asking. "What would you want me to know about how the forecast gets built now?"

Plan follow-ups, not a script. Pick two things you'd ask next depending on what you hear. Then leave a pause.

Check your follow-ups too. A good first question can be undone by the second. "So you just didn't have time?" brings the judgement back in through the side door. Listen for the moment you start finishing her sentences.

A rewritten opening might sound like this: "The forecast has gone in late for three weeks. I'd like to understand how it comes together each week, so can you talk me through last Thursday?" It states what you've seen, makes no claim about why, and asks for an account.

When the answer surprises you

Suppose she tells you the forecast now waits on a data feed that moved to a new owner in September. It arrives on Thursday evening, not Wednesday. Illustrative scenario.

It isn't what you expected, and your theory has just lost its footing. The pull is to defend the theory, or to test her answer on the spot. Try three things instead.

Say it's new to you. "I didn't know that. Thank you." It costs little, and it tells her the answer is welcome.

Ask one more question. "What else changed along with it?" A surprise tells you where your picture had gaps.

Check it later, and hold the decision. Test what you heard against the dates and the records after the conversation, not during it. You don't have to accept it blindly. You do have to check it before you choose.

Some surprises won't point away from the person. Part of your view may hold up. The approach is the same.

If what you hear points to the situation around her, What If the Behaviour Isn't the Problem? covers what to do next.

Opptimise helps managers walk into a people conversation ready to listen, not to confirm what they already think.

The first question is part of the understanding. Better understanding comes before better action.

Sources

• Loftus, E. F. & Palmer, J. C. (1974). Reconstruction of automobile destruction: An example of the interaction between language and memory. Journal of Verbal Learning and Verbal Behavior, 13(5), 585-589. doi:10.1016/S0022-5371(74)80011-3. Full text read (Experiment I and Table 1).
• Opptimise, Behaviour Intelligence: Understanding Behaviour Before Deciding What to Do.
• Opptimise, Performance Is a Signal. Not a Diagnosis.
• Opptimise, What If the Behaviour Isn't the Problem?

Drafted and illustrated with AI assistance and reviewed by the Opptimise team.