Product Improvement Execution Plan
April 14, 2026 · View on GitHub
Generated: 2026-04-13
This document converts docs/product-improvement-roadmap.md into an execution plan.
It is intended to answer:
- what to build first
- what each milestone should include
- what success looks like before moving on
- which workstreams can be parallelized safely
This is not a speculative wishlist. It is an implementation sequence for the next round of product work.
Working Assumptions
This plan assumes the current architecture remains in place:
- PWB remains the system of record for listings, search, and enquiries
- EmDash remains the CMS and admin extension surface
- Astro remains the public frontend composition layer
The execution plan also assumes these product goals are still primary:
- reduce setup friction
- make the admin more productized
- improve the frontend’s commercial quality
- increase trust and lead-generation value
- make content and listing curation feel unified
Delivery Strategy
The roadmap should not be implemented as one large redesign. It should be delivered as five sequential phases.
Each phase should:
- ship usable value on its own
- leave the system in a stable state
- unlock the next layer of product work
Recommended sequence:
- Setup and website control surface
- Homepage merchandising and trust blocks
- Lead capture and conversion surfaces
- Stronger theme families and page-level polish
- Content-data cohesion and reporting
Phase Overview
| Phase | Focus | Outcome |
|---|---|---|
| 1 | Setup and website control surface | Faster onboarding and clearer admin structure |
| 2 | Homepage merchandising and trust | More compelling out-of-the-box public site |
| 3 | Leads and conversion | Stronger CTA paths and better enquiry value |
| 4 | Theme differentiation | More premium and more distinct frontend presentation |
| 5 | Cohesion and optimization | Better curation across listings/content and better measurement |
Phase 1: Setup and Website Control Surface
Goal
Create a clearer first-run and day-to-day website management experience.
Why first
This phase improves the base product for every future customer and every later feature. It has the highest leverage because it reduces confusion across the whole admin.
Milestone 1.1: Launch Checklist
Scope
Add an admin-facing launch checklist or site readiness panel.
The checklist should cover:
- brand identity present
- office details present
- homepage hero/content present
- menu configured
- theme chosen
- property connection healthy
Suggested implementation shape
- new plugin admin screen or dashboard widget
- read-only status derived from current settings/content state
- deep links to the exact pages that resolve missing items
Deliverables
- checklist UI
- readiness state computation helper
- links into existing admin screens
Acceptance criteria
- a first-time user can tell what is missing without guessing
- every missing item links to a useful editor/settings destination
- no checklist item depends on hidden/manual developer knowledge
Milestone 1.2: Website Settings Information Architecture
Scope
Make key website controls easier to find and understand.
Group or relabel surfaces around:
- Brand
- Office
- Homepage
- Theme
- Search Experience
- SEO
Suggested implementation shape
- add a “Website” grouping or index page in admin
- improve titles and descriptions of existing plugin settings pages
- add “View page” / preview affordances
Deliverables
- clearer navigation labels
- short descriptive copy for important settings screens
- preview links from relevant settings pages
Acceptance criteria
- common website edits require fewer clicks
- settings pages explain what they affect
- the admin feels less like generic CMS navigation
Milestone 1.3: Better Seed and Default Experience
Scope
Improve the default seeded experience so the site looks commercially plausible before custom work begins.
Suggested implementation shape
- strengthen
seed/seed.jsondemo copy and sections - ensure brand and office defaults are coherent
- preconfigure a homepage with stronger trust and merchandising balance
Deliverables
- improved demo seed content
- improved default homepage section mix
- better branded defaults for trust/contact areas
Acceptance criteria
- a fresh seed looks intentionally designed
- the homepage does not feel sparse or generic
- there are fewer “demo smell” strings and placeholders
Phase 1 validation
Before moving to Phase 2:
- first-run editing should feel guided
- the core website controls should be discoverable
- the seeded product should be presentation-ready
Phase 2: Homepage Merchandising and Trust
Goal
Make the public-facing site stronger without requiring bespoke development.
Why second
Once setup and navigation are clearer, the next highest-leverage improvement is the quality of the homepage and trust story.
Milestone 2.1: Featured Listing Curation
Scope
Give editors direct control over featured listing sections.
Support:
- manual featured selection
- newest listings
- sale-only / rental-only
- area/category-based selections
Suggested implementation shape
- homepage section configuration
- reusable listing collection helper
- controlled fallback behavior when selection is thin
Deliverables
- curated listing section controls
- rendering support for featured collections
- graceful fallback rules
Acceptance criteria
- editors can control homepage listing emphasis without code
- sparse inventory still renders acceptably
- featured sections stay visually coherent across themes
Milestone 2.2: Trust Block Library
Scope
Add reusable trust-building content blocks.
Recommended first set:
- testimonials
- office credibility / local expertise
- recent sales or key stats
- agent/team intro section
Suggested implementation shape
- add page parts or reusable content modules
- make these available on the homepage and general pages
Deliverables
- at least 2-4 trust-oriented reusable blocks
- seed/demo content using those blocks
Acceptance criteria
- homepage and core pages can include trust signals without custom development
- trust blocks fit the current frontend language
- editors can reuse them easily
Milestone 2.3: Homepage Composition Pass
Scope
Rebalance the homepage as a productized page template rather than an ad hoc page.
Recommended section order:
- Hero
- Featured listings
- Area or lifestyle positioning
- Trust/testimonials
- Secondary CTA
- Latest editorial or insights
Deliverables
- improved homepage composition defaults
- section ordering guidance in docs/admin
Acceptance criteria
- homepage feels commercially persuasive
- the page has a clear narrative and CTA progression
Phase 2 validation
Before moving to Phase 3:
- the homepage should look stronger without custom code
- trust-building modules should be part of the standard product
- editors should have real curation control over homepage listings
Phase 3: Leads and Conversion
Goal
Make lead-generation pathways clearer and more configurable.
Why third
After setup and presentation improve, the next leverage point is conversion.
Milestone 3.1: Property Page CTA System
Scope
Standardize CTA behavior on property detail pages.
Support:
- primary CTA type
- CTA label
- placement rules
- mobile behavior
Recommended CTA types:
- book viewing
- request callback
- WhatsApp/chat
- general enquiry
- valuation request
Deliverables
- CTA configuration surface
- updated property detail CTA rendering
- responsive behavior rules
Acceptance criteria
- CTA behavior is configurable without code edits
- property pages always have a clear primary action
- mobile CTA treatment feels intentional, not bolted on
Milestone 3.2: Contact and Lead Trust Pass
Scope
Improve contact and enquiry-related trust surfaces.
Potential improvements:
- richer office panel
- clearer team or office identity on contact pages
- trust signals near enquiry forms
Deliverables
- improved contact page composition
- stronger contextual trust cues near enquiry pathways
Acceptance criteria
- contact-oriented pages feel credible and complete
- enquiry forms are supported by visible trust signals
Milestone 3.3: Enquiry Attribution Baseline
Scope
Capture minimal useful source information with enquiries.
Baseline fields:
- page type
- property slug if applicable
- CTA source identifier
- optional referrer/campaign when available
Deliverables
- attribution model for enquiries
- source propagation into submission flow
Acceptance criteria
- an enquiry can be tied back to its page and CTA source
- no sensitive tracking assumptions are required for baseline attribution
Phase 3 validation
Before moving to Phase 4:
- property pages should have a clear and configurable CTA strategy
- contact and enquiry surfaces should feel more trustworthy
- baseline enquiry source context should exist
Phase 4: Theme Differentiation and Frontend Polish
Goal
Make themes feel meaningfully different and raise the visual ceiling of the product.
Why fourth
Once setup, homepage composition, and conversion are stronger, frontend refinement has more leverage and less risk of being polish on top of weak product structure.
Milestone 4.1: Theme Families
Scope
Define a small number of visibly different theme families.
Suggested first set:
- Editorial
- Coastal / Lifestyle
- Urban / Modern
Each family should affect more than tokens:
- homepage rhythm
- card proportions
- typography use
- hero treatment
- section spacing behavior
Deliverables
- theme family definitions
- corresponding frontend variations
- updated admin descriptions/presets
Acceptance criteria
- two themes should be recognizably different at a glance
- theme choice should influence layout feel, not only colors
Milestone 4.2: Mobile-first Quality Pass
Scope
Refine mobile behavior for the most important frontend flows:
- property detail page
- gallery
- search and filter
- sticky CTA behavior
Deliverables
- spacing and hierarchy refinements
- mobile interaction fixes
- improved gallery/search ergonomics
Acceptance criteria
- mobile property browsing feels deliberate and usable
- CTA behavior on small screens is strong without being noisy
Milestone 4.3: Interaction and Motion Pass
Scope
Improve perceived quality in high-visibility interactions:
- gallery changes
- search results updates
- homepage section reveals
- card hover/selection behavior
Deliverables
- meaningful motion improvements
- no gratuitous animation
Acceptance criteria
- interaction quality improves perceived polish
- animations support hierarchy and feedback rather than distracting from content
Phase 4 validation
Before moving to Phase 5:
- theme families should feel distinct
- mobile core flows should be strong
- key interactions should feel premium
Phase 5: Content-Data Cohesion and Optimization
Goal
Make curation easier across listing data and editorial content, then add basic optimization tooling.
Why fifth
This phase benefits from the stronger setup, homepage, CTA, and theme systems that earlier phases establish.
Milestone 5.1: Curated Listing Collections Everywhere
Scope
Expand curated listing selection beyond the homepage into reusable page-building contexts.
Examples:
- area pages
- landing pages
- lifestyle pages
- featured development pages
Deliverables
- reusable listing collection surface
- configurable page-part integrations
Acceptance criteria
- editors can combine live listing sets with editorial storytelling on multiple page types
Milestone 5.2: Area and Lifestyle Patterns
Scope
Add productized page patterns that combine:
- editorial intro
- local context
- curated listing feed
- CTA and trust sections
Deliverables
- area/lifestyle content pattern definitions
- reusable sections or templates
Acceptance criteria
- the product can support strong location-led pages without custom assembly each time
Milestone 5.3: Lead and CTA Reporting Baseline
Scope
Expose the first useful admin reporting layer for:
- CTA usage
- enquiry source breakdown
- key page conversion touchpoints
Deliverables
- basic reporting surface or dashboard module
- meaningful but limited metrics set
Acceptance criteria
- agencies can see which pathways are creating leads
- reporting remains simple and operationally useful
Phase 5 validation
Phase 5 is complete when:
- content and listing curation feel more unified
- high-value page types can be assembled from reusable patterns
- the product exposes basic operational feedback loops
Parallel Workstreams
Some work can be done in parallel, but only where write scopes stay separate.
Safe parallel lanes
- admin IA and settings copy improvements
- homepage section/block implementation
- seed/default content improvements
- mobile/frontend polish work
- documentation updates
Unsafe or tightly coupled lanes
- multiple simultaneous edits to the same theme/settings plugin
- homepage composition changes while section APIs are still unstable
- CTA system and enquiry attribution if both depend on the same form/request path
Recommended Milestone Packaging
Keep milestone scope small enough to ship intentionally.
Recommended packaging:
- one PR per milestone
- one commit series per milestone
- no mixing of unrelated admin/frontend changes without a clear product reason
Recommended naming:
Phase 1 Milestone 1 launch checklistPhase 2 Milestone 1 homepage featured curationPhase 3 Milestone 1 property CTA system
Validation Commands
At the end of each milestone, run:
pnpm run test:run
pnpm run typecheck
If docs changed, also run:
pnpm run test:run -- src/docs-validation.test.ts
If frontend/admin behavior changed materially, also perform browser verification on the affected flow.
Suggested Immediate Next Milestones
If work starts now, use this exact order:
- Phase 1 Milestone 1: Launch Checklist
- Phase 1 Milestone 2: Website Settings Information Architecture
- Phase 1 Milestone 3: Better Seed and Default Experience
- Phase 2 Milestone 1: Featured Listing Curation
- Phase 2 Milestone 2: Trust Block Library
That sequence provides the best balance of:
- visible product improvement
- low architectural churn
- strong leverage for all later work
Definition of Success
This execution plan is successful if, after these phases:
- a new site can be set up quickly and confidently
- the admin feels clearly shaped for property-site outcomes
- the homepage and property pages feel commercially credible
- lead paths are obvious and configurable
- editors can curate listings and content together without friction
If a proposed task does not materially improve one of those outcomes, it should be questioned before being added to the active plan.