easyfeedback - Complete Guide for Effective Communication and Improvement

Giving feedback that actually changes behaviour is a skill, and like any skill it benefits from structure. The reason most feedback fails is not bad intent — it is vagueness, poor timing, and an emotional charge that puts the other person on the defensive before you have said anything useful. This guide walks through the models and mechanics that turn feedback from a dreaded conversation into a routine improvement tool.

Want expert help putting this into practice? EasyFeedback can guide you through it.

Why unstructured feedback fails

Consider the feedback most people actually receive: "You need to be more proactive." It sounds actionable but it is not. Proactive about what? When? What would it have looked like in a specific instance? The recipient is left to guess, and guessing under criticism usually produces anxiety rather than change. Vague feedback also invites disagreement, because without a concrete example there is nothing shared to point at.

The fix is anchoring every piece of feedback to observable behaviour and its effect, rather than to personality or inferred intent. "You're disorganized" is a character judgment and provokes defence. "The last two reports arrived after the deadline, which pushed the client call back" is an observation, and observations are much harder to argue with.

The SBI model: Situation, Behaviour, Impact

Related: easyfeedback - Complete Guide for Effective Communication and Growth.

SBI is the most reliable structure for delivering a single piece of feedback cleanly. You name the specific situation, describe the observable behaviour without interpretation, and then state the impact it had. The discipline of the model is that it keeps you out of motive-reading.

A worked example: "In yesterday's client call [situation], you answered the pricing question before the client had finished asking it [behaviour]. The client went quiet afterwards and I think they felt cut off [impact]." Notice what is absent — no "you're rude," no "you always interrupt." Just the moment, the action, and the consequence. That gives the person something concrete to work with and preserves their dignity.

SBI works for positive feedback too, and using it that way is underrated: "In the standup [situation], you summarized the blocker in one clear sentence [behaviour], which meant we resolved it in two minutes instead of the usual ten [impact]." Specific praise reinforces the exact behaviour you want repeated, where "great job" reinforces nothing.

The hardest discipline in SBI is keeping the behaviour step free of interpretation. "You weren't listening" is already a conclusion about someone's inner state, and you can't actually observe listening — you observe that they checked their phone while the client spoke. Stay at the level of what a camera would have recorded. The moment you slip from "here is what you did" into "here is what that means about you," the other person's defences rise and the conversation stops being about the work.

The COIN model for harder conversations

When feedback needs to lead somewhere — a change, a commitment — the COIN model extends SBI with a forward step: Context, Observation, Impact, and Next steps. The first three mirror SBI, and the fourth turns the conversation from diagnosis into agreement. Ending with an explicit "what happens now" is what separates feedback that changes behaviour from feedback that merely vents.

The next-steps portion should be collaborative, not dictated: "What do you think would help you catch these before the deadline?" invites ownership. When the person proposes the fix themselves, they are far more likely to follow through than if you hand them a solution.

Neither model is a script to recite robotically. Said stiffly, "In situation X, your behaviour was Y, and the impact was Z" sounds like an HR form read aloud, and people can hear the template. The point of the structure is to organize your own thinking before you speak, so that what comes out is clear and specific — not to turn a human conversation into a fill-in-the-blanks exercise. Prepare with the model, then talk like a person.

Timing, setting, and dosage

See also: easyfeedback - Tips and Strategies for Effective Feedback.

The best model delivered at the wrong moment still fails. A few rules on delivery mechanics:

• Close to the event. Feedback loses precision and relevance with every passing day. Aim for within 24 to 48 hours while the situation is fresh for both of you.
• Praise in public, correct in private. Public correction triggers shame, which blocks learning. Private correction keeps the focus on improvement.
• One thing at a time. Dumping five criticisms at once overwhelms. Pick the highest-leverage item and let the rest wait.
• Ask permission for heavier feedback. "Can I share an observation about the presentation?" gives the person a moment to shift into a receptive stance.

Avoid the feedback sandwich

The classic advice to bury criticism between two compliments is worth abandoning. People quickly learn the pattern and start discounting the praise as mere packaging, so you lose the value of both the compliment and the correction. Worse, the real message gets diluted until the recipient walks away unsure whether there was a problem at all.

A cleaner approach is to be direct and kind at the same time. State the observation plainly, connect it to impact, and show you are invested in the person's success. Kindness lives in the tone and the intent, not in cushioning the message until it disappears. "I'm telling you this because I think you can nail this next time" carries more warmth than a fake compliment ever will.

Turning feedback into measurable improvement

Feedback that is delivered but never revisited rarely produces change. Close each conversation with a specific, observable commitment and a date to check in. "Let's aim for reports submitted the day before the deadline, and I'll check in with you in two weeks" creates a concrete target and a natural follow-up point. At the check-in, acknowledge progress explicitly — noticing improvement is what tells someone the effort was worth it.

Keep a lightweight record of feedback given and commitments made, so patterns become visible over time and reviews contain no surprises. Tools like EasyFeedback help capture these threads and track whether agreed changes actually happened, which turns scattered conversations into a coherent development story. The models give you the words; the follow-through is what converts those words into genuine improvement, one specific, observable behaviour at a time.