# How to Choose Between Great Ideas

> **Formats:** Markdown (canonical) | DOCX | PDF
> **Updated:** 2026-03-22
> **License:** CC BY 4.0 -- Kate Makrigiannis / k8mak.com

A 4-step quick start guide for prioritizing features, initiatives, or ideas using a 2x2 matrix. Helps your team move from "everything is important" to a clear, defensible action sequence.

## When to use this

You have more good ideas than capacity to build them. Your team needs to decide what to do first, what to defer, and what to kill. This guide walks you through a 2x2 prioritization exercise you can run in 30-45 minutes with sticky notes, a whiteboard, or any digital collaboration tool.

---

## Process

### Step 1: Collect items and pick your axes

Start by listing everything that needs prioritizing. Features, stories, initiatives, experiments -- whatever is competing for attention.

**Write each item on a separate sticky note or card.** One idea per card. Include just enough context to evaluate it (a short title and 1-2 sentence description).

**Then pick your two axes.** These define the 2x2 grid your items will land on.

| Axis Pair | Best for | X-axis (horizontal) | Y-axis (vertical) |
|---|---|---|---|
| **Impact / Feasibility** | Feature prioritization, sprint planning | How easy is this to build? | How much does this move the needle for users? |
| **Importance / Urgency** | Competing demands, executive requests | How critical is this to our goals? | How time-sensitive is it? |
| **Pain / Frequency** | Bug triage, UX improvements | How often do users hit this problem? | How painful is it when they do? |

Pick the pair that matches your decision. If none of these fit, define your own two axes -- just make sure each one is something your team can evaluate with evidence, not just gut feel.

---

### Step 2: Vertical stack-rank

Place all items along the vertical axis first. Ignore the horizontal axis for now.

**How to do it:**
- Start with any two items. Which one scores higher on the vertical axis? Place the higher one above the lower one.
- Add the next item. Where does it fall relative to the ones already placed? Slide it into position.
- Keep going until every item is stacked vertically from highest to lowest.

**Why vertical first?** Evaluating one dimension at a time is easier than evaluating two simultaneously. It reduces arguments and speeds up the exercise.

**Facilitation tip:** If the team disagrees on placement, ask: "What evidence would change your mind?" If the answer is "none," someone is anchoring on opinion. Name it and move on.

---

### Step 3: Horizontal slide

Now evaluate each item on the horizontal axis. Slide items left or right based on their score.

**How to do it:**
- Work through the stack from top to bottom
- For each item, ask: "Where does this fall on the horizontal axis?"
- Slide it left (low) or right (high) without changing its vertical position
- Items will naturally cluster into four quadrants

**What the quadrants mean (Impact/Feasibility example):**

```
                    High Impact
                         |
    ┌────────────────────┼────────────────────┐
    │                    │                    │
    │    BIG BETS        │    QUICK WINS      │
    │    High impact,    │    High impact,     │
    │    hard to build   │    easy to build    │
    │    → Plan these    │    → Do these FIRST │
    │                    │                    │
    ├────────────────────┼────────────────────┤
    │                    │                    │
    │    MONEY PIT       │    FILL-INS        │
    │    Low impact,     │    Low impact,      │
    │    hard to build   │    easy to build    │
    │    → KILL these    │    → Do if spare    │
    │                    │                    │
    └────────────────────┼────────────────────┘
                    Low Impact
         Hard to build       Easy to build
```

---

### Step 4: Read the quadrants and build your action list

The matrix tells you what to do. Read it quadrant by quadrant.

**Quick Wins (high impact, high feasibility):** Do these first. They deliver the most value with the least effort. If you only have capacity for one thing, pick from here.

**Big Bets (high impact, low feasibility):** Plan these carefully. They're worth doing but need more time, research, or resources. Break them into smaller pieces if possible.

**Fill-Ins (low impact, high feasibility):** Do these when you have spare capacity. They're easy but don't move the needle much. Don't let them crowd out Quick Wins or Big Bets.

**Money Pit (low impact, low feasibility):** Kill these. They cost a lot and deliver little. This is the hardest decision for most teams -- but saying no here is what makes everything else possible.

**The KILL decision matters most.** Teams that can't say no to Money Pit items end up with bloated roadmaps and scattered focus. If an item lands in the Money Pit, remove it from the backlog entirely. Don't defer it. Don't revisit it next quarter. Kill it.

**Build your action sequence:**

| Priority | Item | Quadrant | Next Step |
|---|---|---|---|
| 1 | (Quick Win item) | Quick Win | Start this sprint |
| 2 | (Quick Win item) | Quick Win | Start this sprint |
| 3 | (Big Bet item) | Big Bet | Spike/discovery next sprint |
| -- | (Money Pit item) | Money Pit | KILLED -- removed from backlog |

---

## Worked example

**Context:** A health-tech product team needs to prioritize 6 features for the next quarter.
**Axes:** Impact (on patient outcomes) vs. Feasibility (engineering effort + data availability)

**Items:**

| Feature | Impact | Feasibility | Quadrant |
|---|---|---|---|
| Medication reminder notifications | High -- reduces missed doses | High -- push notification infra exists | Quick Win |
| AI-powered symptom checker | High -- could catch issues early | Low -- needs ML model, regulatory review | Big Bet |
| Dark mode for mobile app | Low -- cosmetic, no health outcome | High -- standard UI work | Fill-In |
| Real-time vitals dashboard for clinicians | High -- faster clinical decisions | Medium -- requires device integrations | Big Bet |
| Custom avatar for patient profiles | Low -- engagement only | Low -- needs design system overhaul | Money Pit |
| Care plan PDF export | Medium -- requested by 3 clinic partners | High -- existing PDF library | Quick Win |

**Action sequence:**
1. **Medication reminders** -- Quick Win. Ship in Sprint 1.
2. **Care plan PDF export** -- Quick Win. Ship in Sprint 2.
3. **Real-time vitals dashboard** -- Big Bet. Run a 2-week spike to scope device integrations.
4. **AI symptom checker** -- Big Bet. Defer to Q3 pending regulatory guidance.
5. **Dark mode** -- Fill-In. Add to backlog, pick up if capacity allows.
6. **Custom avatar** -- KILLED. Removed from backlog.

---

## Facilitator tips

- **Run this exercise with the whole squad.** PM, design, engineering, QA. Different perspectives change where items land on the grid. That's the point.
- **Timebox the debates.** If the team can't agree on placement after 2 minutes, note the disagreement and move on. You can revisit contentious items after all items are placed -- context from the full picture often resolves the argument.
- **Use physical movement.** Sticky notes on a whiteboard (or draggable cards in Miro) beat spreadsheets for this exercise. Moving things with your hands makes the prioritization feel real.
- **The KILL conversation is the most valuable part.** If you skip it, you haven't actually prioritized -- you've just sorted. Make space for the team to say "we are choosing not to do this" out loud.
- **Revisit quarterly, not weekly.** A 2x2 is a strategic tool. If you're re-running it every sprint, you're using it to avoid commitment, not to build it.
- **Don't over-index on precision.** The goal is directional clarity, not pixel-perfect placement. If an item is "somewhere between Quick Win and Big Bet," that's useful information -- it means feasibility is the open question.

---

## How did it go?

After running your 2x2 exercise, check:

- [ ] Every item on the board has a clear quadrant placement, not a "somewhere in the middle"
- [ ] The team explicitly discussed and agreed on the KILL decisions
- [ ] Quick Wins have owners and sprint assignments
- [ ] Big Bets have a next step (spike, research, or scoping) -- not just "we'll get to it"
- [ ] The action sequence is short enough to actually execute (3-5 items, not 15)
- [ ] Fill-Ins are clearly labeled as "if capacity allows" -- not quietly treated as priorities
- [ ] The team walked away with fewer things to do, not more

---

*Part of the [k8 Agent Toolkit](https://k8mak.com/agent-toolkit). Download other formats at k8mak.com/resources.*