Turbo AI PM
Kits / workflow

Copilot / drafting tool

Drafts text or edits a person reviews before using. Work top to bottom, or jump to the decision in front of you — each one links to the metrics, tool, and playbook that answer it.

The riskSuggestions users accept and then quietly rewrite — usage looks healthy while time saved is zero.
Used in product scenariosWorkflow Automation
Take it to your PRD, Notion, or launch review.

The chain

What has to be true at each layer for this product to create value. The hypotheses between layers are yours to hold — and to check after launch.

Model / SystemIs the AI behaving correctly?
P-01 Latency (p50 / p95 / p99)↔ C-01S-01 Hallucination Rate↔ S-03Q-07 BLEU / ROUGE↔ B-06
If suggestions are fast and on target, users accept them without heavy rewriting.
Product / UserIs the user accomplishing the task?
B-07 Acceptance Rate↔ B-06B-06 Human Override / Edit Rate↔ B-07B-01 Adoption Rate↔ B-04
If accepted drafts replace manual writing, time saved compounds into retention and seat expansion.
BusinessIs the product creating meaningful value?
Time saved per user · retention · seat expansion

↔ marks each metric's shadow metric — the one to watch alongside it so the number can't improve while the product gets worse.

01

Before you build

Should this be AI at all — and which capability fits?

Before scoping a model, check that the problem has a pattern to learn, data you can reach at decision time, and a cost of being wrong you can live with. If a rule would get most of the value, the rule is the better product.

For CopilotThe test for a copilot: do people write this by hand often enough, and is a near-miss draft genuinely faster to fix than starting fresh? If editing costs more than writing, there is no value to capture.
Check whether this needs AI →
Evidence, business implication, deeper reading
Evidence you need: Examples of the decision being made today, where the data lives and whether it is available at decision time, and an honest estimate of what a wrong answer costs.
Business implication: Choosing AI for a problem rules could solve buys you cost, latency, and a maintenance burden with no extra value.

Check whether the unit economics can work

Estimate cost per completed task against the value of that task before building — margins are set by architecture, not tuning.

MetricsC-02 Cost per Resolved Task↔ B-02C-01 Cost per Inference↔ P-01
Estimate ROI & unit economics →
Evidence, business implication, deeper reading
Evidence you need: Volume estimate, token estimate per task, model pricing, and the human cost (or revenue) per task.
Business implication: A feature that works but destroys gross margin is a failed product.

Make the case to build it

Ask for one specific decision by a specific date: the problem in their units, why AI and not rules, the expected value with its assumptions visible, the risk said out loud, and the checkpoint you will report back at.

MetricsB-07 Acceptance Rate↔ B-06
Build the business case →Make the case & defend the work →
Evidence, business implication, deeper reading
Evidence you need: The baseline (what today costs and achieves), a value range with the two assumptions that move it most, cost per completed task vs. the human cost, the rules-vs-model comparison, and the top failure modes with their mitigations.
Business implication: A named checkpoint turns a bet into a staged decision — much easier to approve than an open-ended one. For this product: Time saved per user · retention · seat expansion.

Define what the model may do, must never do, and needs to know

Write the operating envelope before anyone writes a prompt: context the model sees, unforgivables, and what happens when it fails.

For CopilotDefine voice and tone, and exactly which parts of the user's document the copilot sees.
Write the AI PRD →
Evidence, business implication, deeper reading
Evidence you need: Agreed context spec, a written list of unforgivables, and a fallback for each failure mode.
Business implication: Scope decided here drives both cost and risk for the life of the feature.
02

Evaluate

Choose what to measure

Pick one metric per layer and a shadow metric for each, so no number can improve while the product gets worse.

MetricsB-07 Acceptance Rate↔ B-06B-06 Human Override / Edit Rate↔ B-07B-02 Task Completion Rate↔ B-06B-05 CSAT / Thumbs-up Rate↔ B-03P-01 Latency (p50 / p95 / p99)↔ C-01C-01 Cost per Inference↔ P-01
Calculate any of 38 metrics →
Evidence, business implication, deeper reading
Evidence you need: The starter metrics below, each paired with the metric that catches it being gamed.
Business implication: The layer-3 metric is the one leadership will ask about — name it now. For this product: Time saved per user · retention · seat expansion.

Establish a baseline worth beating

A number means nothing without a comparison. The simple heuristic and the incumbent are the two baselines that get skipped.

For CopilotUsers writing from scratch: time-to-draft on comparable tasks without the copilot.
MetricsP-01 Latency (p50 / p95 / p99)↔ C-01B-07 Acceptance Rate↔ B-06
Evidence, business implication, deeper reading
Evidence you need: Zero-rule, heuristic, and incumbent scores on the same eval set, measured the same way.
Business implication: If a heuristic gets 80% of the value, the model has to justify its extra cost.

Design an evaluation you can trust

Representative data, a scoring method that matches the task, and a holdout nobody tunes against.

MetricsP-01 Latency (p50 / p95 / p99)↔ C-01S-01 Hallucination Rate↔ S-03Q-07 BLEU / ROUGE↔ B-06
Evidence, business implication, deeper reading
Evidence you need: A golden set with an owner and a size target, slices defined up front, and a judge calibrated against humans.
Business implication: A weak eval means you ship blind and cannot explain failures afterward.
03

Ship

Decide whether it is ready to ship

The bar depends on blast radius and reversibility, not on the accuracy number alone.

For CopilotUsually green: a human reviews every draft. Ship earlier, with easy edit and dismiss.
MetricsP-01 Latency (p50 / p95 / p99)↔ C-01S-01 Hallucination Rate↔ S-03
Decide go / no-go →Make the case & defend the work →
Evidence, business implication, deeper reading
Evidence you need: Your quadrant, the threshold for it, the eval result against the baseline, a working fallback, and the numbers you promised at the funding checkpoint.
Business implication: A documented bar turns post-launch blame into post-launch learning.

Design what users see when it is wrong

Every AI flow needs a designed failure state: specific copy, a way to recover, and a human path.

For CopilotA visually distinct draft state — one keystroke to accept, one to dismiss.
Design for non-determinism →
Evidence, business implication, deeper reading
Evidence you need: Fallback copy for timeouts, low confidence, out-of-scope requests, and refusals.
Business implication: Users forgive visible, recoverable failures; they abandon silent ones.
04

In production

Set up monitoring that separates "the model changed" from "the world changed"

Alert on breakage, review drift daily, review quality weekly, review business impact monthly.

MetricsP-01 Latency (p50 / p95 / p99)↔ C-01S-01 Hallucination Rate↔ S-03Q-07 BLEU / ROUGE↔ B-06
Evidence, business implication, deeper reading
Evidence you need: Error and latency alerts, input-mix and confidence drift, slice quality, and the layer-3 metric on a monthly cadence.
Business implication: Degradation you catch in a week costs far less than one you find in a quarterly review.

Turn production signal into a better next version

Design what each interaction captures and where it goes — implicit, explicit, and outcome signal feeding the eval set.

For CopilotEdit diffs on accepted drafts are free, high-quality training signal.
MetricsB-07 Acceptance Rate↔ B-06B-06 Human Override / Edit Rate↔ B-07
Evidence, business implication, deeper reading
Evidence you need: A named owner for the signal pipeline and a path from user action to golden set.
Business implication: The flywheel is the only part of an AI feature that compounds over time.
!

Something looks wrong

Start from what you're seeing. Each symptom points to its most likely cause and the metrics that confirm it.

Acceptance is high, but users are not faster
Accept-then-rewrite: users take the draft and redo most of it.
CheckB-06 Human Override / Edit Rate↔ B-07B-07 Acceptance Rate↔ B-06
Judge quality score up, acceptance flat
The judge rewards something users do not value — often length.
CheckB-07 Acceptance Rate↔ B-06Investigate →
Usage drops after the second week
Novelty wears off once users learn where the copilot is unreliable.
CheckB-04 Retention (AI feature)↔ B-01B-06 Human Override / Edit Rate↔ B-07

Other kits

RAG / knowledge assistant Agents & multi-step pipelines Customer-facing chatbot / support Classifier / triage & moderation Coding assistant / IDE copilot