360-degree feedback

The mental model

A 360-degree feedback cycle centers on a subject — the person being assessed — who receives feedback from a group of responders drawn from different relationships around them (for example their manager, their peers, the people who report to them, and optionally the subject themselves). Feedback from each of these perspectives is gathered and then brought together into a single report for the subject, so they can see how they are perceived across different vantage points rather than from just one source.

360 feedback has its own dedicated management area for admins, separate from other program tooling, because it carries extra rules around structure and anonymity that don’t apply to most other content types. That said, a 360 assessment is still delivered to participants as part of a program’s normal flow — it can also be surfaced as an item on a participant’s journey alongside their other activities, so the person taking it experiences it as one more step in their program rather than a separate system.

What an admin configures

At a high level, setting up a 360 for a program involves a small number of conceptual decisions:

Assessment structure. Whether the cycle is a classic full 360 (manager, peers, team members, and self all contribute), an upward-only cycle (team members giving feedback about their manager), or a custom mix of relationships tailored to the program.
Content. Whether responders answer structured competency ratings (drawn from a competency framework), free-text open-ended questions, or a mix of both.
Self-assessment. Whether the subject is required, invited, or not asked to rate themselves alongside their responders — useful for comparing self-perception against how others see them.
Who the responders are and how they’re grouped. Responders are organized by relationship/perspective (manager, peer, team member, other), and an admin can choose how flexible subjects are in managing their own responder list — for example, whether they can freely add and remove people, can only add (not remove) names that HR has already set up for them, or have no ability to change the list at all.
How the responder list is populated. Responder lists can be built up in a few ways: the subject nominates people themselves, the system suggests likely responders based on the org chart (manager, peers, direct reports), or an admin pre-sets/imports names in bulk. Programs can also choose to invite responders from outside the platform (by name and email) rather than restricting feedback to platform users only.
Minimum responder counts. Admins can set how many responders are needed — as an overall minimum and, optionally, more granular minimums or caps for a specific relationship group (e.g. requiring at least a certain number of peers). These thresholds exist specifically to protect anonymity (see below), and a subject cannot see their results until they’re met.
Rating scale. The numeric scale used for ratings (e.g. a 1–5 or 1–10 range) and the labels shown at each point on that scale.
Visibility and reporting. How much detail different roles around the subject (their coach/support giver, their manager, HR business partners) can see of the resulting report — ranging from no visibility at all, to a high-level summary, to the full breakdown — and whether AI-generated narrative insights are layered on top of the raw results.
Lifecycle and reminders. Whether a 360 automatically kicks off when someone joins a program, when it’s considered due, whether it automatically wraps up at the end of a program, and whether reminder emails go out to responders who haven’t yet completed their feedback.

Anonymity, in plain terms

Protecting the people who give feedback is a core design goal of 360, not an afterthought. The system is built so that:

In the report, individual responses are not shown as “so-and-so said this.” Feedback is gathered from a group of responders and presented to the subject in aggregated form, so that to the people reading the report no single response can be traced back to the person who gave it.
A subject cannot see any results — scores, comments, or breakdowns — until a minimum number of responders has provided feedback. This minimum is admin-configurable (with a sensible built-in default), and until it’s met, the subject and other viewers simply see that feedback is still being collected rather than any partial results.
This same principle extends to more granular views. Where a report breaks results down further (for example, into smaller org-level groupings), the same kind of minimum-response safeguard applies — those breakdowns stay hidden until enough responses exist to protect the individuals involved.
Different viewers (the subject’s coach, their manager, HR) can be given different levels of access to a 360 report — from nothing at all, to a summary, to the full detail — but all of those levels sit on top of the same underlying anonymity protections; broader access never bypasses the minimum-response safeguard.

Admins should think of the minimum-responder threshold as the anonymity “gate” for the whole cycle: results simply do not exist, from any viewer’s perspective, until enough people have responded.

Where this fits in a program

A 360 assessment is attached to a program as a specific type of content item. It has its own configuration, its own responder-management screen, and its own report/analytics views — but it is scheduled, assigned, and tracked using the same program and cohort mechanics as everything else, and can appear directly on a participant’s journey like any other program item.