Skills / Product-management

GIST Planning

GIST is the result of my search for a better planning system — one that focuses product teams on business goals while enabling them to experiment rapidly and learn their way to success.

GIST Planning is a four-layer product management framework by Itamar Gilad that replaces the traditional feature roadmap with an evidence-driven, hypothesis-testing system. It connects company OKRs (Goals) to a scored Idea Bank, time-boxed validation experiments (Steps), and executable tasks — ensuring every build decision is grounded in learning, not opinion. Designed for product managers, founders, and growth teams who are tired of shipping features that don't move the needle.

By Itamar Gilad  ·  Free

Use this skill See it in action

Specimen 01 · Live diagnosisGIST Planning

Input

“We want to grow our user activation rate. We're a B2B SaaS product and too many signups never reach the 'aha moment'. It's killing our paid…”

Diagnosis

Let's build your GIST plan.

Full transcript ↓

Calibrated referenceagent-skills.ai

The gap

Replace feature roadmaps with evidence-tested bets — Goals, Ideas, Steps, Tasks

GIST organizes product planning into four hierarchical layers that flow top-down and feed back bottom-up. Goals are outcome-based OKRs (Objectives + 2–4 Key Results) set at company and team level — they answer *why* you are building anything. Ideas are hypotheses about how to achieve those goals, stored in a persistent Idea Bank and prioritized using the ICE formula (Impact × Confidence × Ease ÷ 3), not gut instinct or HiPPO opinions. Steps are short, time-boxed 'step-projects' (1–10 weeks) that run the minimum experiment needed to validate the top idea before full investment — each step has a written hypothesis, a specific success metric, a duration, and an investment cap. Tasks are the day-to-day stories and tickets that execute the current step. Learning from each Step updates Idea confidence scores and Goal alignment, creating a continuous evidence loop that replaces the rigid quarterly roadmap.

The problem

Most product teams are trapped in roadmap theater: a feature list approved by the highest-paid person in the room, built on output commitments instead of outcome hypotheses, with no mechanism to kill bad ideas before they consume six months of engineering time. The result is teams shipping features that don't move the needle while the real problems go unvalidated. GIST was explicitly designed to break this cycle by making every build decision an evidence-tested bet.

The solution

By the end of a GIST session you will have a scored Idea Bank and a Step-Project Brief for your top idea — a concrete 1–10 week experiment with a clear hypothesis and success metric, so you invest in learning before you commit to building.

You bring

Your current product or business context (stage, team size, market)
An existing roadmap, backlog, or list of ideas you are considering
Your business objectives or OKRs (even rough ones)
Honest uncertainty about which ideas are most likely to work

You get

A Goal Tree that cascades your company OKR down to actionable team-level Key Results
An ICE-scored Idea Bank with every idea ranked by Impact, Confidence, and Ease
A Step-Project Brief for your top idea, including hypothesis, success metric, time-box, and investment cap
A diagnosis of planning anti-patterns (HiPPO decisions, roadmap feature-lock, vague goals) with GIST-aligned rewrites

In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01  —  The same situation, two ways

Specimens A · B

Specimen AVague

Unaided

“You have a 40-item feature backlog, a Q2 roadmap approved by the CEO based on a customer call he had in January, and three engineers who keep asking 'but what's the most important thing?' You're shipping regularly but your activation and retention metrics haven't moved in two quarters.”

×Vague — nothing concrete to act on.

Specimen BCalibrated

With GIST Planning

“Your Idea Bank has 12 ideas, ICE-scored and ranked — you can defend every prioritization decision with explicit assumptions, not opinions. Your top idea has a Step-Project Brief: a 2-week, 40-hour experiment with a clear go/no-go threshold. If it validates, you scale. If it doesn't, you've spent 2 weeks learning instead of 3 months building the wrong thing.”

◦Specific, grounded, and ends with one clear next step.

Fig.02  —  A real session, unedited

In-app · assistant

GIST PlanningClaude

GIST Planning

Let's build your GIST plan. Before we touch any ideas or features, I need to anchor everything in a Goal. What business outcome are you actually trying to move in the next quarter — not what you're building, but what metric would tell you that you won, and why does it matter to the company? Take a first pass and we'll sharpen it into an OKR.

We want to grow our user activation rate. We're a B2B SaaS product and too many signups never reach the 'aha moment'. It's killing our paid conversion.

Good — that's a real outcome problem, not a feature request. Let's write it as a GIST Goal. Here's a draft:

Objective: Make every new signup reach their first value moment within 7 daysKey Result 1: Increase 7-day activation rate from 23% to 40%Key Result 2: Reduce time-to-first-value from 4.2 days to 1.5 daysKey Result 3: Increase trial-to-paid conversion rate from 8% to 14%

This is the top of your Goal Tree. Every idea we consider will have to trace back to at least one of these KRs — if an idea can't connect, it doesn't go in the Idea Bank. Does this capture the real target, or are the numbers off?

Those numbers look right. We have tons of ideas for fixing this — in-app tours, email sequences, a setup wizard, a 'quick wins' checklist. How do we decide which to actually do?

We score them with ICE: Impact × Confidence × Ease, each on a 1–10 scale. Here's your Idea Bank so far — I'll walk you through the scores:

| Idea | Impact | Confidence | Ease | ICE Score ||---|---|---|---|---|| Interactive setup wizard | 8 | 5 | 4 | 5.7 || Onboarding email sequence (5-touch) | 6 | 7 | 8 | 7.0 || In-app product tour (tooltip-based) | 7 | 4 | 6 | 5.7 || 'Quick wins' checklist widget | 7 | 6 | 7 | 6.7 |

The email sequence scores highest right now — mainly because Confidence is 7 (you've seen email work in analogous products) and Ease is 8 (you can run it without engineering). But before we accept this ranking, tell me: what's your actual evidence for each Confidence score? Gut feeling, customer research, or data?

Honestly, the email sequence confidence is based on best practices we read about, not our own data. We've never tested it on our users.

That's exactly what GIST is designed to surface. If that Confidence is really a 4 (industry analogy, no direct evidence), the email sequence drops to ICE 6.0. Meanwhile the quick-wins checklist stays at 6.7 because you have qualitative signals from user interviews. Let me update the bank:

| Idea | Impact | Confidence | Ease | ICE Score ||---|---|---|---|---|| 'Quick wins' checklist widget | 7 | 6 | 7 | 6.7 || Onboarding email sequence | 6 | 4 | 8 | 6.0 || Interactive setup wizard | 8 | 5 | 4 | 5.7 || In-app product tour | 7 | 4 | 6 | 5.7 |

The checklist is now your top idea. Let's design a Step-Project to validate it before you build the full version — what's the smallest experiment that would tell you whether it actually drives activation?

We could mock it up and get it in front of 50 users in two weeks. That feels doable without a big engineering investment.

Perfect. Here's your Step-Project Brief:

Step-Project: Quick Wins Checklist — Activation ValidationHypothesis: If new signups see a 5-item 'quick wins' checklist in the dashboard, they will complete at least 3 items within 7 days, increasing activation rate by ≥8 percentage points.Success Metric: 7-day activation rate for checklist-exposed cohort vs. control groupGo threshold: +8pp activation OR ≥60% of users completing 3+ checklist itemsDuration: 2 weeksInvestment cap: 40 eng-hours (static mockup + basic instrumentation, no persistence layer)Next step if validated: Full engineering build with dynamic checklist personalizationNext step if not validated: Revisit Confidence scores for email sequence, design Step 2

This is the GIST loop in action: you've gone from vague goal to scored idea bank to a bounded, measurable bet — in one session. The 40-hour cap ensures a failed experiment costs two weeks of learning, not two quarters of building.

This is way more concrete than our usual roadmap. I can actually bring this to the team and defend every decision.

Reply to GIST Planning…

What the skill is doing

01Goal Tree Construction

02ICE-Scored Idea Bank

03Step-Project Design

Fig.03  —  The same skill, in the channel you already use

WhatsApp · live

9:41

GPGIST Planningtyping…

Messages are end-to-end encrypted. No one outside of this chat can read them.

Today

We want to grow our user activation rate. We're a B2B SaaS product and too many signups never reach the 'aha moment'. It's killing our paid conversion…9:14✓✓

Before we touch any ideas or features, I need to anchor everything in a Goal.9:14

Take a first pass and we'll sharpen it into an OKR.9:15

Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

◦Reads the situation, names the pattern, returns one concrete next move.

◦Delivered in seconds, inside a conversation that already exists.

Specimen · WhatsApp Business API · live

Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

Goal Tree Construction

Guides you through writing a GIST Goals layer: one Objective (qualitative direction) and 2–4 Key Results (measurable outcomes) at both company and team level. The result is a Goal Tree that makes it immediately visible whether any idea actually supports a business goal — eliminating work that exists only because someone asked for it.

◦ MethodBased on Gilad's Goals layer in GIST, which uses OKR format (Objective + Key Results) and requires every idea in the Idea Bank to trace back to at least one active Key Result.

CapabilityC-02

ICE-Scored Idea Bank

Elicits your full idea set — from anywhere: customer requests, competitor features, internal suggestions, data insights — and scores each one using the ICE formula: Impact (how much will this move the Key Result?), Confidence (how sure are we?), Ease (how cheap/fast to test?). The output is a ranked, visible Idea Bank where prioritization is explicit and arguable, not opaque.

◦ MethodDirectly implements Gilad's ICE scoring model (Impact × Confidence × Ease ÷ 3), the primary prioritization mechanism of the Ideas layer, designed to replace HiPPO-driven backlog ordering.

CapabilityC-03

Step-Project Design

Takes your top-ranked idea and builds a Step-Project Brief: a structured hypothesis card that defines exactly what you're testing, what metric would confirm the idea works, how long the experiment runs, and how much you're willing to invest before deciding to scale or pivot. A Step is not a sprint — it is a bounded bet.

◦ MethodImplements Gilad's Steps layer, where each step-project must include a written hypothesis, a success metric tied to a Key Result, a 1–10 week duration, and an investment cap — enforcing minimum viable experimentation before full build commitment.

CapabilityC-04

Roadmap-to-GIST Conversion

Audits an existing feature roadmap and restructures it as GIST: output items are reframed as idea hypotheses, features without goal alignment are flagged, and committed delivery dates are replaced by step-project experiments. The converted artifact is a GIST-compatible Idea Bank extracted from what was previously a delivery calendar.

◦ MethodApplies Gilad's explicit anti-roadmap principle: roadmaps are 'a list of features masquerading as a strategy'; the GIST Idea Bank replaces them with ranked, evidence-keyed hypotheses.

CapabilityC-05

HiPPO & Anti-Pattern Diagnostics

Scans your current planning process for the five most common GIST anti-patterns: HiPPO decisions (Highest Paid Person's Opinion overrides evidence), output roadmaps (features committed without validation), vague goals (objectives that can't be measured), the hero idea trap (one big bet with no backup), and confidence inflation (ICE scores based on hope, not data). Each flagged pattern comes with a GIST-aligned correction.

◦ MethodGrounded in Gilad's documented critique of traditional PM planning, including his named concept of HiPPO-driven decision-making and the 'planning-execution gap' he identifies as the primary failure mode in product organizations.

Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces

OutputD-01

ICE-Scored Idea Bank

A ranked table of all your ideas, each scored on Impact (1–10), Confidence (1–10), and Ease (1–10) with an ICE composite score. Includes a goal-alignment column showing which Key Result each idea supports. Instantly shows where team energy should go — and which ideas are just noise.

OutputD-02

Step-Project Brief

A one-page hypothesis card for your top idea: the specific claim being tested, the leading metric that would confirm it, the time-box (e.g., 3 weeks), the investment cap (e.g., $4K / 120 eng-hours), and the go/no-go decision rule. The Step-Project Brief replaces a feature spec with a learning commitment.

OutputD-03

Goal Tree

A visual cascade from your company-level Objective down to team-level OKRs, showing exactly which Key Results each team owns and how they roll up to the North Star Metric. Makes goal misalignment visible before it becomes a roadmap conflict.

OutputD-04

Evidence Board

A validation status matrix across your Idea Bank: which ideas have been tested, at what Step, what the result was, and how the outcome updated the idea's Confidence score. Replaces tribal knowledge about 'what we tried before' with a durable, team-wide learning record.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Itamar Gilad

Itamar Gilad is a product management coach, consultant, and author who spent eight years as a Senior Product Manager at Google and held senior product roles at Booking.com. He introduced the GIST framework in a 2016 Medium post that became one of the most widely shared pieces in the product management community, and codified the system in his 2023 book Evidence-Guided. He writes The Evidence-Guided Newsletter on Substack and advises product teams worldwide.

Status · Inspired by Itamar Gilad’s work — not yet claimed. Are you Itamar Gilad?

Primary sourceS-01

Evidence-Guided: Creating High Impact Products in the Age of Uncertainty (2023)

by Itamar Gilad

Former Senior PM at Google and Booking.com; author of Evidence-Guided (2023); originator of the ICE scoring model and GIST framework adopted by product teams globally.

Read the original ↗

Citationitamargilad.com

In the build queue

Be first to run it.

GIST Planning is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail

At launchI have a backlog full of ideas and a roadmap my team doesn't really believe in. Can we run through the GIST framework to figure out what we should actually be working on? I want to start by getting our goals right.