Key concepts

This article is a glossary you can return to whenever something in the product is unfamiliar. Every concept below is defined the way it is used inside Performance Blocks, with notes on lifecycle, scope, and how each piece relates to the others.

If you are brand new, skim the section headings first to build a mental map, then come back for detail when you need it.

Observations

An observation — also called a performance block — is the smallest unit of feedback in Performance Blocks. One observation captures one thing, about one person, at one point in time.

Fields

Every observation has the following fields:

Field	Description
Type	Either Strength or Opportunity. There is no neutral or "FYI" type — every observation is one or the other.
Observation text	What you saw, in your own words.
Impact	Why it mattered — to the team, the customer, the project.
Recommended action	What should happen next. For Strengths this is often "keep doing this" or "share with the team." For Opportunities it is the concrete change you would like to see.
Attributes	One or more competencies the observation relates to. See Attributes below.
Observation date	The date the underlying behavior happened, not the date you wrote it down.
Visibility	Who can see it: the author only, or shared with the employee.

Lifecycle

1. Draft — captured but not yet shared. Drafts are private to the author.
2. Shared with employee — the employee can see it in their portal.
3. Included in a summary — referenced inside one or more summaries. Inclusion does not change the observation; it links to it.
4. Archived — kept for the record but excluded from new summaries by default.

Observations are immutable in spirit but editable in practice — you can fix typos and adjust wording, and the audit log captures the change.

Why structured fields matter

The four-field shape (text, impact, action, attributes) is what makes observations useful at synthesis time. A loose note like "great job on the launch" cannot be aggregated. An observation that says "drove the launch handoff (Strength) — unblocked three engineering teams in the final week (Impact) — promote this pattern in the engineering all-hands (Action) — Leadership, Communication (Attributes)" can be quoted directly in a summary, counted toward an attribute trend, and surfaced in calibrations.

Summaries

A summary is a synthesized performance write-up about one employee for a chosen time window. It pulls in observations, objectives, and 360 feedback for that person and produces a single document you can share, present, or hand to an org admin for approval.

Three kinds of summary

• Individual summary — one employee. Owned by their manager. Available on both plans.
• Team summary — aggregate across one manager's direct reports. Useful for skip-level conversations and for a manager's own self-reflection.
• Skip-level summary — one level deeper than a team summary, covering a manager's reports' reports. Available on the Agentic plan only.

Lifecycle

1. Draft — being written or generated. Only the author sees it.
2. In review — submitted to the org admin review queue.
3. Approved — finalized and visible to the employee (if shared).
4. Archived — kept for record after the review cycle closes.

On the Agentic plan, Henry can generate a draft summary in seconds based on the underlying observations and feedback. You still review and edit it — Henry is a starting point, not the finished work.

Review items

A summary that has been submitted for approval becomes a review item in the org admin's review queue. Admins can approve, send back for edits, or comment. See the admin guide for the full review workflow.

Conversations and 1:1s

A conversation is an asynchronous messaging thread between a manager and one of their direct reports. Conversations are also referred to as 1:1s in the UI — the two terms are interchangeable.

What conversations are good for

• Following up on a specific observation.
• Setting an agenda for a synchronous meeting.
• Capturing decisions and action items so they do not get lost.
• Holding space for back-and-forth on objectives or growth areas.

What they are not

Conversations are not a chat product. They are not for real-time messaging, group threads, or social discussion. Each conversation has exactly two participants: the manager and the employee.

Structure

A conversation contains a series of conversation blocks — short, dated entries. Each block can stand on its own or be a reply. Blocks can reference observations, objectives, or other conversation blocks. This makes a conversation easy to skim later when you are writing a summary.

Objectives

An objective is a goal. Performance Blocks supports three scopes:

• Org objectives — set by org admins. Visible to everyone.
• Manager objectives — set by a manager for their team. Visible to that team.
• Employee objectives — set by or with an individual employee. Visible to the employee and their manager.

Objectives have a title, a description, an optional target date, and a status (on track, at risk, completed, abandoned). Updates to status are timestamped and feed into summaries automatically — if an employee marked an objective "at risk" three months ago and "completed" last week, that arc shows up in their next summary.

Objectives are intentionally lightweight. There is no rigid OKR scoring or weighting model — the goal is to capture intent and outcome without the bureaucracy.

360 feedback

A 360 feedback cycle solicits structured peer input about an employee from a chosen group of colleagues — typically people who work closely with that employee but are not their manager.

How a cycle works

1. A manager (or org admin) opens a cycle for one employee.
2. They select reviewers — usually 3 to 8 peers, cross-functional collaborators, or direct reports.
3. The cycle has a deadline (default two weeks).
4. Reviewers receive a request and submit responses through a structured form. They can also use Henry to draft their response on the Agentic plan.
5. Responses become available to the manager and feed into the next summary.

Reviewer identity is visible to the manager by default; you can configure cycles to be anonymous in Settings → Feedback.

Cadence

There is no enforced cadence. Most teams run cycles aligned with their summary or review cadence — quarterly, semi-annually, or annually. Some teams run lightweight cycles ad hoc when an employee is up for promotion or moving teams.

Attributes

An attribute is a competency or skill tag attached to observations. Examples: Execution, Communication, Technical depth, Leadership, Customer focus, Mentorship. Org admins define the attribute set for the organization; managers and employees use them when capturing observations.

Attributes are how Performance Blocks aggregates qualitative feedback over time. If an employee has 18 observations tagged Communication — 14 Strengths and 4 Opportunities — that distribution becomes a meaningful signal in their next summary.

A handful of best practices:

• Keep the attribute list short. Twelve to twenty attributes is plenty for most organizations. More than thirty becomes hard to use.
• Map attributes to your performance rubric or career framework if you have one.
• Avoid duplicates and near-synonyms (Communication and Written communication will dilute each other).

Review cycles

A review cycle is a defined period during which managers write summaries and org admins approve them. The cycle has a name (e.g. "H1 2026"), a window of dates, and a set of employees in scope.

Cycles are optional. You can use Performance Blocks without ever opening a formal cycle — managers can write summaries on their own cadence and admins can approve them ad hoc. Cycles are useful when you want all summaries written and approved in the same window for calibration purposes.

Henry agent

Henry is the AI assistant built into Performance Blocks. The product reaches you in five places:

• The right-side panel in the web app.
• The Slack integration — DM Henry or invoke him from a channel.
• The Microsoft Teams integration.
• The Chrome extension — capture observations from any tab.
• Email — forward a thread to Henry to convert it into an observation.

Henry's main jobs:

• Drafting: turn a few sentences into a structured observation, conversation reply, or summary.
• Searching: answer questions about people on your team using their observations and feedback.
• Guided flows: walk you step by step through structured tasks like writing a summary or starting a 360 cycle.
• Reasoning over your knowledge base: when you upload internal docs (career frameworks, role expectations, company values), Henry uses them as grounding context.

Plan availability: Henry is available on the Agentic plan only. Team plan customers can do everything Henry helps with — they just do it manually.

For the full guide, see the Henry agent docs.

Knowledge base

The knowledge base is the set of documents you upload to give Henry organizational context. Supported formats include PDF, DOCX, and images. Documents are chunked, embedded, and stored so Henry can retrieve relevant passages when answering a question or drafting content.

Typical contents:

• Your performance rubric or competency framework.
• Career ladders or leveling guides.
• Company values and operating principles.
• Role expectations.
• Onboarding and handbook excerpts that are relevant to people decisions.

The knowledge base is Agentic only and is configured in Settings → Knowledge base.

Notifications

Performance Blocks sends notifications when something needs your attention: a new observation has been shared with you, a 360 request has arrived, a conversation has a new message, a summary is awaiting your review. You control delivery in Settings → Notifications:

• In-app — always on.
• Email — configurable per event type.
• Slack or Teams — configurable per event type if the integration is connected.

Notifications batch where possible to avoid noise.

Mental model: how it all flows

Once the vocabulary is in place, the product fits together like this:

1. Attributes define what you measure people on.
2. Objectives define what each person is trying to do.
3. Observations capture what actually happens — tagged with attributes, dated, structured.
4. 360 feedback cycles add peer perspective at moments that matter.
5. Conversations keep ongoing dialog between managers and employees.
6. Summaries synthesize observations, objectives, and feedback into a write-up.
7. Review cycles and the review queue govern approval and finalization.
8. Henry, on the Agentic plan, accelerates every step above.
9. Knowledge base documents give Henry the grounding to do that well.
10. Notifications keep the loop moving so nothing gets lost.

If you understand that flow, you understand Performance Blocks. The rest of the documentation is detail on each step.

Where to go next

• Roles and access — what each role can do.
• Quick tour — a 15-minute walkthrough for new managers.
• Manager guide — capturing observations, running 1:1s, writing summaries.
• Employee guide — viewing your own data and responding to feedback.
• Admin guide — configuring people, attributes, cycles, and the review queue.