Giving feedback that lands: the SBI model

The standup had just ended. Your manager pulled you aside, voice low, expression tight. “I just feel like you’re not being proactive enough,” she said. “You need to take more ownership.”

You nodded. You said you would work on it. You walked back to your desk with absolutely no idea what that meant in practice.

Three weeks later, in your performance review, the same sentence appeared under “areas for development.” You had not changed anything, because nothing specific had been pointed to. You were not defensive. You were genuinely lost.

That is the hidden cost of vague feedback. It is not kind. It is not diplomatic. It is an instruction with no address on it, dropped into someone’s hands, and then held against them when it does not arrive.

What makes feedback vague

Vague feedback has two signatures. The first is that it names a character trait instead of a behaviour. “You are careless.” “You are not a team player.” “You seem disengaged.” These statements describe a quality someone allegedly has, not anything they did. The person receiving them cannot act on a character assessment. They can only argue with it or absorb it as identity.

The second signature is missing context. “That presentation was not great.” When? Which one? What specifically happened in it? A feedback sentence with no timestamp and no scene is not feedback — it is a verdict.

Both forms feel like they are about honesty, but they are actually about avoidance. The person giving the feedback is protecting themselves from the discomfort of specificity. Naming a character trait lets them skip the harder work of observing a behaviour and tracing its downstream effect. It feels direct. It lands like noise.

The SBI model

SBI stands for Situation, Behavior, Impact. It was developed by the Center for Creative Leadership (CCL), a research organisation that has studied leadership development since 1970. The model is deceptively simple: before you give feedback, fill in three slots.

Situation is the specific context. Not “in meetings” as a general statement. The specific meeting, the specific date, the identifiable moment. “In Tuesday’s client review” or “during the sprint planning session last week.” This does the same work that a timestamp does for a log file — it makes the feedback searchable and reproducible.

Behavior is the observable action. Not an inference about intent. Not a character judgment. What did the person actually do or say that could be confirmed by a third-party observer in the room? “You answered the client’s budget question before the project lead had a chance to respond” is a behavior. “You steamrolled everyone” is an interpretation. The test: could a camera have recorded it? If yes, it is a behavior. If it requires mind-reading, it is not.

Impact is the concrete consequence. On the team, the client, the deliverable, the project. Not your emotional reaction framed as an objective fact, but the downstream effect that a reasonable person could agree on. “The client interpreted that as our team not having a single point of contact, and they followed up by emailing three different people” is an impact. “It made me feel undermined” is worth sharing but it is your experience, not a consequence the other person can independently verify.

Put together: “In Tuesday’s client review, you answered the budget question before Priya had responded. The client then emailed three of us separately asking who owned the numbers, which cost Priya two hours of re-alignment work.”

That sentence can be heard. The person receiving it knows exactly what happened, when it happened, and what it cost. They may disagree, but they cannot be confused.

The transformation in practice

Here is the contrast spelled out plainly.

Vague version: “You are careless with the work you hand off.”

SBI version: “In last Thursday’s handoff call, you shared the draft report without running the reconciliation check we agreed on in the kickoff. Rishi spent forty minutes finding the mismatch the night before the client deadline, which meant he submitted at eleven-thirty and missed his son’s school event.”

Notice what the SBI version does not do. It does not say the person is careless. It does not infer that they do not care about quality. It does not generalise to “you always do this.” It names one moment, one observable gap, and one specific cost. The person receiving it knows what to change. They may or may not be a careless person in general — that is irrelevant to this conversation.

The second version also carries more credibility. It is harder to dismiss. It is not your opinion of someone’s character, which they can reject by producing a counter-opinion. It is a description of an event, which they either accept or dispute on its own terms.

The three delivery decisions that make or break it

Getting the content right is half the job. The other half is how you deliver it.

Timing matters more than people admit. SBI feedback works best when the situation is still fresh enough that both parties remember the details clearly. Waiting until a quarterly review to deliver behavioral feedback about something from six weeks ago is worse than not giving it at all — the person cannot reconstruct the context, you cannot either, and the gap between the event and the conversation makes it feel like a stored grievance. In most workplace settings, within twenty-four to seventy-two hours is the right window. Fast enough to be legible, far enough for both parties to have settled down.

Private almost always beats public. This seems obvious until you watch how often it is violated. A correction delivered in front of the team — even an SBI-compliant one — does two jobs: it communicates the feedback, and it demonstrates to everyone else that this is how people are treated here. The second job defeats the first. The person receiving the feedback is now managing their status in the room, not thinking about what to change. Save public feedback for public recognition. Reserve critique for one-to-one.

Ask before advising. This is the least used and most effective addition to the SBI framework. Before you deliver the impact statement, pause. “Can I share an observation from Tuesday’s session?” That single question changes the shape of the conversation. It signals that this is a collaborative exchange, not a verdict. It gives the other person agency in the interaction. And it dramatically lowers the chance of a defensive response, because they walked into the room voluntarily rather than being ambushed.

When SBI is the receiver’s tool, not the giver’s

Most people learn SBI in the context of giving feedback downward — a manager to a direct report. But the model is just as useful going sideways or upward, and it is especially powerful when you are on the receiving end of vague feedback.

When a manager tells you “I need you to be more strategic,” you are allowed to ask for SBI without using those words. “Can you give me a specific example — a situation where you would have hoped for a different response from me?” That is not a challenge. It is a request for the information you need to actually change something.

Most people giving vague feedback are not withholding detail maliciously. They genuinely have not thought through what they observed. Asking for specificity does them a service as much as it helps you. It turns a vague discomfort into a solvable problem.

The limits of the model

SBI does not solve everything. If the underlying relationship is fractured, a well-structured feedback sentence will not fix it. If the organisational culture rewards conflict avoidance and punishes honesty, SBI will feel like a foreign language nobody wants to speak. If the behavior in question involves values violations — dishonesty, discrimination, harassment — SBI gives you structure but you also need a process, documentation, and probably an HR conversation.

The model also does not tell you how hard to push if the person disagrees with your impact statement. That requires judgment, and judgment is not something you can outsource to a framework.

What SBI does is remove the most common failure mode: giving feedback that is actually about the giver’s anxiety rather than the receiver’s development. It forces the giver to have done the observational work before opening their mouth. That discipline alone changes the quality of most feedback conversations.

The reason it is worth the discomfort

Giving specific feedback is harder than giving vague feedback. It requires you to have paid attention. It requires you to have thought about cause and effect. It requires you to say out loud, with precision, something that you would rather smooth over.

But that discomfort is where the developmental work happens — for the person receiving it, and for you. A manager who can give a clean SBI observation is a manager who has actually watched how their team works. A peer who can articulate behavioral impact is a peer who can be trusted with something real.

The goal is not to become a feedback machine or to turn every workplace interaction into a performance review moment. The goal is that when something matters — when a pattern is costing the team something, when someone is on a trajectory you can see more clearly than they can — you have the tools to say it clearly.

“Be more proactive” is not feedback. It is a wish. SBI turns a wish into a conversation someone can actually use.