# Humane Product Canvas

**A snapshot, not a blueprint.** Fill a few cells at a time, revisit as you learn.
Blanks are honest signal — leave `_(not yet)_` on anything you don't know. Never
fabricate an answer to make it look finished.

Humane Product Canvas by Ajesh Shah · CC BY-SA 4.0 · https://boss.build

---

**Idea:**
**Date:**
**Version:**

---

## 1 · Human Foundation

*Who you serve, the tension they carry, the value you promise.*

### People — who are you designing for, and what matters to them?

Who *exactly* has the painful problem? Be specific — not "everyone." And who feels it
worst *today* — the early adopters you could actually reach this month, not the market
you'd eventually serve. **How many of them are there, and how do you know?** Count from
the bottom up: a list, a forum, a conference, a job title you can filter on. *"$50B
market"* is not an answer; *"about 4,000 of these exist and I can name where they
gather"* is. A number whose source you can't say out loud doesn't go in this cell.

>

### Problem — what real human tension are you solving?

Is it urgent, frequent, expensive, or emotionally painful? What do they use *today*
instead? And who else is selling a fix — not just tools, but the spreadsheet, the
agency, the intern, and doing nothing. **For each real one, say why they might win.**
A competitive picture where everyone else is dismissed is a picture you drew rather
than looked at.

>

### Promises — what value will it deliver?

The sharp promise: *"We help X do Y without Z."*

>

---

## 2 · Product Expression

*How it shows up in a life, how people engage, how it sustains itself.*

### Story — how does this show up in someone's life?

What changed that makes this newly possible or urgent? What's the smallest compelling
workflow that solves it?

>

### Modes of Engagement — how do people interact with it, humanely?

Does it respect time, attention, autonomy, data? Why are you the right person or team
to do this — and **what in your background makes that credible to someone who doesn't
know you?** Not a CV: the specific thing you've seen, built, sold or lived. What *kind*
of relationship is this — self-serve, hands-on, community, a tool they forget they're
using? Name it, because it sets what you'll owe them later.

>

### Business Model — how will you sustain this without compromising your promise?

**Take the branch that fits. Not every project is trying to earn, and the question is
the same either way.**

**If it will earn:** Who pays, how much, why is it worth it? How do the first 100 find
you? What makes usage compound? And once the first 100 are in, which channels keep
working — and which were just you hustling?

**If it won't** — open source, Creative Commons, research, a thing for friends, a thing
for fun: **What keeps this alive?** Your own hours are the budget: how many a week, and
for how long — a season, a year, indefinitely? Name it, because "indefinitely" is the
answer that ends projects. **Who else could carry it?** One maintainer is a single point
of failure. **What happens when you get bored, busy, or hit by a bus** — archive it, hand
it over, let it sit read-only? **And what would make you stop?** Not failure: a stated
condition, so stopping stays a decision rather than a slow fade.

>

### Cost structure *(once there's a price, or a real cost)*

What does it actually cost to serve one person well — and does that cost hold as more
arrive? Fixed vs. variable. Revenue without cost isn't a model, it's a price.

>

### What it takes to deliver *(once the answer isn't "just me and a laptop")*

The few resources and repeated activities the promise depends on — the data you need
access to, the thing you must do every week, the skill you can't outsource.

>

### Key partnerships *(only if someone else is load-bearing)*

Is there anyone whose cooperation this cannot work without? A supplier, a platform, a
regulator, a distribution partner, a data source. **Most ventures have none** — blank is
the common and correct answer.

>

---

## 3 · Stewardship

*Impact, risks, and the values that guide decisions.*

### Metrics — what does meaningful success look like, for people as well as the business?

Real pull: activation, retention, conversion, willingness to pay. **And the regenerative
ones, weighted as seriously:** what people *learn* or gain in well-being, and what makes
the venture more resilient over time — growth that renews rather than extracts.

>

### 🔴 Risks & Harms — what could go wrong, and who might be harmed or excluded?

**One of the two cells no conventional canvas has.** Rank the assumptions that could
kill the idea. Be honest about harm at scale. **Name the worst-served person, not an
abstraction.** Ask plainly: who is this worst for? Who could be excluded by how it
works? What happens to someone who uses it at 3am, or in crisis, or without the context
everyone else has? What does it cost them in attention, autonomy, or dignity? What's the
harm if it works exactly as designed — and what's the harm if it fails?

>

### Build or buy? *(for tool-shaped ideas)*

Is the honest move to *build* this — or would using something that already solves it
serve the person better? Most people want the problem *gone*, not a system to maintain.
The humane answer is sometimes "don't build it yourself." For a genuine product venture,
building **is** the move — skip this one.

>

### 🔴 Principles — what values will guide your decisions?

**The other cell no conventional canvas has.** The non-negotiables you'll hold **even
when holding them costs you something.** A principle you've never paid for isn't one yet.

>

---

## The riskiest assumption

Everything above is downstream of this one.

**If this is wrong, the idea doesn't work:**

>

**The experiment this week** — the smallest test that would prove or disprove it. Often a
15-minute conversation with the right person.

>

**What result would change the plan?** *(decide before you run it)*

>

---

*Humane Product Canvas by Ajesh Shah — licensed CC BY-SA 4.0. Share it, adapt it, use it
commercially; keep the credit and share adaptations under the same licence.*
