Jacob Kaplan-Moss

Three Feedback Models

Giving feedback is perhaps the most important part of a manager’s job. Feedback is the main mechanism by which managers help the folks on their team improve, so effective feedback leads to individual career growth and better team results: win-win.

But giving feedback effectively is difficult. Even under the best of circumstances, talking about performance is awkward. Done incorrectly feedback is even worse: inaccurate, unfair, or poorly-delivered feedback (rightfully) leads to resentment and a breakdown of trust.

I find models, templates, and scripts to be incredibly helpful in potentially-difficult situations like this. So here are three models that I like for delivering feedback. Each is valuable on its own and would make a great starting point for anyone who wants to build their feedback muscle. Together, they highlight some common factors in effective feedback models and show off a couple of “special features” that can help your feedback be particularly effective.

Two general notes before we dive into the specific models:

  1. Remember that feedback can be either positive (“that thing was effective; keep doing it”) or negative (“that thing wasn’t effective; please stop”). All too often we associate feedback with just the negative. This is wrong. Positive feedback alone can be surprisingly effective! And, most feedback you deliver should be positive; studies seem to show that feedback is most effective when the ratio is about 5 pieces of positive feedback to each piece of negative feedback.

    For that reason, I’m going to use a piece of positive feedback – imagining someone on my team wrote a really good pull request review – as a running example throughout this article.

  2. Feedback (like any communication) is personal; the most effective style will depend on both parties in the communication. Small differences in wording can make a big difference in effectiveness. While there are wrong ways to give feedback, there’s no single “right” way. The goal of this article isn’t to get you to use one of these models as-is. You may want to use one as a starting point, but my goal is to highly the similarities and special features of each so that you can work out a system that works best for you and your team.

With that: let’s dive into the models:

Situation-Behavior-Impact (SBI)

The Situation-Behavior-Impact model, developed by The Center for Creative Leadership, is very simple and focuses tightly on behavior and impact. It has three steps:

  1. Situation: describe the situation, e.g. “In your review of Jane’s pull request…”
  2. Behavior: describe the person’s actual specific behavior, e.g. “… you gave her clear advice on improving her test coverage …”
  3. Impact: “.. which helps us in meeting our team’s goal of producing better-tested code.”

Read more:

Manager Tools Feedback Model

The Manager Tools model is also behavior- and impact-focused. It adds two additional pieces: asking for permission to give feedback, and asking for next steps. Their model has four parts:

  1. Ask for permission (and respect a no): “May I give you some feedback?”
  2. Describe behavior, e.g. “When you gave Jane advice on test coverage …”
  3. Describe impact, e.g. “… this helps her, and us as a team, produce better-tested code.”
  4. Next steps: for positive feedback, thank them, e.g. “thanks, keep it up”. For negative feedback, ask for a change, e.g., “can you do something different next time?”

Read More:

Lara Hogan’s Feedback Equation

Lara Hogan, an author, public speaker, and coach for managers, has a Feedback Equation that is quite simple:

  1. Observation of behavior, e.g. “In your review of Jane’s pull request, you gave her clear advice on test coverage…”
  2. Impact of the behavior, e.g. “… this helped her improve her code, which helps with our team’s goal of better-tested code.”
  3. Question or Request, e.g. “thank you; more like that!” (for positive feedback) or “in the future, can you … instead?” (for negative feedback).

Read more:

What these all have in common

You’ll notice that these models are all pretty similar. That’s not an accident; they all share some core concepts of effective feedback:

  • They’re behavioral: focused on something the person did or said (rather than who they are, or your interpretation of their behavior) about something the person did or said,
  • They’re impact-focused: they have you discuss the impact of what the person did, rather than their intent. By focusing on impact they also require you to think about and understand that impact. If the behavior didn’t have an impact, it doesn’t need feedback!
  • Each model encourages sharp focus on one specific behavior and impact. Vague or overly broad feedback isn’t effective. If you tell someone “you’re rude” they’ll probably get defensive (justifiably so), and even if they don’t, they’ll be highly unlikely to know what to do to change it. If, on the other hand, you tell them “you interrupted James four times today”, the fix is obvious.
  • Each model is very short; any of these can be said in under a minute. Sometimes feedback can (and should) lead to a longer conversation, but much of the time it’s best to simply give the feedback and give your direct space to process and take it on board. Nearly every study on feedback shows that more feedback is better, so if feedback delivery is very short, you’re more likely to do more of it.
  • They encourage the same format and tone for both positive and negative feedback. Feedback is always about future behavior (more on this in a moment). Your tone should always be pleasant and optimistic – remember, forward-looking. Using the same format means you can practice this tone on positive feedback, so when you need to give some negative feedback you’ll have an effective script ready to go.

Special features that I like

I find SBI to be the most basic and easy-to-learn model. The other two have a couple of extra “special features” that I find particularly valuable:

  1. Asking for permission (MT) before giving feedback is brilliant. People aren’t always in a place where they can hear feedback, so making sure they’re ready to hear feedback ahead of delivering it makes the feedback much more important. And, asking for permission first implies – correctly – that feedback is as much for the direct as it is for the manager.

    If you do this – and you should – remember that honoring a “no” is incredibly important. If you give someone a choice, it needs to be a real choice; overriding a “no” will destroy any trust you’ve built.

  2. Ending with a request (MT & LH) helps drive the feedback towards something that’s future-oriented. Feedback should never be about punishment or litigating the past; it’s all about the future. Ending with an explicit request for the future drives this message home. It also helps frame any potential post-feedback conversation around “what’s next”.

    The Manager Tools model suggests, for negative feedback, that rather than making a specific suggestion you instead leave the specifics up to the direct by saying something like “can you do that differently next time?” This can be quite effective: usually, telling people what to do doesn’t introduce a lasting change as well as letting them come up with the specifics does.

    I’ve found, though, that sometimes I do need to be directive – especially with more junior staff, or with more serious performance issues. I suggest trying to be open-ended but don’t be coy about specific requests if you’ve got them.

What works for you?

If you’re looking for a way (or a better way) to give feedback, I hope these models help. I’d love to hear about how these work for you: if you’ve used these models, how well (or poorly) do they work? Are there other models you’ve found useful? What situations have you seen where one model works differently? Get in touch - <jacob @ this domain> - and thanks!