What colleagues experience, and what the company sees

1. The terms come first

   Before the first question, the colleague sees who the report is for, what the company receives, how their words are handled, and what happens if they close the page. The same terms the company agreed to.

2. A conversation, not a form

   The interviewer asks about concrete experiences: what the manager does well, where working together gets hard, and which change would help most. It follows up where a form would stop, with neutral questions about a topic, such as how approvals usually work. It never says what anyone else said, and it does not score or rank. Voice is the default. Typing asks the same questions, needs no microphone, and suits an open-plan office.

3. A draft they control

   The conversation becomes a short list of editable points. The colleague can change, remove or add any of them. Each point carries identity protection by default, which generalizes identifying cues and leaves a point out if it cannot be made safe. A colleague can switch protection off for a specific point to include it even if they could be recognized.

4. Submit, or walk away

   Nothing from the conversation or the draft is saved in Chorda's systems before the colleague submits. Close the page and it is gone. After submit, only the approved points are stored, and only for as long as the cycle needs them.

5. Nothing more is asked

   No follow-up questions after Submit. The company sees broad progress for the cycle, never a list of who has finished.

Same behaviour, different words

One colleague mentions a vendor switch that waited two weeks for a yes. Another says people have stopped asking for approval on small things. A third describes the manager as careful about decisions. Chorda recognizes one pattern in three vocabularies and writes it up once, as an observation about sign-off, instead of three separate remarks.

Weighed, not counted

Independent observations from different people carry more weight than one person's repeated point. Chorda notes whether a point came up unprompted or after a neutral question, and a specific event counts differently from a general impression. None of this weighting appears as a number. Each finding carries cues in words. A support cue, independently recurring, means the point was raised unprompted by colleagues who work with the manager in different ways. A scope cue, context-dependent, means the experience differs by situation; it says nothing about strength. The report never says how many people said something.

Differences explained

When experiences disagree, the report says so and names the context where it safely can. The same carefulness that is valued on a decision that is hard to undo can be a bottleneck on a small, reversible one. Feedback that arrives in the moment on launch work can arrive only at review time on routine work. Both experiences are reported with their context, and neither is presented as the consensus. Where a context could point to a person, it is generalized. A single useful observation can appear on its own, as long as it can be made safe.

Grounded and checked before release

Findings are written as experiences colleagues reported, in general terms. They are not verified facts, diagnoses, or claims about intent. Before a report is released, automatic checks confirm every finding is grounded in approved points, respects each point's sharing choice, and that the report as a whole does not point to a person or a relationship. Where it does, the finding is generalized or left out. If the checks fail, the report is not released.

Each cycle produces an individual report for the manager and their coach. Across cycles, Chorda shows People teams and leadership the themes that recur and how many manager cycles share them, so that a pattern can be read for what it is. If sign-off waits appear in twelve of forty cycles, that is a reason to look at the shared approval process before treating it as twelve individual development points. It is a lead, not proof of the cause. A reorganization or a new tool lands differently in different teams, and the company view shows how the same moment was experienced, without pointing at a team or a manager.

Themes can be followed from one cycle to the next. If a theme is less widespread after a leadership programme, the company view shows that change. It is a reason to look closer, not proof that the programme caused it. Counts are always of manager cycles, never of the colleagues who spoke. The figures here are illustrative.

What the company view never does

• Rank managers or cluster them by name
• Show which manager a theme comes from
• Expose individual submissions, points, or who took part
• Replace the individual report, which stays with the manager and their coach

A full 360 runs once or twice a year. From it, the manager and their coach choose two or three experiments: one concrete change in behaviour each, with signs of progress agreed in advance.

Every four to six weeks, two or three colleagues answer a few minutes of questions about that experiment only. Each check-in knows which experiment the manager chose and what the previous check-in found, so the questions build on each other rather than starting from zero. Chorda carries forward the experiment and the synthesized findings, not the conversations. The manager learns whether the change is noticeable, then keeps it, adapts it, or moves on to the next one. The check-in is not a survey, not a monitoring tool, and it produces no score.

At the next full 360, last cycle's themes and the experiments tried sit alongside the new findings, so progress is visible without anyone keeping a spreadsheet.

Sometimes a colleague describes behaviour that is more serious than a development point. That feedback follows the same consent, grounding and privacy rules as everything else: the colleague decides whether it is submitted and how it is shared, and the report treats it as a reported experience.

Chorda is not a formal reporting or emergency channel. Colleagues see the company's separate reporting route before they begin, and nothing is forwarded to anyone automatically.

• Not an anonymity guarantee. Privacy checks reduce recognition risk. On a small team, they cannot remove it, and colleagues are told so before they decide what to share.
• Not verified facts. Findings describe experiences as reported. They are not investigations, diagnoses or judgments of intent.
• Not a complete picture. A point is evidence the report may use, not a guarantee that it appears. Points are left out when they cannot be made safe.
• Not a substitute for the conversation. The report gives the manager and coach somewhere to start. The development happens between people.