"360 feedback" sounds sophisticated. But the way most organizations implement it, here's what happens: someone finds a generic 360 survey template online, sends it to the participant's boss, peers, and direct reports, averages all the scores, and hands the participant a report with a number like "3.7 out of 5."
That approach is fundamentally broken. Averaging scores from different sources flattens the most valuable information in a 360—the differences between perspectives. Your boss thinks you're "results-driven." Your team thinks you're "controlling." Your peers think you're "not collaborative." Put those three together and you have a diagnosis. Average them into 3.7 and you have nothing.
The Four Feedback Roles
360 assessment is built on multi-source feedback. Each source sees a different side of you:
Sometimes a fifth role is added—external clients or cross-functional partners—useful for roles that require boundary-spanning collaboration.
Dimension Design: Don't Use Off-the-Shelf Templates
Plenty of generic 360 templates exist—"Ten Leadership Competencies," "Eight Management Dimensions," and so on. Using them as-is usually underwhelms because they rate generic capabilities rather than the capabilities your organization and your roles actually need.
Dimension design should start from two questions:
- What are this role's core responsibilities? Responsibilities determine which capabilities to assess
- What does our culture expect? The same role in different cultures has different priorities
For example, "middle manager" in an execution-focused company might assess "goal decomposition," "process control," "delivery." In an innovation-focused company, the same role might assess "stimulating creativity," "guidance through failure," "cross-boundary collaboration."
Aim for 5-7 dimensions with 3-4 items each—15-25 items total. Beyond 30 items, rater fatigue sets in and data quality falls off a cliff.
Rating Scales: Behavior Anchoring Is Essential
Just like with needs assessment, 360 rating scales must be behavior-anchored. Without behavior anchors, you're collecting impression scores.
Take the dimension "empowering direct reports." Without behavior anchors:
"Rate the ability to empower direct reports: 1 2 3 4 5"—the rater scores based on pure gut feel.
With behavior anchors:
- 1: Rarely delegates, prefers to do it themselves
- 2: Occasionally delegates but micromanages details, limited autonomy for team
- 3: Delegates but doesn't always match the right person or resources to the task
- 4: Systematically assigns tasks and resources, follows up without micromanaging
- 5: Actively creates growth opportunities, tailors development paths to individual strengths
When raters see behavior anchors, scoring has a reference standard and inter-rater consistency improves dramatically.
The Report: Focus on Differences, Not Scores
The most valuable part of a 360 report isn't "your overall score is 3.8." It's the gap analysis between different rater sources.
Common discrepancy patterns and their diagnostic meaning:
Self high, others low: Self-awareness gap. Could be overconfidence, or simply not knowing how their behavior lands on others. These people need specific behavioral feedback to "hold up the mirror."
Supervisor high, direct reports low: Strong at "managing up" but weak at "managing down." Performs well for the boss but doesn't invest enough in the team or is overly harsh with subordinates. This pattern is extremely common.
Peers low, others high: Cross-functional collaboration is the weak spot. May treat their team well and deliver for the boss, but doesn't cooperate well across departments. Often a silo mentality—only cares about their own team's interests.
All sources low: Not a specific-area problem—overall capability needs development. Requires a systematic development plan.
Report design tip: Use a radar chart to display multi-source comparisons across all dimensions. The advantage of a radar chart is you can see the "shape" instantly—an irregular shape signals large discrepancies between sources, meaning the gap between self-perception and external perception is wide.
The Four 360 Pitfalls
Pitfall 1: Using 360 for performance review. 360 was designed for development, not evaluation. The moment it's tied to promotions or raises, raters strategize: direct reports fear retaliation and score high, peers fear conflict and score high, supervisors protect their people and score high. Everyone ends up above 4.0 and the data is completely distorted.
Pitfall 2: Poor rater selection. Letting participants choose their own raters guarantees they'll pick people who like them. Raters should be designated by HR or the supervisor, and must have sufficient work history with the participant—at least 3 months of collaboration—to have a basis for assessment.
Pitfall 3: Insufficient anonymity. If there are only 1-2 direct reports, even nominally anonymous feedback can be traced. You need at least 3 direct report raters to ensure genuine anonymity. If the team is too small, consider adding cross-functional partners to supplement the rater pool.
Pitfall 4: No follow-up after the report. The biggest waste in 360 is "assess and abandon." After receiving the report, the participant should have a 1:1 debrief with their supervisor or a coach, identify 1-2 dimensions to improve, and create a 3-month behavior change plan. A 360 without follow-up is worse than no 360 at all—it raises expectations and delivers nothing.
360 assessment has two applications within a training system. First, pre-training needs analysis—identify the dimensions with the largest gaps as training priorities. Second, post-training effect tracking—run the same 360 tool before and after training, compare dimension score changes, and you have the most powerful L3 evidence of behavior change. One tool, two uses: diagnosis and evaluation.
🛠️ Build 360 Assessment in FormLM
From questionnaire design to report generation, FormLM supports the full 360 workflow:
- Use scale fields to design behavior-anchored rating scales where each score corresponds to a specific behavioral description
- Use multi-role collection to let supervisors, peers, and direct reports fill the same questionnaire separately
- Use radar charts to display multi-source comparisons across dimensions—gaps are instantly visible
- Use AI reports to auto-generate personalized 360 feedback reports with discrepancy diagnosis and development recommendations
- Use Insight pre-post comparison to run 360 before and after training for quantified behavior change
✅ Key Takeaways
- 360's core value is cross-source discrepancy analysis, not averaged scores
- Four roles, four perspectives: self (awareness), supervisor (results), peers (collaboration), direct reports (management)
- Design dimensions from job responsibilities and company culture—not off-the-shelf templates. Aim for 5-7 dimensions
- Rating scales must be behavior-anchored, or you're collecting "impression scores"
- 360 is a development tool, not a performance review tool—tying it to compensation ruins the data
- Follow-up is non-negotiable: debrief + behavior change plan, or don't bother