Incompass was built around how feedback actually works — around the science of what makes data accurate, fair, and worth acting on. Here's what that means in practice.
.png)
93% of companies run performance reviews. Only 6% think they're worth the time. The process hasn't kept up with how work actually happens.
Every design choice in Incompass exists for a documented reason rooted in behavioral science backed by The Wharton School statisticians. Each one addresses a specific failure in how performance is typically measured. Understanding them is understanding why the output is different.
In a traditional review, feedback flows one way — from manager to employee. What the manager thinks is what goes on record. It's fast to run, but it's a single perspective, shaped by proximity, visibility, and bias.
Incompass gathers feedback from the people who actually work alongside someone — peers, direct reports, cross-functional collaborators. Each rating is then weighted by how closely the reviewer works with that person, so the signal reflects real working relationships, not just reporting lines.
One perspective. Shaped by personal bias, visibility gaps, and the quality of the manager relationship. What the manager thinks becomes the official record.
Peers, reports, and collaborators each contribute. Ratings are weighted by familiarity. The result reflects how someone actually shows up at work.
Most tools ask a rater to evaluate each person from start to finish — then move to the next person, and start over. By the time they reach the fifth colleague, they're working from a different mental state than when they started. Fatigue, shifting expectations, and context drift quietly distort the data.
Incompass structures the process differently. Raters move through one dimension across the entire team before advancing to the next. That way, every score is anchored to the same frame of reference — not to wherever the rater happened to be mentally when they got to that person.
Context resets with every new person. Fatigue distorts later ratings. No consistent reference point across the team.
Same reference point across the whole team. People are compared to each other, not to a shifting internal standard.
When people know their feedback can be traced back to them, they don't say what they actually think — they say what's safe. This isn't a personality flaw; it's a rational response to social risk. Incompass removes that risk entirely.
We keep three things hidden: who reviewed you, what score they gave, and what they said in writing. Research shows a 47% increase in participation and 58% higher likelihood of applying feedback when anonymity is guaranteed. Qualitative comments become dramatically more candid — which is where the most actionable signal lives.
Raters self-censor. Feedback reflects what's safe to say, not what's true. Written comments are vague and non-committal.
+47% participation. +58% more likely to act on feedback. Written responses become dramatically more candid and useful.
Some managers rate everyone a 9 out of 10. Others cluster around 5. Raw averages treat these identically — which means a talented person on a tough grader's team looks worse on paper than a mediocre person on a generous grader's team.
Incompass corrects for this automatically across three dimensions: scoring style normalization, familiarity weighting, and sparse-data stabilization. The result is scores that reflect actual performance — not who reviewed you or how generous they tend to be.
Scoring style differences create false comparisons. A tough grader's team looks worse on paper regardless of actual performance.
Scoring style normalization. Familiarity weighting. Sparse-data stabilization. Scores reflect performance — not reviewer tendencies.
Six dimensions where the methodology diverges — and why each one matters for the quality of your data.
Talk to us about running a cycle at your organization.
