diff --git a/exercises/04.conflicting-constraints/01.problem.map-the-tradeoffs/README.mdx b/exercises/04.conflicting-constraints/01.problem.map-the-tradeoffs/README.mdx
new file mode 100644
index 0000000..b2ef3bb
--- /dev/null
+++ b/exercises/04.conflicting-constraints/01.problem.map-the-tradeoffs/README.mdx
@@ -0,0 +1,48 @@
+# Map the Tradeoffs
+
+👨💼 The narrow host-login plan from Exercise 3 solved one problem: returning
+hosts can get back to schedules more reliably. But the broader pressure has not
+gone away.
+
+Some people now want a wider accounts system because it sounds more
+trustworthy, more professional, and more scalable for repeat usage. Others want
+to preserve the product's low-friction advantage and avoid dragging account
+complexity into flows that currently work.
+
+🧝♀️ Set the playground to `extra/03.improved-implementation`. Use the planning
+artifacts you have already produced as context:
+
+-
+-
+-
+-
+
+🐨 Create a dedicated
+artifact that compares the competing directions in front of the team.
+
+At minimum, compare:
+
+1. The narrow host-focused path
+2. A broader account/login expansion for repeat users
+
+Your map should cover:
+
+1. The priorities each path is trying to optimize
+2. The success criteria each path would claim as justification
+3. The likely harms or regression risks each path creates
+4. The assumptions or evidence gaps hidden inside each argument
+5. The mitigations that could reduce downside, if that path were chosen
+
+Requirements:
+
+- Keep the comparison grounded in the app's real flows on `/`,
+ `/s/{scheduleKey}`, and `/s/{scheduleKey}/{hostKey}`.
+- Include finalized-plan rate, create-flow speed, attendee completion, trust,
+ and implementation/support cost somewhere in your analysis.
+- Be explicit about which constraints are truly competing rather than merely
+ inconvenient.
+- Do not collapse everything into one "balanced" answer yet. Surface the
+ tension first.
+
+💰 Good tradeoff work starts by making the conflict legible. If both options
+sound equally good in your document, you have probably hidden the real cost.
diff --git a/exercises/04.conflicting-constraints/01.problem.map-the-tradeoffs/tradeoff-map.md b/exercises/04.conflicting-constraints/01.problem.map-the-tradeoffs/tradeoff-map.md
new file mode 100644
index 0000000..7ee5bec
--- /dev/null
+++ b/exercises/04.conflicting-constraints/01.problem.map-the-tradeoffs/tradeoff-map.md
@@ -0,0 +1,43 @@
+# Tradeoff Map
+
+This document should make the competing constraints visible before the team
+pretends the decision is obvious.
+
+## Decision context
+
+- What new pressure or request is creating this conflict:
+- Which earlier decision or scope boundary is now under pressure:
+- Which user outcomes are most at risk if the team gets this wrong:
+
+## Option A
+
+- Path name:
+- What this path is trying to optimize:
+- Success criteria this path would claim:
+- Which flows or routes it touches most:
+- Likely harms or regressions it introduces:
+- Key assumptions or missing evidence:
+- Mitigations if this path is chosen:
+
+## Option B
+
+- Path name:
+- What this path is trying to optimize:
+- Success criteria this path would claim:
+- Which flows or routes it touches most:
+- Likely harms or regressions it introduces:
+- Key assumptions or missing evidence:
+- Mitigations if this path is chosen:
+
+## Constraint collisions
+
+- Which priorities are directly in tension:
+- Which route or workflow feels most fragile under this decision:
+- Which costs are immediate versus delayed:
+- Which downside is easiest for a team to underestimate:
+
+## Open questions
+
+- What evidence would most reduce uncertainty:
+- What would make the broader path clearly worth it:
+- What would make the narrower path clearly insufficient:
diff --git a/exercises/04.conflicting-constraints/01.solution.map-the-tradeoffs/README.mdx b/exercises/04.conflicting-constraints/01.solution.map-the-tradeoffs/README.mdx
new file mode 100644
index 0000000..f65720c
--- /dev/null
+++ b/exercises/04.conflicting-constraints/01.solution.map-the-tradeoffs/README.mdx
@@ -0,0 +1,12 @@
+# Map the Tradeoffs
+
+👨💼 Nice work. You made the conflict explicit instead of letting the team talk
+about "better trust" and "low friction" as if they were automatically
+compatible.
+
+🦉 This is the judgment move this exercise is training. Strong engineers do not
+hide the cost of a more ambitious path, and they do not romanticize simplicity
+without naming what it fails to solve. They force both sides of the tradeoff to
+be comparable.
+
+In the next step, you'll stop mapping and start deciding.
diff --git a/exercises/04.conflicting-constraints/01.solution.map-the-tradeoffs/tradeoff-map.md b/exercises/04.conflicting-constraints/01.solution.map-the-tradeoffs/tradeoff-map.md
new file mode 100644
index 0000000..a866781
--- /dev/null
+++ b/exercises/04.conflicting-constraints/01.solution.map-the-tradeoffs/tradeoff-map.md
@@ -0,0 +1,126 @@
+# Tradeoff Map
+
+This document makes the competing constraints visible before the team pretends
+the decision is obvious.
+
+## Decision context
+
+- What new pressure or request is creating this conflict:
+ - The team can now justify a narrow host-login improvement, but stakeholders
+ are pushing to go further and add broader accounts for repeat users.
+- Which earlier decision or scope boundary is now under pressure:
+ - The exercise 2 decision to avoid a full account system and the exercise 3
+ plan to keep login narrowly host-focused.
+- Which user outcomes are most at risk if the team gets this wrong:
+ - Finalized-plan rate
+ - Create-flow completion speed
+ - Attendee submission completion
+ - Host trust that they can reliably get back to the right schedule
+
+## Option A
+
+- Path name:
+ - Stay narrow: host-focused continuity improvement only
+- What this path is trying to optimize:
+ - Better return access for hosts
+ - Lower risk to the create and attendee flows
+ - Smaller implementation and support surface area
+- Success criteria this path would claim:
+ - Returning hosts recover schedules more reliably
+ - Finalized-plan rate holds steady or improves
+ - Create and attendee completion remain effectively unchanged
+- Which flows or routes it touches most:
+ - `/s/{scheduleKey}/{hostKey}` for host management and claiming
+ - A host-only login entry point
+ - Optional post-create continuity prompts on `/`
+- Likely harms or regressions it introduces:
+ - Some repeat users may still feel the product lacks the polish or permanence
+ of a full account system
+ - Hosts may still need to understand both host-link and account-based access
+ - The narrower move may not fully solve long-term repeat-usage expectations
+- Key assumptions or missing evidence:
+ - The real pain is primarily host return access, not universal identity
+ - Most attendee value still comes from fast, no-account participation
+ - The product does not yet need cross-schedule history badly enough to justify
+ broader scope
+- Mitigations if this path is chosen:
+ - Make host-link purpose and claiming behavior clearer in the UI
+ - Keep login optional and host-only
+ - Measure host-return success and revisit if the narrow path proves
+ insufficient
+
+## Option B
+
+- Path name:
+ - Expand to broader repeat-user accounts
+- What this path is trying to optimize:
+ - Stronger sense of trust and permanence
+ - Easier schedule recovery across multiple visits
+ - Future optionality for identity-based features
+- Success criteria this path would claim:
+ - More returning users can find schedules without saved links
+ - Repeat usage feels more durable and professional
+ - The product gains a clearer long-term identity model
+- Which flows or routes it touches most:
+ - `/` because account creation or login prompts tend to leak into creation
+ - `/s/{scheduleKey}` because attendee identity expectations become harder to
+ avoid
+ - `/s/{scheduleKey}/{hostKey}` because host-link access now competes with
+ authenticated ownership
+ - New auth, recovery, and account-management routes
+- Likely harms or regressions it introduces:
+ - Create-flow friction rises as account concepts compete with quick schedule
+ setup
+ - Attendee completion may drop if the public flow feels more account-shaped or
+ less obviously guest-friendly
+ - Support and implementation cost grow materially because the team now owns
+ auth, recovery, identity state, and confusion between access paths
+ - Route purpose becomes easier to blur across public, host-only, and account
+ surfaces
+- Key assumptions or missing evidence:
+ - More identity always equals more trust in this product category
+ - Repeat users actually want accounts enough to offset the added friction
+ - The team can add broader auth without harming mobile-friendly participation
+ - Future optionality is valuable enough right now to justify present cost
+- Mitigations if this path is chosen:
+ - Keep attendee participation explicitly guest-friendly
+ - Preserve host-link fallback during rollout
+ - Gate account prompts away from the first-create and first-respond moments
+ - Add instrumentation and rollback triggers before broad exposure
+
+## Constraint collisions
+
+- Which priorities are directly in tension:
+ - Host-return reliability and future trust signals versus low-friction create
+ and attendee completion
+ - Product permanence versus route clarity and workflow simplicity
+ - Future optionality versus current implementation and support cost
+- Which route or workflow feels most fragile under this decision:
+ - `/s/{scheduleKey}` is the most fragile because its value comes from being
+ obviously lightweight and guest-oriented
+- Which costs are immediate versus delayed:
+ - Immediate costs: auth complexity, support burden, route confusion, create
+ friction
+ - Delayed benefits: stronger repeat-user continuity and future identity-based
+ product expansion
+- Which downside is easiest for a team to underestimate:
+ - Leakage of account expectations into flows that currently succeed precisely
+ because they are simple and no-account
+
+## Open questions
+
+- What evidence would most reduce uncertainty:
+ - How often hosts fail to return without the private host link
+ - Whether attendee completion drops when identity language is introduced
+ - Whether repeat users in this social-planning context actually expect account
+ persistence
+- What would make the broader path clearly worth it:
+ - Repeated evidence that host-return problems materially hurt finalized-plan
+ rate and cannot be solved by a host-only path
+ - Clear demand for cross-schedule history or identity-based recovery from a
+ meaningful share of active users
+- What would make the narrower path clearly insufficient:
+ - Hosts still fail to re-find schedules even after a careful host-only
+ continuity improvement
+ - The product proves that repeat-user continuity matters more than
+ ultra-lightweight first-use participation
diff --git a/exercises/04.conflicting-constraints/02.problem.make-the-call/README.mdx b/exercises/04.conflicting-constraints/02.problem.make-the-call/README.mdx
new file mode 100644
index 0000000..41ac63a
--- /dev/null
+++ b/exercises/04.conflicting-constraints/02.problem.make-the-call/README.mdx
@@ -0,0 +1,34 @@
+# Make the Call
+
+👨💼 You have mapped the conflict. Now the team needs a decision, not another
+round of vague discussion.
+
+🧝♀️ Use your existing to
+create a dedicated
+artifact.
+
+This is the moment to choose which tradeoff the team should make now and to
+justify it clearly enough that other engineers, product partners, and future
+you can understand why.
+
+Your decision brief should cover:
+
+1. Which path should ship now
+2. Why that tradeoff is correct at the current product stage
+3. The success criteria that would prove the decision was right
+4. The downside you are knowingly accepting
+5. The mitigations required to keep the chosen path safe
+6. What stays explicitly deferred and what evidence would reopen it later
+7. The rollback boundary if the decision harms the product
+
+Requirements:
+
+- Use the competing constraints from your tradeoff map rather than re-framing
+ the problem from scratch.
+- Name the constraint you are prioritizing most and the constraint you are
+ willing to disappoint.
+- Keep the reasoning grounded in the scheduler's real routes and flows.
+- Do not defend the decision with vague future optionality alone.
+
+💰 Tradeoff articulation is not "everyone is right." It is "this cost is worth
+paying now, and this other cost is not."
diff --git a/exercises/04.conflicting-constraints/02.problem.make-the-call/tradeoff-decision.md b/exercises/04.conflicting-constraints/02.problem.make-the-call/tradeoff-decision.md
new file mode 100644
index 0000000..410c63a
--- /dev/null
+++ b/exercises/04.conflicting-constraints/02.problem.make-the-call/tradeoff-decision.md
@@ -0,0 +1,35 @@
+# Tradeoff Decision
+
+This brief should turn the tradeoff map into a clear recommendation the team can
+act on.
+
+## Decision
+
+- Which path should ship now:
+- Why this is the right tradeoff at the current product stage:
+- Which constraint are we prioritizing most:
+- Which constraint are we knowingly not maximizing:
+
+## Success criteria
+
+- What must improve if this decision is right:
+- What must not regress while we pursue it:
+- Which indicators or behaviors would give us confidence:
+
+## Accepted downside
+
+- What cost are we deliberately accepting:
+- Why that cost is preferable to the alternative right now:
+- Which users or workflows are least well served by this decision:
+
+## Mitigations
+
+- How we reduce the main downside of the chosen path:
+- Which guardrails or rollout choices are required:
+- What fallback or rollback boundary we need:
+
+## Deferred path
+
+- What remains deferred for now:
+- What evidence would reopen that path later:
+- What would make us regret this decision:
diff --git a/exercises/04.conflicting-constraints/02.problem.make-the-call/tradeoff-map.md b/exercises/04.conflicting-constraints/02.problem.make-the-call/tradeoff-map.md
new file mode 100644
index 0000000..a866781
--- /dev/null
+++ b/exercises/04.conflicting-constraints/02.problem.make-the-call/tradeoff-map.md
@@ -0,0 +1,126 @@
+# Tradeoff Map
+
+This document makes the competing constraints visible before the team pretends
+the decision is obvious.
+
+## Decision context
+
+- What new pressure or request is creating this conflict:
+ - The team can now justify a narrow host-login improvement, but stakeholders
+ are pushing to go further and add broader accounts for repeat users.
+- Which earlier decision or scope boundary is now under pressure:
+ - The exercise 2 decision to avoid a full account system and the exercise 3
+ plan to keep login narrowly host-focused.
+- Which user outcomes are most at risk if the team gets this wrong:
+ - Finalized-plan rate
+ - Create-flow completion speed
+ - Attendee submission completion
+ - Host trust that they can reliably get back to the right schedule
+
+## Option A
+
+- Path name:
+ - Stay narrow: host-focused continuity improvement only
+- What this path is trying to optimize:
+ - Better return access for hosts
+ - Lower risk to the create and attendee flows
+ - Smaller implementation and support surface area
+- Success criteria this path would claim:
+ - Returning hosts recover schedules more reliably
+ - Finalized-plan rate holds steady or improves
+ - Create and attendee completion remain effectively unchanged
+- Which flows or routes it touches most:
+ - `/s/{scheduleKey}/{hostKey}` for host management and claiming
+ - A host-only login entry point
+ - Optional post-create continuity prompts on `/`
+- Likely harms or regressions it introduces:
+ - Some repeat users may still feel the product lacks the polish or permanence
+ of a full account system
+ - Hosts may still need to understand both host-link and account-based access
+ - The narrower move may not fully solve long-term repeat-usage expectations
+- Key assumptions or missing evidence:
+ - The real pain is primarily host return access, not universal identity
+ - Most attendee value still comes from fast, no-account participation
+ - The product does not yet need cross-schedule history badly enough to justify
+ broader scope
+- Mitigations if this path is chosen:
+ - Make host-link purpose and claiming behavior clearer in the UI
+ - Keep login optional and host-only
+ - Measure host-return success and revisit if the narrow path proves
+ insufficient
+
+## Option B
+
+- Path name:
+ - Expand to broader repeat-user accounts
+- What this path is trying to optimize:
+ - Stronger sense of trust and permanence
+ - Easier schedule recovery across multiple visits
+ - Future optionality for identity-based features
+- Success criteria this path would claim:
+ - More returning users can find schedules without saved links
+ - Repeat usage feels more durable and professional
+ - The product gains a clearer long-term identity model
+- Which flows or routes it touches most:
+ - `/` because account creation or login prompts tend to leak into creation
+ - `/s/{scheduleKey}` because attendee identity expectations become harder to
+ avoid
+ - `/s/{scheduleKey}/{hostKey}` because host-link access now competes with
+ authenticated ownership
+ - New auth, recovery, and account-management routes
+- Likely harms or regressions it introduces:
+ - Create-flow friction rises as account concepts compete with quick schedule
+ setup
+ - Attendee completion may drop if the public flow feels more account-shaped or
+ less obviously guest-friendly
+ - Support and implementation cost grow materially because the team now owns
+ auth, recovery, identity state, and confusion between access paths
+ - Route purpose becomes easier to blur across public, host-only, and account
+ surfaces
+- Key assumptions or missing evidence:
+ - More identity always equals more trust in this product category
+ - Repeat users actually want accounts enough to offset the added friction
+ - The team can add broader auth without harming mobile-friendly participation
+ - Future optionality is valuable enough right now to justify present cost
+- Mitigations if this path is chosen:
+ - Keep attendee participation explicitly guest-friendly
+ - Preserve host-link fallback during rollout
+ - Gate account prompts away from the first-create and first-respond moments
+ - Add instrumentation and rollback triggers before broad exposure
+
+## Constraint collisions
+
+- Which priorities are directly in tension:
+ - Host-return reliability and future trust signals versus low-friction create
+ and attendee completion
+ - Product permanence versus route clarity and workflow simplicity
+ - Future optionality versus current implementation and support cost
+- Which route or workflow feels most fragile under this decision:
+ - `/s/{scheduleKey}` is the most fragile because its value comes from being
+ obviously lightweight and guest-oriented
+- Which costs are immediate versus delayed:
+ - Immediate costs: auth complexity, support burden, route confusion, create
+ friction
+ - Delayed benefits: stronger repeat-user continuity and future identity-based
+ product expansion
+- Which downside is easiest for a team to underestimate:
+ - Leakage of account expectations into flows that currently succeed precisely
+ because they are simple and no-account
+
+## Open questions
+
+- What evidence would most reduce uncertainty:
+ - How often hosts fail to return without the private host link
+ - Whether attendee completion drops when identity language is introduced
+ - Whether repeat users in this social-planning context actually expect account
+ persistence
+- What would make the broader path clearly worth it:
+ - Repeated evidence that host-return problems materially hurt finalized-plan
+ rate and cannot be solved by a host-only path
+ - Clear demand for cross-schedule history or identity-based recovery from a
+ meaningful share of active users
+- What would make the narrower path clearly insufficient:
+ - Hosts still fail to re-find schedules even after a careful host-only
+ continuity improvement
+ - The product proves that repeat-user continuity matters more than
+ ultra-lightweight first-use participation
diff --git a/exercises/04.conflicting-constraints/02.solution.make-the-call/README.mdx b/exercises/04.conflicting-constraints/02.solution.make-the-call/README.mdx
new file mode 100644
index 0000000..a45ce00
--- /dev/null
+++ b/exercises/04.conflicting-constraints/02.solution.make-the-call/README.mdx
@@ -0,0 +1,12 @@
+# Make the Call
+
+👨💼 Nice work. You chose deliberately instead of letting the team drift toward
+the biggest-sounding solution by default.
+
+🦉 This is where engineering judgment becomes legible. A good decision memo does
+not claim the chosen path has no downside. It explains which downside is worth
+accepting now, which one is too dangerous, and how the team will know if that
+logic stops being true.
+
+In the next step, you'll pressure-test the decision so it can survive future
+pushback instead of collapsing into vague compromise.
diff --git a/exercises/04.conflicting-constraints/02.solution.make-the-call/tradeoff-decision.md b/exercises/04.conflicting-constraints/02.solution.make-the-call/tradeoff-decision.md
new file mode 100644
index 0000000..e292376
--- /dev/null
+++ b/exercises/04.conflicting-constraints/02.solution.make-the-call/tradeoff-decision.md
@@ -0,0 +1,80 @@
+# Tradeoff Decision
+
+This brief turns the tradeoff map into a clear recommendation the team can act
+on.
+
+## Decision
+
+- Which path should ship now:
+ - Ship the narrow host-focused continuity path now and keep broader
+ repeat-user accounts deferred.
+- Why this is the right tradeoff at the current product stage:
+ - The product still wins on low-friction scheduling. The most credible problem
+ we have is host return reliability, not universal identity. A broader
+ account move would charge complexity across all major flows before the
+ product has earned that cost.
+- Which constraint are we prioritizing most:
+ - Protect create-flow speed and attendee completion while improving host
+ return reliability enough to help finalized-plan rate.
+- Which constraint are we knowingly not maximizing:
+ - We are not maximizing long-term identity optionality or the polish signal of
+ a full account model yet.
+
+## Success criteria
+
+- What must improve if this decision is right:
+ - Returning hosts re-find and manage schedules more reliably than before.
+- What must not regress while we pursue it:
+ - Schedule creation remains lightweight and skippable past any login prompts.
+ - Attendee participation on `/s/{scheduleKey}` remains obviously
+ guest-friendly and fast to complete.
+ - Finalized-plan rate does not fall because the access model becomes more
+ confusing.
+- Which indicators or behaviors would give us confidence:
+ - Host-return success rises.
+ - Claim or login completion for hosts is healthy without displacing host-link
+ fallback usage.
+ - Create abandonment and attendee drop-off remain flat.
+
+## Accepted downside
+
+- What cost are we deliberately accepting:
+ - Some repeat users may still feel the product lacks the permanence and
+ convenience of a broader account system.
+- Why that cost is preferable to the alternative right now:
+ - It is safer to under-build identity than to over-build it into the create
+ and attendee flows that currently make the product work.
+- Which users or workflows are least well served by this decision:
+ - Power users who want cross-schedule history or universal schedule recovery.
+
+## Mitigations
+
+- How we reduce the main downside of the chosen path:
+ - Improve host-link clarity, make claiming easy, and ensure returning hosts
+ can get back to the right schedule without generic account chrome.
+- Which guardrails or rollout choices are required:
+ - Keep login optional and host-only.
+ - Preserve host-link fallback.
+ - Keep account concepts out of `/s/{scheduleKey}`.
+ - Make any post-create continuity prompt secondary and easy to dismiss.
+- What fallback or rollback boundary we need:
+ - If create completion, attendee completion, or finalized-plan rate worsens
+ after rollout, remove the host-login prompt first and fall back to the
+ host-link model while keeping any claimed ownership data intact.
+
+## Deferred path
+
+- What remains deferred for now:
+ - Participant accounts
+ - Cross-schedule history for all users
+ - Broader account-management surfaces
+ - Identity-based product expansion justified mainly by future optionality
+- What evidence would reopen that path later:
+ - The narrow host path fails to solve return reliability.
+ - Repeat-user continuity becomes a demonstrated product need across a
+ meaningful share of usage.
+ - Broader identity can be introduced without materially harming guest
+ participation.
+- What would make us regret this decision:
+ - Discovering that the product's next stage depends far more on repeat-user
+ continuity than on frictionless first-time participation.
diff --git a/exercises/04.conflicting-constraints/02.solution.make-the-call/tradeoff-map.md b/exercises/04.conflicting-constraints/02.solution.make-the-call/tradeoff-map.md
new file mode 100644
index 0000000..a866781
--- /dev/null
+++ b/exercises/04.conflicting-constraints/02.solution.make-the-call/tradeoff-map.md
@@ -0,0 +1,126 @@
+# Tradeoff Map
+
+This document makes the competing constraints visible before the team pretends
+the decision is obvious.
+
+## Decision context
+
+- What new pressure or request is creating this conflict:
+ - The team can now justify a narrow host-login improvement, but stakeholders
+ are pushing to go further and add broader accounts for repeat users.
+- Which earlier decision or scope boundary is now under pressure:
+ - The exercise 2 decision to avoid a full account system and the exercise 3
+ plan to keep login narrowly host-focused.
+- Which user outcomes are most at risk if the team gets this wrong:
+ - Finalized-plan rate
+ - Create-flow completion speed
+ - Attendee submission completion
+ - Host trust that they can reliably get back to the right schedule
+
+## Option A
+
+- Path name:
+ - Stay narrow: host-focused continuity improvement only
+- What this path is trying to optimize:
+ - Better return access for hosts
+ - Lower risk to the create and attendee flows
+ - Smaller implementation and support surface area
+- Success criteria this path would claim:
+ - Returning hosts recover schedules more reliably
+ - Finalized-plan rate holds steady or improves
+ - Create and attendee completion remain effectively unchanged
+- Which flows or routes it touches most:
+ - `/s/{scheduleKey}/{hostKey}` for host management and claiming
+ - A host-only login entry point
+ - Optional post-create continuity prompts on `/`
+- Likely harms or regressions it introduces:
+ - Some repeat users may still feel the product lacks the polish or permanence
+ of a full account system
+ - Hosts may still need to understand both host-link and account-based access
+ - The narrower move may not fully solve long-term repeat-usage expectations
+- Key assumptions or missing evidence:
+ - The real pain is primarily host return access, not universal identity
+ - Most attendee value still comes from fast, no-account participation
+ - The product does not yet need cross-schedule history badly enough to justify
+ broader scope
+- Mitigations if this path is chosen:
+ - Make host-link purpose and claiming behavior clearer in the UI
+ - Keep login optional and host-only
+ - Measure host-return success and revisit if the narrow path proves
+ insufficient
+
+## Option B
+
+- Path name:
+ - Expand to broader repeat-user accounts
+- What this path is trying to optimize:
+ - Stronger sense of trust and permanence
+ - Easier schedule recovery across multiple visits
+ - Future optionality for identity-based features
+- Success criteria this path would claim:
+ - More returning users can find schedules without saved links
+ - Repeat usage feels more durable and professional
+ - The product gains a clearer long-term identity model
+- Which flows or routes it touches most:
+ - `/` because account creation or login prompts tend to leak into creation
+ - `/s/{scheduleKey}` because attendee identity expectations become harder to
+ avoid
+ - `/s/{scheduleKey}/{hostKey}` because host-link access now competes with
+ authenticated ownership
+ - New auth, recovery, and account-management routes
+- Likely harms or regressions it introduces:
+ - Create-flow friction rises as account concepts compete with quick schedule
+ setup
+ - Attendee completion may drop if the public flow feels more account-shaped or
+ less obviously guest-friendly
+ - Support and implementation cost grow materially because the team now owns
+ auth, recovery, identity state, and confusion between access paths
+ - Route purpose becomes easier to blur across public, host-only, and account
+ surfaces
+- Key assumptions or missing evidence:
+ - More identity always equals more trust in this product category
+ - Repeat users actually want accounts enough to offset the added friction
+ - The team can add broader auth without harming mobile-friendly participation
+ - Future optionality is valuable enough right now to justify present cost
+- Mitigations if this path is chosen:
+ - Keep attendee participation explicitly guest-friendly
+ - Preserve host-link fallback during rollout
+ - Gate account prompts away from the first-create and first-respond moments
+ - Add instrumentation and rollback triggers before broad exposure
+
+## Constraint collisions
+
+- Which priorities are directly in tension:
+ - Host-return reliability and future trust signals versus low-friction create
+ and attendee completion
+ - Product permanence versus route clarity and workflow simplicity
+ - Future optionality versus current implementation and support cost
+- Which route or workflow feels most fragile under this decision:
+ - `/s/{scheduleKey}` is the most fragile because its value comes from being
+ obviously lightweight and guest-oriented
+- Which costs are immediate versus delayed:
+ - Immediate costs: auth complexity, support burden, route confusion, create
+ friction
+ - Delayed benefits: stronger repeat-user continuity and future identity-based
+ product expansion
+- Which downside is easiest for a team to underestimate:
+ - Leakage of account expectations into flows that currently succeed precisely
+ because they are simple and no-account
+
+## Open questions
+
+- What evidence would most reduce uncertainty:
+ - How often hosts fail to return without the private host link
+ - Whether attendee completion drops when identity language is introduced
+ - Whether repeat users in this social-planning context actually expect account
+ persistence
+- What would make the broader path clearly worth it:
+ - Repeated evidence that host-return problems materially hurt finalized-plan
+ rate and cannot be solved by a host-only path
+ - Clear demand for cross-schedule history or identity-based recovery from a
+ meaningful share of active users
+- What would make the narrower path clearly insufficient:
+ - Hosts still fail to re-find schedules even after a careful host-only
+ continuity improvement
+ - The product proves that repeat-user continuity matters more than
+ ultra-lightweight first-use participation
diff --git a/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/README.mdx b/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/README.mdx
new file mode 100644
index 0000000..0f38a6a
--- /dev/null
+++ b/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/README.mdx
@@ -0,0 +1,36 @@
+# Pressure-Test the Decision
+
+👨💼 The real test of a tradeoff is whether it survives pressure after the
+meeting. Stakeholders will keep asking why the team is not simply doing the
+bigger thing now.
+
+🧝♀️ Use your existing
+
+-
+-
+
+to create a dedicated
+artifact.
+
+This review should pressure-test the decision you just made. Treat it like a
+future-facing note to the team explaining how the chosen path could still fail,
+what protections it needs, and what evidence would force a revisit.
+
+Your review should cover:
+
+1. The main ways the chosen tradeoff could still hurt the product
+2. The signals that would tell you the team picked the wrong side of the
+ constraint
+3. The mitigations or rollout conditions that are non-negotiable
+4. What would trigger a revision, narrowing, or rollback
+5. What future evidence would justify reopening the deferred path
+
+Requirements:
+
+- Keep the pressure test anchored in the real user flows and route boundaries.
+- Name at least one failure mode for create, attendee response, and host return.
+- Be explicit about what you would change first if the decision starts going bad.
+- Do not restate the decision memo. Stress it until you can see where it breaks.
+
+💰 Judgment gets better when decisions can survive contact with future evidence,
+not just the confidence of the person who wrote the first brief.
diff --git a/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/tradeoff-decision.md b/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/tradeoff-decision.md
new file mode 100644
index 0000000..e292376
--- /dev/null
+++ b/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/tradeoff-decision.md
@@ -0,0 +1,80 @@
+# Tradeoff Decision
+
+This brief turns the tradeoff map into a clear recommendation the team can act
+on.
+
+## Decision
+
+- Which path should ship now:
+ - Ship the narrow host-focused continuity path now and keep broader
+ repeat-user accounts deferred.
+- Why this is the right tradeoff at the current product stage:
+ - The product still wins on low-friction scheduling. The most credible problem
+ we have is host return reliability, not universal identity. A broader
+ account move would charge complexity across all major flows before the
+ product has earned that cost.
+- Which constraint are we prioritizing most:
+ - Protect create-flow speed and attendee completion while improving host
+ return reliability enough to help finalized-plan rate.
+- Which constraint are we knowingly not maximizing:
+ - We are not maximizing long-term identity optionality or the polish signal of
+ a full account model yet.
+
+## Success criteria
+
+- What must improve if this decision is right:
+ - Returning hosts re-find and manage schedules more reliably than before.
+- What must not regress while we pursue it:
+ - Schedule creation remains lightweight and skippable past any login prompts.
+ - Attendee participation on `/s/{scheduleKey}` remains obviously
+ guest-friendly and fast to complete.
+ - Finalized-plan rate does not fall because the access model becomes more
+ confusing.
+- Which indicators or behaviors would give us confidence:
+ - Host-return success rises.
+ - Claim or login completion for hosts is healthy without displacing host-link
+ fallback usage.
+ - Create abandonment and attendee drop-off remain flat.
+
+## Accepted downside
+
+- What cost are we deliberately accepting:
+ - Some repeat users may still feel the product lacks the permanence and
+ convenience of a broader account system.
+- Why that cost is preferable to the alternative right now:
+ - It is safer to under-build identity than to over-build it into the create
+ and attendee flows that currently make the product work.
+- Which users or workflows are least well served by this decision:
+ - Power users who want cross-schedule history or universal schedule recovery.
+
+## Mitigations
+
+- How we reduce the main downside of the chosen path:
+ - Improve host-link clarity, make claiming easy, and ensure returning hosts
+ can get back to the right schedule without generic account chrome.
+- Which guardrails or rollout choices are required:
+ - Keep login optional and host-only.
+ - Preserve host-link fallback.
+ - Keep account concepts out of `/s/{scheduleKey}`.
+ - Make any post-create continuity prompt secondary and easy to dismiss.
+- What fallback or rollback boundary we need:
+ - If create completion, attendee completion, or finalized-plan rate worsens
+ after rollout, remove the host-login prompt first and fall back to the
+ host-link model while keeping any claimed ownership data intact.
+
+## Deferred path
+
+- What remains deferred for now:
+ - Participant accounts
+ - Cross-schedule history for all users
+ - Broader account-management surfaces
+ - Identity-based product expansion justified mainly by future optionality
+- What evidence would reopen that path later:
+ - The narrow host path fails to solve return reliability.
+ - Repeat-user continuity becomes a demonstrated product need across a
+ meaningful share of usage.
+ - Broader identity can be introduced without materially harming guest
+ participation.
+- What would make us regret this decision:
+ - Discovering that the product's next stage depends far more on repeat-user
+ continuity than on frictionless first-time participation.
diff --git a/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/tradeoff-map.md b/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/tradeoff-map.md
new file mode 100644
index 0000000..a866781
--- /dev/null
+++ b/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/tradeoff-map.md
@@ -0,0 +1,126 @@
+# Tradeoff Map
+
+This document makes the competing constraints visible before the team pretends
+the decision is obvious.
+
+## Decision context
+
+- What new pressure or request is creating this conflict:
+ - The team can now justify a narrow host-login improvement, but stakeholders
+ are pushing to go further and add broader accounts for repeat users.
+- Which earlier decision or scope boundary is now under pressure:
+ - The exercise 2 decision to avoid a full account system and the exercise 3
+ plan to keep login narrowly host-focused.
+- Which user outcomes are most at risk if the team gets this wrong:
+ - Finalized-plan rate
+ - Create-flow completion speed
+ - Attendee submission completion
+ - Host trust that they can reliably get back to the right schedule
+
+## Option A
+
+- Path name:
+ - Stay narrow: host-focused continuity improvement only
+- What this path is trying to optimize:
+ - Better return access for hosts
+ - Lower risk to the create and attendee flows
+ - Smaller implementation and support surface area
+- Success criteria this path would claim:
+ - Returning hosts recover schedules more reliably
+ - Finalized-plan rate holds steady or improves
+ - Create and attendee completion remain effectively unchanged
+- Which flows or routes it touches most:
+ - `/s/{scheduleKey}/{hostKey}` for host management and claiming
+ - A host-only login entry point
+ - Optional post-create continuity prompts on `/`
+- Likely harms or regressions it introduces:
+ - Some repeat users may still feel the product lacks the polish or permanence
+ of a full account system
+ - Hosts may still need to understand both host-link and account-based access
+ - The narrower move may not fully solve long-term repeat-usage expectations
+- Key assumptions or missing evidence:
+ - The real pain is primarily host return access, not universal identity
+ - Most attendee value still comes from fast, no-account participation
+ - The product does not yet need cross-schedule history badly enough to justify
+ broader scope
+- Mitigations if this path is chosen:
+ - Make host-link purpose and claiming behavior clearer in the UI
+ - Keep login optional and host-only
+ - Measure host-return success and revisit if the narrow path proves
+ insufficient
+
+## Option B
+
+- Path name:
+ - Expand to broader repeat-user accounts
+- What this path is trying to optimize:
+ - Stronger sense of trust and permanence
+ - Easier schedule recovery across multiple visits
+ - Future optionality for identity-based features
+- Success criteria this path would claim:
+ - More returning users can find schedules without saved links
+ - Repeat usage feels more durable and professional
+ - The product gains a clearer long-term identity model
+- Which flows or routes it touches most:
+ - `/` because account creation or login prompts tend to leak into creation
+ - `/s/{scheduleKey}` because attendee identity expectations become harder to
+ avoid
+ - `/s/{scheduleKey}/{hostKey}` because host-link access now competes with
+ authenticated ownership
+ - New auth, recovery, and account-management routes
+- Likely harms or regressions it introduces:
+ - Create-flow friction rises as account concepts compete with quick schedule
+ setup
+ - Attendee completion may drop if the public flow feels more account-shaped or
+ less obviously guest-friendly
+ - Support and implementation cost grow materially because the team now owns
+ auth, recovery, identity state, and confusion between access paths
+ - Route purpose becomes easier to blur across public, host-only, and account
+ surfaces
+- Key assumptions or missing evidence:
+ - More identity always equals more trust in this product category
+ - Repeat users actually want accounts enough to offset the added friction
+ - The team can add broader auth without harming mobile-friendly participation
+ - Future optionality is valuable enough right now to justify present cost
+- Mitigations if this path is chosen:
+ - Keep attendee participation explicitly guest-friendly
+ - Preserve host-link fallback during rollout
+ - Gate account prompts away from the first-create and first-respond moments
+ - Add instrumentation and rollback triggers before broad exposure
+
+## Constraint collisions
+
+- Which priorities are directly in tension:
+ - Host-return reliability and future trust signals versus low-friction create
+ and attendee completion
+ - Product permanence versus route clarity and workflow simplicity
+ - Future optionality versus current implementation and support cost
+- Which route or workflow feels most fragile under this decision:
+ - `/s/{scheduleKey}` is the most fragile because its value comes from being
+ obviously lightweight and guest-oriented
+- Which costs are immediate versus delayed:
+ - Immediate costs: auth complexity, support burden, route confusion, create
+ friction
+ - Delayed benefits: stronger repeat-user continuity and future identity-based
+ product expansion
+- Which downside is easiest for a team to underestimate:
+ - Leakage of account expectations into flows that currently succeed precisely
+ because they are simple and no-account
+
+## Open questions
+
+- What evidence would most reduce uncertainty:
+ - How often hosts fail to return without the private host link
+ - Whether attendee completion drops when identity language is introduced
+ - Whether repeat users in this social-planning context actually expect account
+ persistence
+- What would make the broader path clearly worth it:
+ - Repeated evidence that host-return problems materially hurt finalized-plan
+ rate and cannot be solved by a host-only path
+ - Clear demand for cross-schedule history or identity-based recovery from a
+ meaningful share of active users
+- What would make the narrower path clearly insufficient:
+ - Hosts still fail to re-find schedules even after a careful host-only
+ continuity improvement
+ - The product proves that repeat-user continuity matters more than
+ ultra-lightweight first-use participation
diff --git a/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/tradeoff-review.md b/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/tradeoff-review.md
new file mode 100644
index 0000000..04256d1
--- /dev/null
+++ b/exercises/04.conflicting-constraints/03.problem.pressure-test-the-decision/tradeoff-review.md
@@ -0,0 +1,34 @@
+# Tradeoff Review
+
+This review should pressure-test the chosen tradeoff and define how the team
+would respond if the decision starts failing in practice.
+
+## Failure modes
+
+- How could the chosen path still hurt schedule creation:
+- How could the chosen path still hurt attendee participation:
+- How could the chosen path still hurt host return or finalization:
+
+## Signals to watch
+
+- What would tell us we picked the wrong side of the tradeoff:
+- Which behavioral or product signals matter most:
+- Which signal would make you worry first:
+
+## Required protections
+
+- Which mitigations are non-negotiable for the chosen path:
+- What rollout conditions or guardrails must be present:
+- What must stay unchanged to protect user trust:
+
+## Revision and rollback
+
+- What would you narrow or change first if things go badly:
+- What is the rollback boundary:
+- What would stay worth keeping even if the rollout underperforms:
+
+## Revisit triggers
+
+- What future evidence would justify reopening the deferred path:
+- What would have to become true for the bigger option to be the better
+ tradeoff:
diff --git a/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/README.mdx b/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/README.mdx
new file mode 100644
index 0000000..b34322e
--- /dev/null
+++ b/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/README.mdx
@@ -0,0 +1,11 @@
+# Pressure-Test the Decision
+
+👨💼 Nice work. You did not stop at "we made a call." You defined how the team
+would know whether that call was still protecting the product.
+
+🦉 This is what separates judgment from taste. Good engineers do not just pick a
+direction; they specify the conditions under which they would narrow, revise,
+or reverse it.
+
+The exercise is complete once the tradeoff, its risks, and its revision
+boundary are all explicit.
diff --git a/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/tradeoff-decision.md b/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/tradeoff-decision.md
new file mode 100644
index 0000000..e292376
--- /dev/null
+++ b/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/tradeoff-decision.md
@@ -0,0 +1,80 @@
+# Tradeoff Decision
+
+This brief turns the tradeoff map into a clear recommendation the team can act
+on.
+
+## Decision
+
+- Which path should ship now:
+ - Ship the narrow host-focused continuity path now and keep broader
+ repeat-user accounts deferred.
+- Why this is the right tradeoff at the current product stage:
+ - The product still wins on low-friction scheduling. The most credible problem
+ we have is host return reliability, not universal identity. A broader
+ account move would charge complexity across all major flows before the
+ product has earned that cost.
+- Which constraint are we prioritizing most:
+ - Protect create-flow speed and attendee completion while improving host
+ return reliability enough to help finalized-plan rate.
+- Which constraint are we knowingly not maximizing:
+ - We are not maximizing long-term identity optionality or the polish signal of
+ a full account model yet.
+
+## Success criteria
+
+- What must improve if this decision is right:
+ - Returning hosts re-find and manage schedules more reliably than before.
+- What must not regress while we pursue it:
+ - Schedule creation remains lightweight and skippable past any login prompts.
+ - Attendee participation on `/s/{scheduleKey}` remains obviously
+ guest-friendly and fast to complete.
+ - Finalized-plan rate does not fall because the access model becomes more
+ confusing.
+- Which indicators or behaviors would give us confidence:
+ - Host-return success rises.
+ - Claim or login completion for hosts is healthy without displacing host-link
+ fallback usage.
+ - Create abandonment and attendee drop-off remain flat.
+
+## Accepted downside
+
+- What cost are we deliberately accepting:
+ - Some repeat users may still feel the product lacks the permanence and
+ convenience of a broader account system.
+- Why that cost is preferable to the alternative right now:
+ - It is safer to under-build identity than to over-build it into the create
+ and attendee flows that currently make the product work.
+- Which users or workflows are least well served by this decision:
+ - Power users who want cross-schedule history or universal schedule recovery.
+
+## Mitigations
+
+- How we reduce the main downside of the chosen path:
+ - Improve host-link clarity, make claiming easy, and ensure returning hosts
+ can get back to the right schedule without generic account chrome.
+- Which guardrails or rollout choices are required:
+ - Keep login optional and host-only.
+ - Preserve host-link fallback.
+ - Keep account concepts out of `/s/{scheduleKey}`.
+ - Make any post-create continuity prompt secondary and easy to dismiss.
+- What fallback or rollback boundary we need:
+ - If create completion, attendee completion, or finalized-plan rate worsens
+ after rollout, remove the host-login prompt first and fall back to the
+ host-link model while keeping any claimed ownership data intact.
+
+## Deferred path
+
+- What remains deferred for now:
+ - Participant accounts
+ - Cross-schedule history for all users
+ - Broader account-management surfaces
+ - Identity-based product expansion justified mainly by future optionality
+- What evidence would reopen that path later:
+ - The narrow host path fails to solve return reliability.
+ - Repeat-user continuity becomes a demonstrated product need across a
+ meaningful share of usage.
+ - Broader identity can be introduced without materially harming guest
+ participation.
+- What would make us regret this decision:
+ - Discovering that the product's next stage depends far more on repeat-user
+ continuity than on frictionless first-time participation.
diff --git a/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/tradeoff-map.md b/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/tradeoff-map.md
new file mode 100644
index 0000000..a866781
--- /dev/null
+++ b/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/tradeoff-map.md
@@ -0,0 +1,126 @@
+# Tradeoff Map
+
+This document makes the competing constraints visible before the team pretends
+the decision is obvious.
+
+## Decision context
+
+- What new pressure or request is creating this conflict:
+ - The team can now justify a narrow host-login improvement, but stakeholders
+ are pushing to go further and add broader accounts for repeat users.
+- Which earlier decision or scope boundary is now under pressure:
+ - The exercise 2 decision to avoid a full account system and the exercise 3
+ plan to keep login narrowly host-focused.
+- Which user outcomes are most at risk if the team gets this wrong:
+ - Finalized-plan rate
+ - Create-flow completion speed
+ - Attendee submission completion
+ - Host trust that they can reliably get back to the right schedule
+
+## Option A
+
+- Path name:
+ - Stay narrow: host-focused continuity improvement only
+- What this path is trying to optimize:
+ - Better return access for hosts
+ - Lower risk to the create and attendee flows
+ - Smaller implementation and support surface area
+- Success criteria this path would claim:
+ - Returning hosts recover schedules more reliably
+ - Finalized-plan rate holds steady or improves
+ - Create and attendee completion remain effectively unchanged
+- Which flows or routes it touches most:
+ - `/s/{scheduleKey}/{hostKey}` for host management and claiming
+ - A host-only login entry point
+ - Optional post-create continuity prompts on `/`
+- Likely harms or regressions it introduces:
+ - Some repeat users may still feel the product lacks the polish or permanence
+ of a full account system
+ - Hosts may still need to understand both host-link and account-based access
+ - The narrower move may not fully solve long-term repeat-usage expectations
+- Key assumptions or missing evidence:
+ - The real pain is primarily host return access, not universal identity
+ - Most attendee value still comes from fast, no-account participation
+ - The product does not yet need cross-schedule history badly enough to justify
+ broader scope
+- Mitigations if this path is chosen:
+ - Make host-link purpose and claiming behavior clearer in the UI
+ - Keep login optional and host-only
+ - Measure host-return success and revisit if the narrow path proves
+ insufficient
+
+## Option B
+
+- Path name:
+ - Expand to broader repeat-user accounts
+- What this path is trying to optimize:
+ - Stronger sense of trust and permanence
+ - Easier schedule recovery across multiple visits
+ - Future optionality for identity-based features
+- Success criteria this path would claim:
+ - More returning users can find schedules without saved links
+ - Repeat usage feels more durable and professional
+ - The product gains a clearer long-term identity model
+- Which flows or routes it touches most:
+ - `/` because account creation or login prompts tend to leak into creation
+ - `/s/{scheduleKey}` because attendee identity expectations become harder to
+ avoid
+ - `/s/{scheduleKey}/{hostKey}` because host-link access now competes with
+ authenticated ownership
+ - New auth, recovery, and account-management routes
+- Likely harms or regressions it introduces:
+ - Create-flow friction rises as account concepts compete with quick schedule
+ setup
+ - Attendee completion may drop if the public flow feels more account-shaped or
+ less obviously guest-friendly
+ - Support and implementation cost grow materially because the team now owns
+ auth, recovery, identity state, and confusion between access paths
+ - Route purpose becomes easier to blur across public, host-only, and account
+ surfaces
+- Key assumptions or missing evidence:
+ - More identity always equals more trust in this product category
+ - Repeat users actually want accounts enough to offset the added friction
+ - The team can add broader auth without harming mobile-friendly participation
+ - Future optionality is valuable enough right now to justify present cost
+- Mitigations if this path is chosen:
+ - Keep attendee participation explicitly guest-friendly
+ - Preserve host-link fallback during rollout
+ - Gate account prompts away from the first-create and first-respond moments
+ - Add instrumentation and rollback triggers before broad exposure
+
+## Constraint collisions
+
+- Which priorities are directly in tension:
+ - Host-return reliability and future trust signals versus low-friction create
+ and attendee completion
+ - Product permanence versus route clarity and workflow simplicity
+ - Future optionality versus current implementation and support cost
+- Which route or workflow feels most fragile under this decision:
+ - `/s/{scheduleKey}` is the most fragile because its value comes from being
+ obviously lightweight and guest-oriented
+- Which costs are immediate versus delayed:
+ - Immediate costs: auth complexity, support burden, route confusion, create
+ friction
+ - Delayed benefits: stronger repeat-user continuity and future identity-based
+ product expansion
+- Which downside is easiest for a team to underestimate:
+ - Leakage of account expectations into flows that currently succeed precisely
+ because they are simple and no-account
+
+## Open questions
+
+- What evidence would most reduce uncertainty:
+ - How often hosts fail to return without the private host link
+ - Whether attendee completion drops when identity language is introduced
+ - Whether repeat users in this social-planning context actually expect account
+ persistence
+- What would make the broader path clearly worth it:
+ - Repeated evidence that host-return problems materially hurt finalized-plan
+ rate and cannot be solved by a host-only path
+ - Clear demand for cross-schedule history or identity-based recovery from a
+ meaningful share of active users
+- What would make the narrower path clearly insufficient:
+ - Hosts still fail to re-find schedules even after a careful host-only
+ continuity improvement
+ - The product proves that repeat-user continuity matters more than
+ ultra-lightweight first-use participation
diff --git a/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/tradeoff-review.md b/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/tradeoff-review.md
new file mode 100644
index 0000000..3161476
--- /dev/null
+++ b/exercises/04.conflicting-constraints/03.solution.pressure-test-the-decision/tradeoff-review.md
@@ -0,0 +1,75 @@
+# Tradeoff Review
+
+This review pressure-tests the chosen tradeoff and defines how the team would
+respond if the decision starts failing in practice.
+
+## Failure modes
+
+- How could the chosen path still hurt schedule creation:
+ - Even a host-only continuity prompt could slow the post-create moment if it
+ competes with the more important "share this schedule" success state.
+- How could the chosen path still hurt attendee participation:
+ - Identity language could leak into `/s/{scheduleKey}` through copy, chrome,
+ or navigation and make the flow feel less obviously guest-friendly.
+- How could the chosen path still hurt host return or finalization:
+ - Hosts could become confused about whether they should use the host link, the
+ claimed schedule path, or both, which would weaken the reliability gains the
+ change was meant to create.
+
+## Signals to watch
+
+- What would tell us we picked the wrong side of the tradeoff:
+ - Host return still underperforms while complexity and support cost rise
+ anyway.
+- Which behavioral or product signals matter most:
+ - Create completion rate
+ - Attendee submission completion
+ - Host-return success
+ - Finalized-plan rate
+ - Support questions about host-link versus login confusion
+- Which signal would make you worry first:
+ - Any sign that attendee completion drops or that public responders start
+ asking whether they need accounts.
+
+## Required protections
+
+- Which mitigations are non-negotiable for the chosen path:
+ - Login remains optional and host-only.
+ - Host-link fallback stays valid.
+ - `/s/{scheduleKey}` remains free of ownership and account language.
+ - Route purpose stays crisp across `/`, `/s/{scheduleKey}`, and
+ `/s/{scheduleKey}/{hostKey}`.
+- What rollout conditions or guardrails must be present:
+ - Instrumentation must be in place before exposure grows.
+ - The post-create continuity prompt must be secondary and dismissible.
+ - Failed auth must preserve context rather than trapping the host in generic
+ account flows.
+- What must stay unchanged to protect user trust:
+ - Fast schedule creation
+ - Lightweight guest participation
+ - Trustworthy finalization and timezone clarity
+
+## Revision and rollback
+
+- What would you narrow or change first if things go badly:
+ - Remove or hide the post-create prompt first, then simplify the login entry
+ point if hosts still get lost.
+- What is the rollback boundary:
+ - If the rollout causes a meaningful decline in create completion, attendee
+ completion, or finalized-plan rate, revert back to the host-link-first
+ access model while preserving any safely claimed ownership data.
+- What would stay worth keeping even if the rollout underperforms:
+ - Clearer host-link language and any product copy improvements that help hosts
+ understand return access without adding friction elsewhere.
+
+## Revisit triggers
+
+- What future evidence would justify reopening the deferred path:
+ - A strong repeat-user pattern emerges where cross-schedule history or account
+ recovery materially improves product outcomes.
+ - The narrow host path proves too limited despite strong execution.
+- What would have to become true for the bigger option to be the better
+ tradeoff:
+ - Identity continuity would need to become central to product value rather
+ than a secondary trust or convenience improvement, and the team would need a
+ credible way to add it without degrading guest participation.
diff --git a/exercises/04.conflicting-constraints/FINISHED.mdx b/exercises/04.conflicting-constraints/FINISHED.mdx
new file mode 100644
index 0000000..8753156
--- /dev/null
+++ b/exercises/04.conflicting-constraints/FINISHED.mdx
@@ -0,0 +1,29 @@
+# Conflicting Constraints
+
+Great work finishing Exercise 4.
+
+Before moving on, complete this reflection as a required output.
+
+## Required reflection outputs
+
+1. **Constraint collision**
+ - Which two priorities felt most genuinely in conflict?
+ - Which cost was easiest to underestimate at first?
+
+2. **Tradeoff decision**
+ - Which path did you choose and why was it the right choice now?
+ - Which valuable outcome did you knowingly leave under-optimized?
+
+3. **Success criteria**
+ - What signals would make you feel confident the decision was correct?
+ - Which signal would make you revisit the decision first?
+
+4. **Mitigation and rollback**
+ - What protections were non-negotiable for the chosen path?
+ - What would you narrow, revise, or roll back first if the tradeoff started
+ harming the product?
+
+5. **Future revisit**
+ - What new evidence would justify reopening the deferred option?
+ - What did this exercise teach you about making tradeoffs legible instead of
+ pretending they are not there?
diff --git a/exercises/04.conflicting-constraints/README.mdx b/exercises/04.conflicting-constraints/README.mdx
new file mode 100644
index 0000000..d94d6b0
--- /dev/null
+++ b/exercises/04.conflicting-constraints/README.mdx
@@ -0,0 +1,37 @@
+# Conflicting Constraints
+
+In this exercise, you're the engineer stuck between two reasonable but
+competing product instincts.
+
+The scheduling app now has a plausible narrow path for improving host return
+access. That should make the next decision easier, but it rarely does.
+Stakeholders want the product to feel more trustworthy, more durable, and more
+professional for repeat usage. At the same time, the product still wins because
+it is fast to start, easy to share, and lightweight for infrequent users.
+
+That is the tension: the same change can improve one outcome while quietly
+harming another.
+
+Your job is to make that conflict explicit instead of hand-waving it away.
+You'll identify the competing goals, define what success means for each,
+forecast what could get worse, and defend a specific tradeoff with clear
+mitigations and rollback boundaries.
+
+You'll move through a practical decision sequence:
+
+1. Map the competing constraints and what each path optimizes
+2. Make a clear tradeoff decision with explicit success criteria
+3. Pressure-test the decision by defining mitigation and rollback boundaries
+
+By the end, you should be able to explain:
+
+- which priorities are genuinely in conflict rather than merely inconvenient
+- which product quality signals matter most when those priorities collide
+- why the chosen tradeoff is reasonable now, not just eventually
+- what evidence would justify revisiting the decision later
+
+
+ A strong result is not "we found the perfect answer." It is "we made the
+ conflict legible, chose deliberately, and documented how we would know if the
+ decision stopped being the right one."
+