Better Alternatives to the Feedback Sandwich

The feedback sandwich feels like a kindness — wrap the hard truth in two slices of praise and the medicine goes down easier. Research says otherwise. Studies find that recipients of sandwiched feedback walk away thinking they're doing well, praise delivered alongside criticism becomes untrustworthy, and performance measures don't improve. The good news is that structured, direct alternatives exist, and they work better for managers and employees alike.

What the Feedback Sandwich Is and Where It Came From

The compliment sandwich technique, popularized in the 1980s, involves placing a piece of constructive criticism between two positive comments. Mary Kay Ash popularized a version of the approach in the 1960s when she founded Mary Kay Cosmetics, at a time when corporate feedback was rare and often delivered harshly. The cushioning instinct made sense in that context. Modern workplaces, however, have different needs: employees expect candor, psychological safety, and feedback that actually helps them improve.

The formula has persisted for decades largely because it feels considerate. Managers avoid the discomfort of delivering a blunt critique, and the structure gives them a script to follow. But predictability is precisely the problem.

Why It Backfires: Receivers Hear Praise, Not Correction

Researcher Ayelet Fishbach divides students into feedback givers and receivers; givers are told to communicate that the receiver's performance is unsatisfactory, that improvement is needed, and that there are concrete ways to get on track. When the sandwich format is used, receivers consistently walk away believing they performed well. The critical message — the entire point of the conversation — is lost.

Fishbach's explanation is direct: negative feedback gets buried and becomes non-specific when wrapped in positive framing. From the receiver's perspective, the positives register most strongly, and the critique fades into background noise.

This is not a minor misfire. If the goal of corrective feedback is behavior change, a method that leaves recipients feeling praised for substandard work is counterproductive by definition.

The Empirical Record: Perceptions Improve, Performance Does Not

Two studies of written peer feedback with third-year medical students indicate that students think feedback sandwiches positively impact subsequent performance when there is no evidence that they do. The gap between perceived and actual impact is significant: the sandwich makes feedback feel better without making it work better.

The limited body of empirical research on the feedback sandwich method provides evidence that while this method is associated with perceived usefulness and effectiveness, its use may not influence subsequent performance. In other words, everyone leaves the conversation feeling good about it — and nothing changes.

That dissociation between perception and outcome is dangerous for teams. Managers who rely on feel-good signals to gauge whether feedback landed are likely to keep using a method that isn't working.

The Mechanism: Mixed Signals Blur Behavior Change

Research in Organizational Behavior Management shows that the feedback sandwich — positive, corrective, then positive again — can actually weaken performance effects, because mixing praise and correction blurs what behavior should change or continue. When two types of signals are delivered together, neither is processed cleanly.

The research suggests sticking with one type of feedback at a time, because it improves an individual's performance more quickly than mixed-feedback statements do. Clarity, not comfort, is what drives improvement. A manager who separates praise from correction gives both messages a chance to land with full weight.

Think about the cognitive load on the recipient during a feedback conversation. They are managing emotional responses, trying to retain specific information, and deciding what to act on. Injecting praise immediately before and after a critique forces them to switch mental modes repeatedly, which makes it harder to extract and retain the corrective content.

Second-Order Damage: Praise Becomes Untrustworthy

The sandwich doesn't just dilute criticism — it contaminates praise. The formula is so predictable that either the compliments come off as pandering, or employees develop trust issues regarding positive feedback coming from that manager specifically.

Delivering praise and criticism separately allows each to be received genuinely; when praise is consistently coupled with criticism, recipients may begin to distrust all positive feedback. This is a slow erosion. Over months, a manager's compliments become signals that a criticism is coming, rather than genuine recognition of good work. The relational cost compounds quietly.

Author Adam Grant has publicly urged colleagues to stop serving the feedback sandwich, arguing that it makes compliments sound insincere, while HR expert Susan Heathfield writes that the best way to deliver negative feedback is "to provide straightforward, to the point, descriptive communication". Two prominent voices from very different disciplines landing on the same conclusion is notable.

What Employees Actually Prefer: Direct Delivery

People say they like to receive negative feedback directly, and like bosses who deliver negative feedback directly — yet the compliment sandwich does not have very many factors working in its favor. The manager's discomfort with directness is real, but it's not a reliable guide to what serves the employee.

Studies suggest that employees tend to prefer corrective feedback over positive feedback, but managers are often reluctant to give it. This gap between what employees want and what managers provide is one of the central problems in workplace feedback culture. The sandwich thrives in that gap, giving managers a way to feel like they delivered critique without fully committing to it.

Directness does not mean harshness. It means specificity, clarity about what happened, and an honest statement of what needs to change. Those are the qualities employees report finding useful.

The SBI Model: A Leading Structural Alternative

The Situation-Behavior-Impact (SBI) framework, described by Mindtools, gives managers a three-part structure that keeps feedback grounded in observable facts rather than generalizations.

• Situation: Describe the specific context. "In yesterday's client call…"
• Behavior: Name the observable behavior, not a personality trait. "…you interrupted the client three times before they finished their point."
• Impact: State the effect. "That left the client visibly frustrated and we lost momentum on the proposal."

SBI works because each element can be verified. The situation is a shared reference point. The behavior is observable rather than inferred. The impact is traceable rather than speculative. The recipient can argue with an opinion; they find it harder to dispute a sequence of specific facts.

SBI also scales. Use it for a one-on-one correction, a written review comment, or a real-time observation after a meeting. The format remains consistent, which makes it easier to deliver under pressure.

Adapting SBI to Expertise Level

One important refinement: the same feedback structure should not be applied identically to every employee. Research cited in the original I Done This writing on this topic found that the type of feedback people prefer depends on whether they are an expert or a novice. Experts are more likely to seek and respond to corrective feedback. Beginners benefit from encouragement alongside correction, because confidence is a prerequisite for applying new skills.

This means an SBI conversation with a senior analyst should lead with the behavior and impact clearly; a conversation with someone in their first quarter on the job might front-load the situational acknowledgment and be more measured in how impact is framed. The structure stays the same — the calibration changes.

Frequency as a Force Multiplier

No method, however well-designed, compensates for infrequency. According to Gallup research, only 28 percent of people receive feedback a few times a year, while 19 percent say they receive it once a year or less. Feedback that arrives rarely arrives with too much weight — managers over-prepare, employees over-interpret, and the conversation becomes an event rather than a tool.

Employees who receive daily feedback from their manager are 3.6 times more likely to feel motivated to excel than those who receive annual feedback. The implication is not that managers should conduct formal feedback sessions every day, but that brief, specific, timely observations — delivered consistently — have a cumulative motivational effect that formal annual reviews cannot replicate.

High-frequency feedback also reduces the stakes of any single conversation. When correction arrives within hours of a behavior rather than weeks later, it is more specific, more memorable, and easier to act on. The behavior is still fresh. The context is shared. The conversation can be short.

Making Real-Time Feedback Stick

A few practical principles for embedding frequent feedback into everyday management:

Separate channels, separate signals. Deliver praise when there is something genuinely worth praising, not as a preamble to criticism. Reserve correction for moments when a behavior needs to change. Over time, employees learn that both signals mean exactly what they appear to mean.

Specificity over sentiment. "Great job on the deck" is pleasant but vague. "The competitive positioning slide on page four gave the leadership team a clear anchor for the discussion" is specific enough to repeat. The more specific the feedback, the more repeatable the behavior.

Ask before concluding. After delivering SBI feedback, ask the recipient what they're taking away or what they might do differently. This surfaces misunderstanding before it calcifies and gives the employee ownership of the next step.

Document the pattern, not just the incident. A single corrective conversation is a data point. A pattern is evidence. Brief notes after feedback conversations help managers track whether the behavior is improving, staying flat, or declining — and give employees a fair record if a formal review is needed.

Building a Culture Where Feedback Travels Freely

Structured models and calibrated delivery matter, but they operate inside a broader environment. Psychological safety — the shared belief that speaking up, making mistakes, and raising concerns will not result in punishment — is a prerequisite for any feedback model to function well. A team where employees fear judgment will filter feedback in both directions: managers soften critiques, employees hide errors.

Managers build that environment incrementally. Responding to feedback from their own team without defensiveness signals that critique is safe. Acknowledging their own mistakes openly models the behavior they want to see. Asking for input on decisions — and visibly acting on it — demonstrates that feedback flows upward as well as down.

The feedback sandwich, whatever its original intent, signals the opposite: that honest critique requires so much softening that it must be hidden between praise. Replacing it with direct, structured, frequent feedback is not just a technique swap. It is a statement about what the team values and what conversations are possible.

---

Where Oz fits

Oz by Anyreach is an AI teammate you can ask in Slack, Microsoft Teams, email, on the web, or on a call. Oz can use connected tools to help complete work, and every risky action asks for approval first. Learn how I Done This is transitioning to Oz.