> 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/become-a-tability-power-user/how-tos/how-to-write-okrs-when-you-already-have-long-term-goals.md).

# How to write OKRs when you already have long-term goals

A practical guide to layering OKRs on top of long-term plans, KPIs, and milestones — without duplicating a single goal.

Your long-term targets say **where you're going**. OKRs decide **what you'll change this quarter to get there**. That's a different job — not a smaller copy of the same one.

If you already run against a three-year plan with clear targets, OKRs can feel like extra paperwork — like someone asked you to rewrite your goals in a new notation and call it strategy. That reaction is fair. Done badly, OKRs are exactly that.

Done well, they answer a question your long-term plan doesn't: *of everything you could do this quarter, what will you actually push on?* This guide shows how to put OKRs on top of the targets you already have — without duplicating a single one.

### 1. The one idea that fixes the confusion

The "this feels duplicative" feeling comes from treating an OKR as a smaller slice of your plan. It isn't a slice. It's a *choice*.

Your plan and your KPIs describe the destination and the dials you always watch. An OKR is the one or two **bets** you're making right now to move a dial that won't move on its own. Same business, two different jobs.

{% hint style="success" %}
**The distinction to hold onto:** You **run** the business with KPIs and your long-term plan. You **change** the business with OKRs. \
\
Most of what matters stays in the first bucket. OKRs are only for the handful of things you're actively trying to move this cycle. 
{% endhint %}

### 2. Two things, side by side

Seen together, it's clear these aren't the same document written twice. They do opposite things.

|                      | The plan you already have                            | OKRs                                              |
| -------------------- | ---------------------------------------------------- | ------------------------------------------------- |
| **Its job**          | Where you're going, and the health metrics you watch | What you'll change this cycle to get there faster |
| **Time horizon**     | One to three-plus years, rolling                     | A quarter, sometimes a year                       |
| **What it measures** | Destination targets and steady-state KPIs            | The specific movement you're trying to create now |
| **How many**         | Everything that matters to the business              | A short list — only what needs an active push     |
| **Changes**          | Rarely                                               | Every cycle                                       |
| **When you hit it**  | You're where you planned to be                       | You moved the needle you set out to move          |

{% hint style="info" %}
**The unlock:** not every target becomes an OKR. Most of your KPIs should stay KPIs. If everything on your plan turned into an OKR, you'd have rebuilt your plan in new notation — which is exactly the busywork you were right to be suspicious of. 
{% endhint %}

### 3. What actually deserves an OKR

A target earns an OKR only when all three of these are true *this cycle*:

* It won't improve on its own or on the current trajectory.
* You're willing to put real focus and resources behind moving it now.
* You can meaningfully influence it in the next quarter to a year.

Everything else stays where it is — a KPI you monitor, or a milestone you track on the plan. There's no prize for coverage.

{% hint style="info" %}
**A quick test:** if the honest answer to *"what would we do differently because this is an OKR?"* is "nothing," then it isn't one. It's a metric you're already watching, and it should stay that way. 
{% endhint %}

### 4. From long-term target to OKR, without copying

The trick is to never start from the target itself. Start from the gap.

#### 1. Look at the gap, not the goal

Put the destination target next to where you are today. The interesting part isn't the number — it's the gap, and the reason it's there. That reason is where an OKR lives.

#### 2. Name the bet

What single shift, this cycle, would close the most gap? Say it in plain language, with no metrics attached. That sentence is your **Objective**.

#### 3. Decide how you'll know it's working

Pick two to four **Key Results** — outcomes, not tasks — that would prove the bet is paying off. Your destination number never appears here. It stays on the plan. The OKR is the wedge you drive to reach it.

**Worked example — Head of Sales**

*A three-year revenue target becomes one quarter's bet.*

* **On the plan (stays put):** Grow ARR from $10M to $30M over three years. *This is the destination — never written as an OKR.*
* **This year's anchor:** Reach $16M ARR.
* **Q3 Objective:** Establish mid-market as a real revenue channel.
* **Key Results:**
  1. Close 12 mid-market logos over $50k ACV, up from 2.
  2. Grow qualified mid-market pipeline from $1.5M to $5M.
  3. Cut the mid-market sales cycle from 90 to 60 days.

> **Notice:** $30M ARR is nowhere in the OKR. The gap was "mid-market is untouched," so the bet is mid-market. The OKR moves you toward the destination faster than you'd get there by default — it doesn't restate it.

### 5. When your goals are milestones, not numbers

Plenty of long-term plans are milestone-based: launch in three regions, earn a certification, ship a platform. These convert differently, and the fix is simple.

Leave the milestone on the plan as a milestone. Then let the OKR capture the *outcome* you actually care about — so you don't just "ship on time," you "ship something that works."

**Milestone-based example — "Launch in Germany by Q4"**

*From "done" to "worth doing."*

* **Objective:** Make the Germany launch land, not just go live.
* **Key Results:**
  1. 200 activated accounts in the first 30 days.
  2. Localization defect rate under 2%.
  3. Three reference customers signed post-launch.

> **The move** is turning "did we do it?" (binary) into "did it work?" (measurable). If a milestone is genuinely just done-or-not and carries little risk, leave it as a milestone. Don't force an OKR onto it.

### 6. Write Key Results that measure movement, not motion

The most common mistake, by far, is Key Results that are really task lists: "hire two reps," "run three campaigns." Those are the *plan* for hitting a Key Result — not the Key Result itself. A good KR states the result you'd see if the work succeeded.

* **Task (avoid):** Launch a partner program.
* **Outcome (aim for):** 15% of new pipeline sourced through partners.

{% hint style="info" %}
**One test settles it:** could you complete every task on the list and still not move the business? If yes, you've written tasks. Rewrite them as the change you're trying to see. 
{% endhint %}

### 7. Let a tool do the heavy lifting

The hard part usually isn't the concept — it's the blank page, and the nagging "am I even writing these right?" A good AI assistant removes both. You bring the judgment about which bet matters; it handles the drafting and the format.

Hand it three things: your long-term target, where you are now, and the one bet you want to make. Then ask for a draft.

**A prompt you can paste into any AI assistant:**

```
I'm the [VP of Sales]. Our three-year target is to [grow ARR from $10M
to $30M]. We're at [$12M] today and behind because [we haven't touched
mid-market]. This quarter I want to make one bet: [break into mid-market].

Draft one Objective and 3 Key Results. Key Results must be measurable
outcomes, not tasks. Suggest a baseline and target for each, and flag
anything that reads like a to-do list rather than a result.
```

#### Doing it where your plan already lives

An assistant in a blank chat is great for a first draft. **Tability's AI mode** goes a step further: it works inside the tool where your plan and targets already sit. It can read your existing goals, draft OKRs from them, suggest measurable Key Results, and — the part that matters most here — keep the long-term plan and the quarterly OKRs linked, so you're never maintaining two copies of anything.

That link is the whole point. When the tool holds the connection between destination and bet, updating one doesn't mean rewriting the other. The duplication you were worried about stops being possible.

{% hint style="warning" %}
**One caveat:** whatever you use, you still decide which bet is worth making. The tool drafts. You choose. 
{% endhint %}

### 8. Traps that make OKRs feel like busywork

Nearly every "OKRs are a waste of time" story traces back to one of these. Avoiding them is most of the battle.

**Turning the whole plan into OKRs**

If every target becomes an OKR, you've rebuilt your plan in a new format. Pick the few things that need a push now; leave the rest as KPIs and milestones.

**Too many of them**

Three Objectives with three Key Results each is plenty for most teams. Past that, nothing gets real focus — which defeats the purpose of choosing at all.

**Key Results that are secretly tasks**

If you can finish them all and nothing about the business changed, they're a to-do list. Rewrite them as outcomes.

**Set and forget**

OKRs earn their keep in the weekly or fortnightly check-in, not the kickoff doc. If you'll write them and never look again, don't bother writing them.

**Grading theater**

The score exists to start a conversation — "why did this stall?" — not to judge people. The moment it feeds performance reviews, everyone sandbags their targets and the honesty is gone.

### 9. Quick start

If you do nothing else, do this.

1. Keep your plan and your KPIs exactly as they are. Nothing here replaces them.
2. Find the one or two targets that are behind, or that need a real push this cycle.
3. For each, name the single bet that closes the most gap — that's your Objective, in plain words.
4. Write two to four Key Results that would prove the bet is working. Outcomes, not tasks.
5. Draft with an AI assistant or Tability's AI mode, then edit down to what's actually real.
6. Check in every week or two. Adjust the work when it's not landing — not the target.

{% hint style="success" %}
**You already did the hard part.**\
\
You have the targets. You know the business. You know where you're behind and why. OKRs don't ask you to redo any of that — they ask a smaller, more useful question: *what are we changing right now to get there?*\
\
Pick a couple of bets, write them plainly, and check in. That's the entire practice. Done right, it should make your quarter clearer — not busier. 
{% endhint %}