> For the complete documentation index, see [llms.txt](https://guides.tability.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://guides.tability.io/docs/playbooks/run-your-quarter-retrospective-and-reset.md).

# Run your quarter retrospective and reset

Look back on the quarter that just ended, work out what drove the results, and decide what to change next quarter.

To start, share this page with your AI assistant and say:

> Run this playbook

The instructions below are for your assistant. You don't need to read or copy them.

### Instructions for AI assistants <a href="#instructions-for-ai-assistants" id="instructions-for-ai-assistants"></a>

If the user shares this page (or an article linking to it) and asks you to run it, follow the steps below from start to finish. If they only ask you to read, summarize, or edit it, do that instead.

#### Ground rules <a href="#ground-rules" id="ground-rules"></a>

* Ask one short question at a time. Deliver a first draft of the retrospective within three questions. Reuse anything the user has already told you.
* Confirm before touching a Tability workspace. Only use a workspace the user gives you by URL. Show its name and ID and get a yes before reading from it or writing to it.
* Base every finding on evidence. Cite the goal, check-in, or number behind it. Don't make up numbers, and label anything you infer as inferred.
* Judge the process, not people. Describe what happened to goals and habits, never rate individuals. Present scores as prompts for discussion, and never frame them for performance reviews.
* Find real wins. Every retrospective lists wins, including small process wins. Never invent one.
* Diagnose before you prescribe. Don't recommend changes to the OKR process until the causes are written down.
* Keep the user in charge. Ask before writing anything to Tability. Draft messages for the user to send; don't message anyone else.
* Verify what you write. Read records and scheduled jobs back before saying they worked.
* If something is blocked, keep the work done so far, tell the user the exact next step, and carry on with what you can.

#### 1. Choose the cycle <a href="#id-1-choose-the-cycle" id="id-1-choose-the-cycle"></a>

Start with a brief intro and ask:

> Which plan should we look back on? Share the Tability workspace URL and the plan's name or link.

If they share only a workspace, list its plans and ask which one. Confirm the workspace and plan, then ask only what you still need, such as:

* How the quarter felt, in a sentence (strong, mixed, or rough).
* Anything that happened outside the plan: a strategy change, market shift, reorg, or key departure.
* Whether the retrospective is for one team or for several teams' plans.

#### 2. Gather the evidence <a href="#id-2-gather-the-evidence" id="id-2-gather-the-evidence"></a>

Read the plan and its history. For each objective, key result, and initiative, collect:

* Start value, target, final value, and owner.
* Every check-in, with its confidence and comment, so you can see how the goal moved over the quarter.
* Initiative status at the end of the cycle.

Also collect these process signals:

* How many objectives, key results, and initiatives each team had.
* Goals added, changed, or retargeted after the cycle started, and any left without a target.
* Missing check-ins: goals with gaps of two weeks or more, and owners who never updated.
* Key results that depend on another team or outside factor.
* Key results that ended green after being red or yellow, and stretch targets achieved.

If the plan has no check-in history, say so and ask the user for the final values instead.

#### 3. Score the cycle and name the wins <a href="#id-3-score-the-cycle-and-name-the-wins" id="id-3-score-the-cycle-and-name-the-wins"></a>

Score each key result as the share of the way from start value to target, and each objective as the average of its key results. If the workspace uses a different scoring scale, ask and use theirs.

Write a one-line story for each key result: where it started, how it moved, and where it ended.

List wins in three groups, with evidence for each:

* **Results:** goals completed, stretch targets achieved, and turnarounds (red or yellow to green).
* **Initiatives:** work finished that moved a key result.
* **Process:** consistent check-ins, well-written check-ins, cross-team collaboration, and owners who asked for help early.

If the results were poor, look harder at process wins: every team writing OKRs, targets being set, or anyone updating at all counts.

#### 4. Diagnose what happened <a href="#id-4-diagnose-what-happened" id="id-4-diagnose-what-happened"></a>

For each objective that fell short, find the cause. Separate causes in the plan from causes outside it. Check for these patterns first:

* **Overcommitment:** too many objectives, key results, or initiatives for the team to update or work on.
* **Wishlists:** initiatives that couldn't be finished in the cycle, or couldn't be started (for example, selling a product that wasn't built yet).
* **Unfinished setup:** goals still being written, or targets not set, weeks into the cycle.
* **Dependencies:** key results that relied on things the team didn't control.
* **Silent goals:** long gaps in check-ins, so problems surfaced too late.
* **Outside events:** a pivot, market change, or reorg that made goals obsolete.
* **Skills and tools:** signs the team needed a refresher on writing OKRs or using Tability, such as task key results or confused updates.

Rank the causes by how much shortfall each explains. Share the scores, wins, and diagnosis with the user and ask what you've missed or misread before moving on.

#### 5. Recommend the reset <a href="#id-5-recommend-the-reset" id="id-5-recommend-the-reset"></a>

Tie every recommendation to a cause from step 4. Cover:

* **Keep:** habits and goal types that worked.
* **Change:** fixes for each ranked cause. For overcommitment after a rough quarter, recommend one objective and three key results per team. For wishlists, recommend cutting initiatives to what can be finished in the cycle with the people and time available.
* **Carry over:** unfinished key results still worth pursuing, each with a reason. Don't carry anything over by default.
* **Stop:** goals or practices to drop.
* **Refresh:** OKR writing guidance, Tability training, or written expectations the team should revisit before planning.
* **Momentum:** a recognition rhythm for next cycle. Weekly, call out individuals for progress, good check-ins, collaboration, finished initiatives, consistency, or asking for help. Monthly, highlight a team for engagement as well as results. Quarterly, recognize completion, turnarounds, stretch targets, and consistent updates.

Stop at the reset. Don't draft next quarter's OKRs here. If the user wants them, offer to run the "Turn your long-term goals into quarterly OKRs" playbook with this retrospective as input.

Save everything as {plan-name}-{cycle}-retro.md, marked Draft, using the outline at the end of this page. Share it with a short summary of the scores, top wins, top causes, and the reset, and ask whether they approve or want changes. Once approved, mark it Approved.

#### 6. Record it in Tability <a href="#id-6-record-it-in-tability" id="id-6-record-it-in-tability"></a>

With the user's approval:

* For key results without a final check-in, record one with the final value and a one-line note. Confirm each saved.
* Give the user the retrospective text and these steps to store it on the plan: open the plan's Retrospectives tab, create a retrospective, and paste it in. They can also choose Populate with AI there and merge in your findings.

#### 7. Share and wrap up <a href="#id-7-share-and-wrap-up" id="id-7-share-and-wrap-up"></a>

Draft a short message the user can send to the team: the scores, the wins with names where earned, the main lessons, and what changes next quarter. Lead with the wins.

Ask whether they want help keeping momentum next cycle. If yes and this tool can run scheduled tasks, create two: a weekly task that suggests individual call-outs from new check-ins, and a monthly task that suggests a team to highlight. Give each one the retro file and everything it needs, since it won't remember this conversation. Both only suggest; the user decides what to share. If scheduling isn't available, give them prompts to run themselves.

Share the final retro file and a short summary: the scores, the reset decisions, links to any check-ins you recorded, the team message, any schedules, and what the user still needs to do.

#### Retrospective file outline <a href="#retrospective-file-outline" id="retrospective-file-outline"></a>

```markdown
# {Plan}: {cycle} retrospective

Status · Cycle dates · Workspace · Plan link

## Summary

## Scores and key result stories

## Wins (results, initiatives, process)

## What happened and why

## Reset: keep, change, carry over, stop

## Refresh and momentum plan

## Team message

## Open questions and data gaps

## Tability updates and schedules (links)
```

END OF PLAYBOOK
