home / skills / coowoolf / insighthunt-skills / frontier-of-understanding

frontier-of-understanding skill

safe

/decision-thinking/frontier-of-understanding

This skill helps teams move from outcome targets to learning and understanding frontier, ensuring goals match levers and reduce risk.

npx playbooks add skill coowoolf/insighthunt-skills --skill frontier-of-understanding

Review the files below or copy the command above to add this skill to your agents.

Files (1)

SKILL.md

3.7 KB

---
name: Frontier of Understanding (NCTs)
description: Before setting outcome goals, identify your understanding level. If you don't know the levers, set a learning goal, not a revenue goal. Don't commit to outcomes you can't control.
---

# The Frontier of Understanding (NCTs)

> "If you don't understand how to move a particular metric, then the right goal is to set a goal to increase your understanding not to move that metric." — Ravi Mehta

## What It Is

Instead of blindly focusing on outcomes, teams should identify their **"Frontier of Understanding."** If the levers are unknown, the goal should be "Understanding Risk" (learning); if known, the goal can be "Execution Risk" (doing) or "Strategic Risk" (outcomes).

## When To Use

- Quarterly planning when leadership demands **metric increase**
- Team has **no clear hypothesis** on how to achieve
- Avoiding "throwing spaghetti at the wall"
- Setting realistic, achievable goals

## The Risk Levels

```
┌─────────────────────────────────────────────────────────┐
│  UNDERSTANDING RISK                                     │
│  "We don't know the levers"                             │
│  → Goal = Insight / Learning                            │
├─────────────────────────────────────────────────────────┤
│  DEPENDENCY RISK                                        │
│  "We know levers but lack tools/resources"              │
│  → Goal = Unblock dependencies                          │
├─────────────────────────────────────────────────────────┤
│  EXECUTION RISK                                         │
│  "We have the tools"                                    │
│  → Goal = High velocity / Quality experiments           │
├─────────────────────────────────────────────────────────┤
│  STRATEGIC RISK                                         │
│  "We are executing well"                                │
│  → Goal = Verify hypothesis moves the metric            │
└─────────────────────────────────────────────────────────┘
```

## How To Apply

```
STEP 1: Identify Your Frontier
└── Do we know what moves this metric?
└── Have we proven the levers work?

STEP 2: Match Goal to Frontier
└── Unknown levers → Learning goal
└── Known levers → Execution goal
└── Proven levers → Outcome goal

STEP 3: Don't Overcommit
└── If in Understanding phase, don't promise revenue
└── Promise insights instead

STEP 4: Move Along the Frontier
└── Each quarter, advance your understanding
└── Eventually you earn the right to set outcome goals
```

## Common Mistakes

❌ Setting an **outcome goal (Revenue)** when in "Understanding Risk" phase

❌ "Throwing spaghetti at the wall" to hit arbitrary targets

❌ Treating all goals as **equally achievable**

## Real-World Example

At Tinder, data showed high spending from a small group. Instead of blindly trying to grow revenue, they set a goal to understand WHY. They found these weren't rich people, but frequent travelers/salespeople. This insight led to "Tinder Platinum."

---
*Source: Ravi Mehta, Former CPO of Tinder, Lenny's Podcast*

Overview

This skill teaches teams to diagnose their "Frontier of Understanding" before setting outcome goals. It reframes goal-setting: if you don't know the levers that move a metric, set a learning goal; only commit to outcome goals when levers are proven. The approach reduces wasted effort and prevents overcommitment to metrics you can't control.

How this skill works

You assess where your team sits on a risk spectrum: Understanding, Dependency, Execution, or Strategic. Based on that assessment you pick an appropriate goal type—insight, unblocking, execution, or outcome—and align quarterly plans to advance understanding one step at a time. The method emphasizes promising learnings when levers are unknown and reserving revenue or target commitments until levers are validated.

When to use it

Quarterly planning when leadership asks for metric growth
When the team lacks a clear hypothesis for how to move a metric
To avoid unfocused experiments or "spaghetti" approaches
When deciding whether to promise outcomes or learning milestones
When dependencies or tooling gaps block progress

Best practices

Start every planning cycle by asking if you know the levers and whether they are proven
Define clear learning goals with measurable signals (tests, experiments, qualitative interviews)
Treat insights as deliverables: document what was learned and next steps
Use dependency goals to explicitly unblock resources or tools before execution
Only convert learning signals into outcome targets after repeated validation

Example use cases

A product team asked to increase revenue but lacking a hypothesis sets a quarter goal to run 8 experiments and validate top 3 levers
A growth squad knows the levers but lacks instrumented events, so the goal is to unblock analytics and tools
A new feature shows promise in small tests; set an execution goal to scale experiments with quality metrics rather than a revenue number
Leadership wants bold targets—present a learning roadmap that earns the right to outcome commitments in subsequent quarters

FAQ

What counts as proof that a lever is known?

Repeated experimental evidence showing the lever moves the target metric reliably, plus instrumentation and a clear causal hypothesis.

How long should a team stay in an Understanding phase?

Long enough to generate reproducible signals—typically 1–3 quarters depending on experiment cadence and signal strength. Move on when you can predictably replicate effects.