Skip to content

Product Principles

beta  ·  Family standing-standards  ·  Phase undefined  ·  Sizes lean  ·  ~7,750 tokens

The short standing set of beliefs a product team agrees once and then uses to settle product trade-offs: what to build, for whom, and what to refuse. Each principle carries a statement, why the team holds it, what it costs, what it rules out and a decision it settled. It is judged by whether a reasonable team could choose the opposite, because a principle nobody would reverse cannot settle an argument.

The short card. The reasoning behind every rule here, with the sources, is in product-principles_companion.md. A fully worked instance is product-principles_example.md. Where this card calls something the library’s position, no source read for the bundle publishes it, and the card says so each time.

What this document decides. A product principles document settles which way a product decision should lean when two good options compete: what the team builds, funds or declines. That reading goes past Cagan’s own definition, which speaks of the values and beliefs that inform product decisions, so it is the library’s. It does not settle how a screen, a message or an interaction behaves for the person using it. That test is the library’s own inference, at medium-low confidence. The sources split product principles from design principles by scope, and no source draws the line in these words. Published sets cross it, so some principles will straddle. The Purpose and Scope section has a field where the document says which ones.

What to call it. Write “product principles” for the standing set that settles what the team builds. One vendor source, The Org, uses “principles of product” for the design phase, so that wording invites confusion with the sibling bundle design-principles.

  • The team keeps arguing the same trade-off between two good options, and nothing written says which way to lean. Cagan’s 2020 position is that principles help a team see which values it prioritises when it makes a trade-off. One vendor’s AI prompt treats three or more debates in a quarter that a new principle would have settled as a reason to add one. That is the vendor’s figure, not a measured one.
  • A request keeps arriving that the team feels it should decline, and nobody can say why in a sentence. The set can hold the reason. Sources assert that principles are used to say no, and this research found no named company case that shows it, so treat that benefit as unproven.
  • Leadership has written a product vision and a strategy and wants the beliefs that sit beside them written down. Cagan places principles beside the vision and the strategy, and his 2020 text has product leaders preparing the set for every team.
  • A new product or team has joined, and nobody can say whether the existing set covers it. The Purpose and Scope section holds the answer.
  • A new product manager needs to learn how the team decides. Intercom puts a first problem statement in its new-joiner onboarding pack, and says it uses its principles to assess and debate options.
  • A set already exists and nobody cites it. Grade it with the rubric below, then rewrite the rows that score 0.
  • The strategy has changed significantly and the set has not been read against it. Productboard’s prompt names that as a condition for revisiting a set, and the Review Trigger section carries it.

Several documents get mistaken for product principles. The table says which one you likely need.

You actually need Because
Design principles Those settle how a screen, a message or an interaction behaves. This document settles what the team builds, funds or declines. The sources split the two by scope (ProductPlan, airfocus, Marc Abraham). The line in these words is the library’s, at medium-low confidence. When a principle straddles, say so in the Purpose and Scope field built for it. See the design-principles guide.
A few rules inside a strategy The Product Principles section of the product-strategy one-pager holds two to five rules inside a strategy. This bundle holds the standing set, with a reason, a cost, a boundary and a settled decision for each principle. The choice is a matter of size, and the library’s.
A project’s own principles Cagan’s 2005 guide places them inside a PRD, “specific to your domain and your particular project”. His later writing has principles not tied to one release. Put a project’s own principles in that project’s PRD. Intercom keeps both a standing set and occasional per-project principles, so the two can coexist.
Mission, vision or goals ProductPlan calls conflating the three with principles “a common mistake”. The product vision says where the product is going, and the product strategy says how it will get there. A principle says which way a trade-off leans on the way. That last reading is the library’s.
Company values The sources read that address values (Abraham, Guru, ProductPlan) put them upstream and broader, and Cagan’s 2020 text has principles state values and beliefs. No source gives a test between a value and a principle. The product-strategy one-pager’s TRAP supplies one, which is the library’s: values describe how you want to behave, and principles decide arguments.
A feature list or a release plan Cagan writes that principles are “not a list of features” and are not tied to one release. Ask whether the line could be shipped and finished. A principle is never reached.
A rule book or a compliance checklist Intercom writes “Principles are not rules - they exist to guide, not control.” A set meant to be checked off item by item is a different document.

This bundle ships one size, lean, so there is no variant to choose. The reason is in product-principles_companion.md section 4. No source read publishes a long and a short version of one set, and confidence in a single size is medium. Four choices remain inside it.

  1. How many principles. Three to five is advice, not a finding, and the template ships three principle blocks. Sources say three to five, up to seven, or fourteen in layers, and none cites a measurement. Add a fourth or fifth only when you can attach a real decision to it.
  2. What level the set covers. Company, product line or product. Sources place sets at each level, and the template asks the author to state which. A team that needs only two or three rules inside its strategy should use the one-pager’s section instead.
  3. How the conflict section resolves. The template asks for a winner in a named kind of decision. The library chose that over a global ranking, and no source opposes ordering outright. A principle written in an “X over Y” form may carry part of its own resolution.
  4. Which fields to leave honest. If a principle has no settled case yet, write that plainly and give a line of why. Do not invent a case, a request or a source.

Run these tests on each principle before you grade

Section titled “Run these tests on each principle before you grade”
  1. The opposite test. Intercom writes “no principle is a truism”, and its test is to flip the principle around to express the exact opposite and see whether it holds up. AWS, writing about tenets, says the same of an opposite an organisation could accept, and a vendor’s AI prompt agrees. Name the team or product that would reasonably choose the opposite. If you cannot, rewrite the statement. One vendor, airfocus, wants a principle that triggers an emotional response, and a line nobody would oppose can pass that test. The template follows the opposite test, because a reader can apply it.
  2. The cost test. Name what the principle gives up and who bears it. Gazit’s rule is to take a stand by preferring one alternative over the other. Box, as Pendo reports it, published a named cost with each principle: “The tradeoff? Expense.” If the cost is blank, the principle has taken no stand. This is a field the library asks for, and published sets show it in several forms.
  3. The say-no test. Write down a choice the team could have made and can no longer make because of the principle. That is the “What it rules out” field. No published set read carries it per principle, so it is the library’s request, and the template says so.
  4. The decision test. Name one real decision the principle settled and the option it turned down. If you cannot attach a decision to a principle, it is probably not yet a principle. That sentence is the library’s advice, not a sourced finding. An honest “no decision has met this yet” is allowed, and row 6 scores it.
  5. The what-does-it-settle test. Ask what the principle decides. If it decides what the team builds, funds or declines, it belongs here. If it decides how a screen, a message or an interaction behaves, it belongs in design-principles. If it decides both, say so in the straddling field.

This bundle ships one size, so all twelve rows apply and no scope table is needed. Score each row 0, 1 or 2. A score under 16 out of 24 means several rows are thin or empty, so the set will not yet settle an argument. A product manager facing a trade-off will find principles that can be read either way, or that cost nothing, or that nobody owns, and the decision goes back to whoever is in the room. The line at 16 is the library’s judgement and is not measured. The total does not say which rows failed. Read the rows that scored 0 first, because a 0 on row 2 (a principle that cannot be reversed) or on row 7 (no way to choose between two principles) leaves a set that cannot decide, even when other rows score well. Grade what the author added to the blank template. The template already prints the field labels and the five conditions in the Review Trigger table, so a row that rewards their presence is satisfied before anyone writes a word.

# Criterion 0 1 2
1 Scope is bounded No product is named, or the scope covers “everything we make” The product and its level are named, but nothing says which kinds of decision the set leaves alone, or how it sits beside the vision, the strategy and the design principles You can point at the sentence naming the product, the level and the people who apply the set, at one kind of decision the set refuses to settle and where that decision goes instead, and at the line saying how the set sits beside the strategy and outlives it
2 Principles can be reversed The statements are ones no thoughtful team would reject, such as “put the customer first”, or the section is blank Some principles can be reversed, and others pass only because nobody would oppose them For every principle, a colleague who did not write the set can state the opposite approach and name a team or product that would reasonably choose it
3 Costs are named The trade-off field is blank, says “none”, or repeats the statement A cost is named, but it is vague, or nobody is said to bear it, or the option it gives up is not stated For every principle you can point at the option it gives up and the people who bear the cost, and one of those people would recognise the cost as real
4 Refusals are real The rules-out field is blank, repeats the statement, or names “bad experiences” Some principles name a choice a team could have made, and others rule out only what nobody would choose For every principle you can point at a choice the team could have made and now cannot. Where a real request has been turned down, the request and who asked are named
5 Reasons are held The reasoning field is blank, says “customers matter”, or repeats the statement A reason is given, but a reader cannot tell what it rests on, or it would justify the opposite principle as well A new product manager who was not in the room can say, from the reason alone, why the team accepts the cost named in row 3, and you can point at the observation, decision or stated belief it rests on
6 Cases are real The field says “many”, describes a case no one can place, or is blank A case is described, but the option it turned down is not named, or the same outcome would have followed without the principle, or the field says plainly that no decision has met the principle yet You can point at the review, request or decision record where the principle settled something, the option it turned down is named, and someone who was there would recognise the account
7 A conflict is resolved The section says every principle matters equally, answers “use judgement”, or is blank Two principles are named and a winner is stated, but the kind of decision is “everything” or no reason is given You can point at two principles, a named kind of decision where they collide, which one wins there and why, and a product manager could apply it to the next decision of that kind without asking the author
8 Use has a place The section says product managers should keep the principles in mind, or is blank Moments of use are named, but the form of a citation is missing, or what a reviewer may do when a decision sets a principle aside, or how a new joiner meets the set You can point at the moments where the set is cited, at the line a decision record carries, including which principle was set aside and why, and at how a new joiner meets the set, and someone who was not in the room could write that line
9 Owner is reachable “The team”, or no owner A role is named without a person, or the place to propose a change or who decides is missing You can point at the role and the person who own the set, the place where anyone proposes a change and who decides, and you could message the owner today
10 Changes keep reasons The log is blank, or each entry is a date and the word “updated” Entries say what changed but not what fired the change, or a merge or retirement leaves no reason The first row gives the date the set was first agreed and who decided it. Each later row says what changed in a sentence a stranger could follow and names what fired it, and a merged or retired principle keeps its reason
11 Triggers carry owners The Review Trigger table still holds the template’s placeholders, or every owner is “the team”, or the only condition is a date Some conditions have a role and an action, but at least one has neither, or a condition that does not apply is left with the template’s wording instead of “N/A” and a reason Every condition has a role as owner and an action that person takes, such as reopening the set or writing a dated line in the change log, and any condition that has already fired points at its line in the change log
12 Decisions, not screens Several principles settle how a screen, a message or an interaction behaves, or the set restates a design principle, or the set holds no principles, or the straddling field is blank The statements settle what is built, funded or declined, but the straddling field says “None” with no reason, or a principle that also decides how a screen behaves is not named in it You can apply the what-does-it-settle test to every principle and each decides what the team builds, funds or declines, and the straddling field either names every principle that also decides how a screen behaves and says why it stays here, or says “None” and gives the reason

Every cell above describes evidence, not a count. The test for each cell is whether someone could satisfy it without improving the document. If they could, the cell is written wrong. The rubric does not grade how many principles the set holds, because the number is advice and not a finding.

Rows 3 to 6 grade fields that published sets carry unevenly. A name and a statement appear in every set the research read, and reasoning is common. A trade-off written into a principle is shown by several publishers. A per-principle rules-out field appears only in a vendor’s AI prompt among the sets and prompts tabulated, and Box’s per-principle “Versus” is a comparable field found outside that table. An examples field appears in two sets: Monzo’s is a case study of a settled decision, and Slack’s is a product example. An honest “no decision has met this yet” scores 1 on row 6, so a new set can clear the line without inventing a case. Grade with someone who did not write the set. The blank template should score 0 on every row.

Each entry gives the failure, the test that catches it and the rubric row it costs. The arguments behind them, and the disputes among sources, are in the companion’s sections 6 and 7.

  1. The truism. A principle whose opposite no reasonable team would choose, which cannot settle an argument however well it lands. Intercom: “no principle is a truism”. Test: name the team that would choose the opposite. Costs row 2.
  2. The feature list or the release plan. A line the team could ship and finish. Cagan says principles are not a list of features and are not tied to one release. Test: could the line ever be done? Costs rows 2 and 4.
  3. Mission, vision or goals relabelled. ProductPlan: “A product principle is never “reached.”” Test: does the line describe a destination or a measure, or a way to choose on the road there? Costs rows 2 and 4.
  4. No stated cost. A principle that names nothing it gives up has taken no stand. Gazit: “Take a stand by preferring one alternative over the other(s).” Test: who loses when this principle wins? Costs row 3.
  5. Too many. Sources say three to five, up to seven, or fourteen in layers, and none cites a measurement. Test: can you attach a real decision to each one? Costs row 6.
  6. A set nobody cites. The library’s inference is that a set with no stated moment of use risks being read once and not applied. The research found no measured effect of a set on decisions. Test: could someone who was not in the room write the citation line? Costs row 8.
  7. Principles treated as rules. Intercom’s podcast says “these are guides, they’re not rules”, and its blog says principles exist to guide and not to control. Test: does a decision cite the principle it leaned on and the one it set aside, with a reason? A citation so mechanical that it becomes a checkbox has lost that reading. Costs row 8.
  8. A claim that principles say no, with no case behind it. Pendo reports that Box does not pursue a feature idea that fails its principles, and PostHog argues that principles make it easier to say no. This research found no named company source describing a principle used to decline a specific request. Test: can you name the request and who asked? Costs rows 4 and 6.
  9. A principle the team consistently violates. A vendor’s prompt treats consistent violation without a decision to change the principle as a reason to reopen the set. The library reads a violated principle as either a wrong principle or a missing decision. Test: does a Review Trigger row cover it, with an owner? Costs row 11.
  10. Project principles inside the standing set. A principle that belongs to one project goes in that project’s PRD. Test: would it still hold on the next product line? Costs row 1.
  11. A set that restates the current strategy. The library’s own caution, not a sourced one. A principle that restates one period’s strategy choice will go stale when that strategy does, and no source ties a principle’s life to a strategy period. Test: would the principle survive a change of strategy? Costs rows 1 and 11.
  12. A design principle in the product set. A principle that settles how a screen, a message or an interaction behaves belongs to design principles. This is the library’s test. Test: the what-does-it-settle question. Costs row 12.
  13. No stated order when two collide. AWS expects tensions between tenets and says they need judgement. The library’s inference is that a set that never says which principle wins leaves the argument to whoever is in the room. Test: does the table name a winner in a named kind of decision? Costs row 7.
  14. A set that never changes, or changes with no record. Intercom’s record shows five principles merged into three, and a later set of four at the top level. Slack’s set differs between two snapshots with no stated reason. A set with no owner and no reopening condition has no way to do either on purpose. Test: does the log keep the reason for each merge or retirement, and does each trigger name an owner? Costs rows 9, 10 and 11.
  15. Borrowing another company’s set. No page the research read states a reuse licence, and Productboard reserves its rights. A set written for another company may not pass your opposite test. Test: write the team’s own words, then run the opposite test again. Costs rows 2 and 5.
  16. Presenting the library’s positions as received practice. A name and a statement appear in every published set read. What a principle rules out appears per principle only in a vendor’s AI prompt among the sets and prompts tabulated, with Box’s “Versus” a comparable field outside that table. A named cost appears in several forms, and an examples field in two sets (a case study at Monzo, a product example at Slack). The role names in Ownership and Changes, the dated change log and the table form of the Review Trigger are the library’s. A filled set may keep them and should not describe them as a standard.

This bundle ships in the standing-standards family as a foundation: a standard that a team’s product decisions are judged against. The reasoning is in ADR 0090 (product principles joins standing-standards as a foundation), and the family’s rules are in the standing-standards contract. Update the set when a Review Trigger condition fires, not on a calendar. The stated intervals disagree: Guru says a set should hold for at least a year, LogRocket says to review it every three months, Intercom says it updates every one to two years, and ProductPlan says an update should be unusual. None cites a measurement. A team that wants a review date may add one as a floor and should label it as its own choice.

The companion’s section 6 lists the open disputes: whether a set is standing or per project, whose set it is, whether a principle should be memorable or arguable, whether to rank, how many, how often to revise, and whether the set covers execution only or the choice of what to build. The research did not find who first wrote about a team keeping its own product principles, who owns a set after it is written or who approves a later change, any worked rule for two principles that conflict, or any measured effect of a set on decisions. It found no tie to a delivery methodology, and it found no research on regulated products or a solo author. The template is untested for both of the last two.

When a product manager who did not write the set can take the next real choice between two good options, name the principle that decides it, say what that principle costs and rules out, and say which principle wins if a second one pulls the other way. And when the page names who owns the set and what event reopens it.

Then delete every HTML comment. Keep the source_template and source_template_version frontmatter fields, because they are the only record of which template and version produced the document.

product-principles_template-lean.md · ~7,750 tokens

---
title: "{{product_name}} Product Principles"
doc_type: product-principles
size: lean
product_name: "{{product_name}}"
owner: "{{owner}}"
status: draft
doc_version: "{{doc_version}}"
created: "{{date}}"
updated: "{{date}}"
related_links: []
source_template: product-principles
source_template_version: 0.1.0
---
<!--
LEAN PRODUCT PRINCIPLES. This bundle ships one size, deliberately. No source the research read publishes
a long and a short version of one set. The sources that add depth add it per principle, not per
document, and so does this template. A long variant would model a weight the research did not find.
Confidence in the single size is medium, because it rests on the absence of a published pair. See
product-principles_companion.md section 4 (Variants and sizing).
A PRODUCT PRINCIPLES DOCUMENT IS A SHORT STANDING SET OF BELIEFS that a product team agrees once and
then uses to settle product trade-offs. Cagan defines the type: "Product principles complement the
product vision by stating the values and beliefs that are intended to inform the many product decisions
that will need to be made." He writes that principles are "not a list of features" and "are not tied to
any one product release", and that they are "aligned with the product vision for an entire product
line". This library reads the set as settling what the team builds and what it refuses. That reading
goes past Cagan's sentence, and Slack limits a set to execution, so it is held at medium-low confidence.
The set is a standard that product decisions are judged against, applied by judgement, not by
checkbox: Intercom writes "Principles are not rules - they exist to guide, not control." See
product-principles_companion.md section 1 (Orientation) and section 6 (Debates and contested boundaries).
SOME OF THIS STRUCTURE IS THIS LIBRARY'S OWN POSITION, NOT RECEIVED PRACTICE.
No source the research read publishes a template for this type, other than a vendor's AI prompt
(Productboard's). So the six sections here are assembled from published sets and method articles. Each
principle below carries five fields. A name and a statement appear in every set read. Reasoning is common.
A trade-off written into the principle is shown by several publishers. A field for what a principle rules
out appears per principle only in Productboard's prompt. Of the eight published sets tabulated, none
carries it: Langfuse publishes one set-level list, and Box, as Pendo reports it, a comparable "Versus".
This template asks for more than published sets do, and says so. Where a section is the library's
position, its comment says so. No source read measures whether keeping such a document improves
decisions. The research found nothing on solo practitioners, regulated products or tailoring by
methodology, so the template is untested for all three.
WHAT THIS IS NOT
It is NOT a feature list or a release plan. ProductPlan writes "You don’t build principles".
It is NOT mission, vision or goals. ProductPlan calls "conflating the company’s mission, product vision,
and goals with product principles" a common mistake.
It is NOT a project's own principles. Cagan's 2005 guide has principles "specific to your domain and your
particular project". This bundle is the standing set. A principle that belongs to one project goes in
that project's PRD (see prd_template-full.md). Intercom does both, which shows they can coexist.
It is NOT a few rules inside a strategy. That is a matter of size: the Product Principles section of the
product-strategy one-pager holds two to five rules inside a strategy, and this bundle holds the standing
set with a reason, a cost, a boundary and a settled decision for each principle.
It is NOT design principles. The line is this library's own test, and the sibling design-principles
bundle states it from the other side: ask what a principle settles. If it settles how a screen, a
message or an interaction behaves, it belongs in design principles. If it settles what the team builds,
funds or declines, it belongs here. No source draws the line in those words. Some principles straddle
it, and this document should say so rather than hide them, in the last field of Purpose and Scope.
Write "product principles" for this set. One vendor source (The Org) uses "principles of product" for the
design phase, so the wording invites confusion.
All names, dates and figures in the GOOD and WEAK illustrations below are invented.
HOW TO FILL THIS IN
1. Read the comment under each heading: WHAT it wants, WHY it matters (with a pointer into
product-principles_companion.md), guiding questions to ASK, a GOOD and a WEAK example, and the TRAP to
avoid. Three tables also carry PRIORITY and ROW HINT.
2. Replace each {{placeholder}} with your content. Draft Purpose and Scope first, then The Principles,
then When Principles Conflict, then the rest. Within The Principles, start from a decision the team
has argued about, ask which option a principle would have chosen, write what it rules out and what it
costs, and draft the statement last.
3. If a section, or one row inside a table, does not apply, write "N/A" and one line of why, rather than
leaving it blank or deleting the section.
4. This is a standing document, reopened on the Review Trigger below. Before you first share it:
self-grade against product-principles_guide.md, then DELETE every HTML comment. Keep the
source_template and source_template_version frontmatter fields: they are the only record of which
template and version produced this document. Write the principles in the team's own words. No page
the research read states a reuse licence, and Productboard reserves its rights, so a set copied from
another company is untested here.
-->
# {{product_name}} Product Principles
## Purpose and Scope
<!-- WHAT The product or products the set covers, the level at which it applies, what kinds of decision it
settles, how it relates to the product vision, the product strategy and the design principles,
and which principles, if any, also settle how a screen behaves.
WHY A set with no stated scope leaves a new product with no way to know whether the set covers it.
That reading is this library's. Sources place sets at different levels. Cagan's 2017 chapter
has principles "aligned with the product vision for an entire product line", and Slack
writes "Product principles should be company-wide, not siloed into the product or design
team." Kindel describes tenets at company, organisation, team and project levels. Stating the level is a
field this library adds, assembled from sources that each place a set at one level. What the
set settles is itself disputed. ProductPlan says principles "guide what you decide to build and
how to build it", and Cagan ties them to the many product decisions a team makes. Slack scopes
them to "inform execution rather than roadmapping, strategic thinking or planning". This
template follows Cagan and ProductPlan, which is the library's reading, held at medium-low
confidence because the sources read it both ways. An author who wants Slack's limit can state
it here. The line with design principles is this library's test. Intercom keeps five product
design principles inside its product principles, so some principles will straddle. The last
field asks the author to name them and say why, or to write "None" with a reason. No source
read asks for it.
Deep dive: product-principles_companion.md section 3 (Anatomy > Purpose and Scope).
ASK Which product or products does this set cover, and is it company-wide, per product line or per
product? Which kinds of decision does it settle, and which does it leave alone? How does it
relate to the product vision, the product strategy and the design principles, and does it
outlast the current strategy? Which principles, if any, also settle how a screen, message or
interaction behaves, and why?
GOOD "These principles cover Kestrel Ledger, the bookkeeping app for freelancers. They apply at the
product level, and the two product managers, the design lead and the engineering lead use
them. They settle what we build, fund or decline. They do not settle how a screen behaves,
which belongs to the design principles. They sit beside the vision and the current strategy,
and they outlast the strategy." For the last field: "Principle 2, File it for them before we
explain it, also reaches into design, because it decides what a button does as well as
whether we build it. We keep it here because its test is whether we fund the filing step,
and the design principles settle how the button reads."
WEAK "These principles apply to everything we make." (names no product, no level and no kind of
decision) For the last field: "None." (gives no reason, so a reader cannot tell whether the
author tested each principle against the what-does-it-settle question or never looked)
TRAP A scope that drifts toward how a screen behaves. That has strayed into design principles,
which is the library's own judgement and no source states it. A close cousin is a scope that
names a single project, which makes this a project's principles, and those belong in that
project's PRD. A last slip is restating the current strategy's choices as scope: the set
outlives any one strategy. -->
**Product or products covered:** {{product_covered}}
**Level of the set (company, product line or product) and who applies it:** {{set_level_and_users}}
**Decisions this set settles, and decisions it does not:** {{decisions_scope}}
**Relation to the product vision, the product strategy and the design principles, and whether the set outlasts the current strategy:** {{neighbour_relations}}
**Principles that also settle how a screen behaves:** {{straddling_principles}}
## The Principles
<!-- WHAT Three to five principles. Each carries five fields: a statement, why we hold it, the trade-off
it accepts, what it rules out, and a decision it settled. This section is the document. Copy
the block below to add a fourth or fifth principle.
WHY Cagan's definition makes principles the thing meant to inform the many product decisions a
team makes, so each has to be specific enough to inform one. The three to five figure is
advice, not a finding. Gazit writes "No more than three to five principles are needed."
Slack caps a set at "no more than seven", ThoughtSpot "landed on seven" after debating the
count, and Intercom holds fourteen in layers. Of the eight published sets tabulated, only
Slack's and PostHog's fall inside three to five. None of the sources cites a measurement.
STATEMENT: every set read carries a name and a statement. WHY WE HOLD IT: Cagan writes "The
product team needs to understand these principles and the reasoning behind each one."
GitLab and Monzo give a reasoning paragraph. THE TRADE-OFF IT ACCEPTS: GitLab writes three of eight in an "X over Y" form, such
as "We value quality over velocity". Linear names a sequence in "Simple first, then
powerful". Box, as Pendo reports it, published a named cost with each principle, such as "The
tradeoff? Expense." Productboard's prompt has a field headed "TENSION". Writing the cost into
the principle keeps it attached when someone quotes the principle, which is this library's
reading of why the form matters, not a finding. WHAT IT RULES OUT: Productboard's prompt has a
field headed "WHAT IT MEANS WE WON'T DO", and it is a vendor's AI prompt, not a published set.
Langfuse's handbook has a list headed "What we don't do", as one list for the whole set. No
published set read carries the field per principle, so asking for it is this library's
position. A DECISION IT SETTLED: Monzo follows two principles with a case study, such as "Case
Study: Refunding customers during the upgrade". Slack's "How it shows up:" gives a product
example. Asking for one real case on every principle is this library's addition.
THE TRUISM TEST, AND THE DISSENT. Intercom writes "no principle is a truism" and "The opposite
of a principle should itself be a principle". AWS, writing about tenets,
says a tenet needs "an opposite which could be acceptable to an organisation". Productboard's
prompt says a principle without a reasonable opposite is "not a principle". airfocus holds a
different view, that principles "should trigger an emotional response". This template follows
Intercom, AWS and Productboard. The library's reasoning is that a reader can apply the
opposite test and cannot apply an emotional response. Distrusting a principle that no proposed
decision has ever strained is this library's inference from the opposite test.
Deep dive: product-principles_companion.md section 3 (Anatomy > The Principles) and section 6
(Debates and contested boundaries).
ASK Which decision has the team argued about, which option did this principle choose, and could a
reasonable team choose the opposite? What does the principle cost, and who bears that cost?
What can the team no longer do once it adopts the principle, and is there a request someone
has already made that it turns down? Which real case did it settle, and what reasoning would
a new product manager need to hold it?
GOOD Principle 1: "Freelancers before accountants." Statement: "When a request would make Kestrel
Ledger faster for a freelancer doing their own books but slower to set up for a bookkeeper who
manages many clients, we build for the freelancer." Why we hold it: "In the invented 2024
customer survey, most paying accounts were sole traders who do their own books, and the
accountants who tried the product left in the first month for tools with batch features."
The trade-off it accepts: "Bookkeeping firms will choose a rival with batch tools, and our
multi-client revenue stays small." What it rules out: "Multi-client workspaces, approval
chains between a bookkeeper and a client, and bulk import wizards for a firm's client list.
A partner firm with sixty clients asked for the first of these." A decision it settled: "In
2025 the partner firm asked for a multi-client switcher. We turned it down, and shipped a
shared-folder export a freelancer can send to their own bookkeeper."
WEAK Principle 1: "Put the customer first." Statement: "We always put our customers first." Why we
hold it: "Customers matter." The trade-off it accepts: "None." What it rules out: "Bad
experiences." A decision it settled: "Many." (a truism: no reasonable team would choose the
opposite, so it rules nothing out, it names no cost, and no field names a real decision or a
real reason)
TRAP A truism, which cannot settle an argument however well it lands. A close cousin is a principle
with no stated cost, which cannot show what it gives up. A third is a feature or a release
plan dressed as a principle, or a principle that restates one period's strategy and will go
stale when the strategy does. That last caution is this library's, not a sourced one. A last
slip is copying a principle from another company's set, which may not pass the team's own
opposite test. -->
**Principle 1: {{principle_1_name}}**
- **Statement:** {{principle_1_statement}}
- **Why we hold it:** {{principle_1_why}}
- **The trade-off it accepts:** {{principle_1_tradeoff}}
- **What it rules out:** {{principle_1_rules_out}}
- **A decision it settled:** {{principle_1_decision_settled}}
**Principle 2: {{principle_2_name}}**
- **Statement:** {{principle_2_statement}}
- **Why we hold it:** {{principle_2_why}}
- **The trade-off it accepts:** {{principle_2_tradeoff}}
- **What it rules out:** {{principle_2_rules_out}}
- **A decision it settled:** {{principle_2_decision_settled}}
**Principle 3: {{principle_3_name}}**
- **Statement:** {{principle_3_statement}}
- **Why we hold it:** {{principle_3_why}}
- **The trade-off it accepts:** {{principle_3_tradeoff}}
- **What it rules out:** {{principle_3_rules_out}}
- **A decision it settled:** {{principle_3_decision_settled}}
## When Principles Conflict
<!-- WHAT A table that says which principle wins when two pull apart, and in which kind of decision.
WHY AWS expects tensions between tenets and says they need judgement. ProductPlan writes "The
principles should clearly state what is most important." Productboard's prompt has a set-level
section headed "When Principles Conflict", though the research cannot say what that section
demands beyond its heading. A set that never says which one wins leaves the argument to
whoever is in the room. That last sentence is this library's inference, not a sourced claim.
THIS SECTION TAKES A SIDE IN A LIVE DISPUTE, AND THE SIDE IS THIS LIBRARY'S POSITION.
ThoughtSpot stack-ranked its candidates "within each category from most to least important",
and the same post says "it didn’t matter if some principles contradicted others". Nielsen
Norman Group, writing about design principles, takes a middle position: principles should "Not conflict with
one another", and where two matter, "then clearly specify in which contexts one of these
principles is more important than the other". No source read opposes ordering outright, and
none gives a worked rule for two principles that conflict. This template asks the author to
say which principle wins in a named kind of decision. That is closest to the NN/g position,
which is a method analogue for design principles and not product evidence, and it is the
library's choice. A resolution can take other forms. GitLab puts the priority inside the
principle with "X over Y", and Eriksson teaches the "even over" form. A ranked list is the
other shape. A resolution scoped to a kind of decision is less brittle than a global order,
and that reading is the library's.
Deep dive: product-principles_companion.md section 3 (Anatomy > When Principles Conflict) and
section 6 (Debates and contested boundaries).
ASK Which two principles are most likely to collide? In which kind of decision do they collide?
Which one wins there, and why? Is the answer different in another kind of decision?
PRIORITY Resolve the most likely collision first, and do not rank every pair. One resolved conflict
in a named kind of decision is worth more than a complete table nobody applies. That is this
library's advice, not a sourced rule.
ROW HINT A good row names the two principles, names the kind of decision, says which principle wins
there and gives the reason. A weak row says both matter equally.
GOOD | Freelancers before accountants, against File it for them before we explain it | A request
for a batch filing tool that files for many clients at once | Freelancers before accountants |
The filing step is in scope for one freelancer, and the batch tool serves a firm, so it waits
until a freelancer can file in two taps. |
WEAK | All of them | Everything | Use judgement | They are all important. | (no pair, no kind of
decision, no winner)
TRAP Saying that all principles are equally important, or ranking every pair. A close cousin is a
row whose winner is the principle that sounds better, with no reason given. -->
| Principles in tension | Kind of decision | Which wins | Why |
|---|---|---|---|
| {{conflict_principles}} | {{conflict_decision_kind}} | {{conflict_winner}} | {{conflict_reason}} |
## Using Them in Decisions
<!-- WHAT Where the set is applied, such as product review, prioritisation, a request the team may decline
and onboarding, how a decision cites a principle, and how a new joiner meets the set.
WHY The sources describe sets in use, mostly in review settings. At Slack, product managers and
designers "reference them in weekly workshops where projects are shared at every stage", and
Slack advises "it’s critical to designate clear checkpoints for assessing your work against
the principles". Cagan reports that teams tell him principles are "the part of the strategic
context that they use most often in their daily discovery work". That is his report of what
teams tell him, not a measurement. The research found no measured effect of a set on
decisions. SAYING NO IS ASSERTED MORE THAN SHOWN. Pendo reports that Box does not pursue
a feature idea that fails its principles. This research found no named company source that
describes a principle declining a specific request. This template names saying no as a moment to record, which is the
library's position, and confidence that sets are used this way is medium-low. ONBOARDING IS A
USE. Intercom's new joiners get an onboarding pack that includes a first problem statement the
company calls an Intermission. HOW A DECISION CITES A PRINCIPLE IS THE LIBRARY'S POSITION.
ProductPlan names the ability to point back to a principle to justify a decision, but no
source read prescribes a citation form. The set is judged against by judgement, not mechanical
compliance. This library infers that a set no decision ever cites is decoration. That
inference is not a sourced finding.
Deep dive: product-principles_companion.md section 3 (Anatomy > Using Them in Decisions).
ASK In which two or three moments is a principle cited: a product review, a prioritisation
discussion, a request the team is about to turn down? How is the citation written down? How
does a new joiner meet the set, and who walks them through it?
GOOD "Applied in the monthly product review, when the product managers rank the backlog, and when
we are about to turn down a request from a customer or a partner. A decision record carries one
line naming the principle relied on and, if a principle was set aside, which one and why. A
reviewer may accept the reasoning without accepting the outcome, and says which. A new product
manager reads the set in their first week and brings one past decision they think it would
have changed."
WEAK "Product managers should keep the principles in mind." (names no moment and no form of
citation)
TRAP A set that is read at adoption and never cited again, or a citation so mechanical that it
becomes a checkbox. A close cousin is a claim that the set is used to say no, with no case. -->
**Where the set is applied:** {{where_applied}}
**How a decision cites a principle:** {{how_decisions_cite}}
**How a new joiner meets the set:** {{new_joiner_use}}
## Ownership and Changes
<!-- WHAT The named owner, how anyone proposes a change, who decides, and a dated change log.
WHY This is where the set's own staleness is handled. No source read names who owns a set after
it is written, or who approves a later change. LogRocket describes leadership signing off a
draft. The role names this template asks for are the library's position. Cagan's 2020 text has product leaders prepare the set. Intercom
describes organic change: its principles "evolve very slowly". THE AUTHORSHIP DEBATE
SITS BEHIND THIS FIELD, AND THIS TEMPLATE DOES NOT SETTLE IT. ProductPlan writes "defined
product principles are driven by the team itself". Product School says "Talk to your team,
not just leadership". LogRocket writes "A hybrid approach works better", in which leadership
presents a draft and sharpens it with the team. The template asks only that an owner be named
and that the way a change is proposed be stated. THE DATED RECORD IS SOURCED IN PRACTICE, NOT
IN METHOD. Intercom gives the one dated record of a set changing. In 2019 it said it used to
have five principles and "merged three of them into one". Slack's set differs by one principle
between its 2021 list and a later list, with no reason given in either source, so that is a
comparison and not a change record. Revision is
invited, and rarely: Kindel's tenets carry "unless you know better ones", and ProductPlan says
"An update should be an unusual and well-considered event, not a regular occurrence." So
record a retirement or a merge as a change and keep the reason. That advice is the library's.
Intercom's merge is the one sourced example.
Deep dive: product-principles_companion.md section 3 (Anatomy > Ownership and Changes).
ASK Who owns this set, by role and by name? Where does anyone propose a change? Who decides? When
was the set first agreed, and what has changed since, including anything merged or retired?
PRIORITY Start the log with the date the set was first agreed, so the first row is never blank.
ROW HINT A good row has a date, says what changed in a sentence a stranger could follow, and names
what fired it. A weak row is a date and the word "updated".
GOOD | 2025-11-04 | Principle 3 merged into principle 1: two decisions in a row were settled by
citing both, and nobody could say how they differed | Review Trigger: a valid exception keeps
recurring | Head of product |
WEAK | 2025-11-04 | Updated | | | (no change a stranger could follow, no trigger and no decider)
TRAP A set with no owner and no condition that reopens it. A change log that shows only additions
hides the retirements and merges. -->
**Owner (role and person):** {{owner_role_and_person}}
**How anyone proposes a change:** {{how_to_propose}}
**Who decides a change:** {{who_decides}}
| Date | What changed | What fired it | Who decided |
|---|---|---|---|
| {{change_date}} | {{change_summary}} | {{change_trigger}} | {{change_decider}} |
## Review Trigger
<!-- WHAT A table of conditions, owners and actions: what event makes this set wrong or incomplete, who
notices, and what they do about it. The conditions are things an owner can observe, not dates.
WHY The standing-standards family requires every member to carry this mechanism, because members
fail by going quietly out of date. No source read names an owner for a trigger. The owner
column is this library's, and the conditions are statements of intent, not logged events.
Where a condition has a source, it is named here. Productboard's AI prompt, a vendor's prompt,
lists three conditions for revisiting a set, and this template restates them in its own rows:
"The company strategy changes significantly", a principle "consistently violated one without
deciding to change it", and a new principle that "would have resolved 3+ debates in the last
quarter". The threshold of three or more debates is the vendor's figure and not a measured
one. A valid exception: Intercom says to "try and re-codify that back in or change a
principle, evolve it". The quotation carries no count, so "recurs" and its threshold of a
second occurrence are this library's reading. A principle does not take hold: Slack says to
observe how it is used and "tune it based on your findings". THE STATED
INTERVALS DISAGREE, AND THIS TEMPLATE TAKES NONE OF THEM. Guru says principles should "hold
their ground for at least a year". LogRocket says "Review your product principles every three
months", which is a review and not a rewrite. Intercom says "We also update them every 1-2
years". ProductPlan says updates should be unusual. None cites a measurement. No calendar
interval is required, and the foundation classification holds. A team that wants a review date
may add one as a floor and should label it as its own choice.
Deep dive: product-principles_companion.md section 3 (Anatomy > Review Trigger).
ASK What could the owner observe that shows this set is wrong? Who notices each condition, by
role? What do they do about it: reopen the set, run a short check against recent decisions, or
write the change into the log?
PRIORITY List the conditions an owner is most likely to see first, and write each as an event
someone can observe, not as a date. The strategy row is the visible link to the product
strategy, which this set outlives and does not restate.
ROW HINT A good row states a condition the owner could observe, names a role as owner, and states an
action. A weak row says "review regularly" and names no one.
GOOD | The strategy changes significantly | Head of product | Read each principle against the new
strategy within a month, keep those still sound, and write any merge, retirement or addition
into the change log with the reason. |
WEAK | Review regularly | The team | Update the principles. | (a condition nobody could observe, no
named role as owner and no action anyone could take)
TRAP Writing a date as the only trigger, or naming one condition and treating it as the whole
mechanism. Tie a row to a line in the change log when it fires. -->
| Condition | Owner | What They Do About It |
|---|---|---|
| The strategy changes significantly | {{strategy_change_owner}} | {{strategy_change_action}} |
| A principle is consistently violated without a decision to change it | {{violation_owner}} | {{violation_action}} |
| A new principle would have settled repeated debates | {{repeated_debate_owner}} | {{repeated_debate_action}} |
| A valid exception recurs | {{exception_owner}} | {{exception_action}} |
| A principle does not take hold | {{not_taking_hold_owner}} | {{not_taking_hold_action}} |

The reasoning, the history and every source, in the repository:

  • Companion - the long-form argument: why these sections, where the sources disagree, and what the bundle refuses to claim
  • History - what changed in this bundle, and when
  • Research log - every source consulted, with what each one actually supports
  • Catalog metadata - the machine-readable record this page is generated from

Catalog record: 6 sections across 1 format(s), methodology generic, typically owned by Product leader (VP or Head of Product).