JefCurtis · January 15, 2026 18:41
diff --git a/gistfile1.txt b/gistfile1.txt
 [David M](twist-mention://4028) and I, are spearheading the migrating of our marketing landing pages from Next.js to Astro! 🎯

 This continues the successful experimentation we've been doing with Astro (Help Center and pairing sessions) and we're now ready to tackle the full landing pages migration.

 **Why Astro?**

 Our current Next.js setup has significant pain points: slow page loads, large client bundles, and poor performance, especially on slower connections and mobile devices. We're hitting Next.js limitations that block us from achieving our performance goals, and the framework just seems to be heading in the wrong direction for what we need it for.

 Astro solves these problems with near-zero JavaScript by default and static generation. Plus, we keep our existing Atomic React components—no full rewrite needed.

 ---

 ## 🛠️ Foundation phase

 Before diving into migration, we'll handle:

 - **Migration strategy document** outlining the incremental rollout approach ✅

    - Platform GH issues will be created as pages are ready to go live and replace the Next.js versions
 - **Performance baseline metrics** (Core Web Vitals, bundle sizes, load times) ♻︎

    - Request Metrics makes this easy for us. We’ll add the before and after to the [migration spreadsheet](https://docs.google.com/spreadsheets/d/11EGImfn0SsLXxrx3ItmzCeytPe0jj6V5y0hWjbchihQ/edit?gid=0#gid=0).
 - **Component compatibility audit** (React Server Components → Astro islands) ✅

 ---

 ## 🤖 AI-Assisted Migration Approach

 We'll be leveraging AI tools to handle the mechanical parts of the migration (component conversions, import updates, prop transformations). However, finding the right balance between AI efficiency and human reviewability is crucial.

 Our approach:

 - **AI handles**: Repetitive component conversions, boilerplate updates, test scaffolding
 - **Humans handle**: Architecture decisions, performance optimizations, edge cases
 - **Tests cover everything**: Comprehensive test suites validate AI-generated changes before review

 Each PR will include:

 - Clear description of what was automated
 - Test coverage demonstrating correctness
 - **Reasonable diff sizes** for effective human review

 This isn't about replacing human judgment, but much of the business logic and solutions have already been battle tested in production and should continue to work well work with Astro. That being said, we need to strike the right balance between PR diff sizes and efficiency to accomplish the lofty goal of **migrating all the pages in one quarter**.

 ---

 ## ✅ Work scope

 Migrate all marketing landing pages incrementally to Astro, starting with higher-traffic pages.

 Everything must be:

 - **Faster**: Minimal client bundles, near-instant server responses
 - **More reliable**: Statically rendered at build time whenever possible, no runtime failures
 - **Better SEO**: Improved Core Web Vitals scores, faster indexing

 Target pages include:

 - Homepage, Features, Pricing
 - Productivity method pages
 - Contentful custom landing pages
 - Integration pages
 - Template pages
 - The entire Inspiration hub
 - Critical upgrade workflows

 ---

 ## 🚢 Deployment plan

 We'll deploy incrementally using CloudFront routing rules to gradually shift traffic from Next.js to Astro.

 **Phase 1**: Deploy Astro build alongside Next.js (separate subdomain or path)
 **Phase 2**: Route high-traffic pages to Astro, monitor metrics and issues
 **Phase 4**: Switch primary traffic, keep Next.js as fallback
 **Phase 5**: Full cutover once metrics validate success

 **Phase 6 (stretch)**: Wipe the Next.js project from the monorepo. 💣💥

 This approach minimizes risk and allows us to rollback at any point.

 ---

 ## 📊 Measures of success

 **Hypothesis**: Migrating to Astro improves performance and user experience.

 Metrics:

 - **Lighthouse scores**: All pages hit 90+ performance scores
 - **Core Web Vitals**: Pass all thresholds (LCP, FID, CLS)
 - **Bundle size**: Reduce from ~2MB to <100KB JavaScript
 - **Page speed**: Sub-1s First Contentful Paint on 4G

 We'll track these weekly during rollout and compare against baseline.

 ---

 **References:**

 - [Solution spec](https://docs.google.com/document/d/1ewURPqXlyQpRaTBrAM7WWekrJtJKF-bnLqUtSujmAgo/edit?tab=t.0)
 - [Migration spreadsheet](https://docs.google.com/spreadsheets/d/11EGImfn0SsLXxrx3ItmzCeytPe0jj6V5y0hWjbchihQ/edit?gid=0#gid=0)

 Questions, concerns, or suggestions? Drop them below! 👇
	[David M](twist-mention://4028) and I, are spearheading the migrating of our marketing landing pages from Next.js to Astro! 🎯

	This continues the successful experimentation we've been doing with Astro (Help Center and pairing sessions) and we're now ready to tackle the full landing pages migration.

	Why Astro?

	Our current Next.js setup has significant pain points: slow page loads, large client bundles, and poor performance, especially on slower connections and mobile devices. We're hitting Next.js limitations that block us from achieving our performance goals, and the framework just seems to be heading in the wrong direction for what we need it for.

	Astro solves these problems with near-zero JavaScript by default and static generation. Plus, we keep our existing Atomic React components—no full rewrite needed.

	---

	## 🛠️ Foundation phase

	Before diving into migration, we'll handle:

	- Migration strategy document outlining the incremental rollout approach ✅

	- Platform GH issues will be created as pages are ready to go live and replace the Next.js versions
	- Performance baseline metrics (Core Web Vitals, bundle sizes, load times) ♻︎

	- Request Metrics makes this easy for us. We’ll add the before and after to the [migration spreadsheet](https://docs.google.com/spreadsheets/d/11EGImfn0SsLXxrx3ItmzCeytPe0jj6V5y0hWjbchihQ/edit?gid=0#gid=0).
	- Component compatibility audit (React Server Components → Astro islands) ✅

	---

	## 🤖 AI-Assisted Migration Approach

	We'll be leveraging AI tools to handle the mechanical parts of the migration (component conversions, import updates, prop transformations). However, finding the right balance between AI efficiency and human reviewability is crucial.

	Our approach:

	- AI handles: Repetitive component conversions, boilerplate updates, test scaffolding
	- Humans handle: Architecture decisions, performance optimizations, edge cases
	- Tests cover everything: Comprehensive test suites validate AI-generated changes before review

	Each PR will include:

	- Clear description of what was automated
	- Test coverage demonstrating correctness
	- Reasonable diff sizes for effective human review

	This isn't about replacing human judgment, but much of the business logic and solutions have already been battle tested in production and should continue to work well work with Astro. That being said, we need to strike the right balance between PR diff sizes and efficiency to accomplish the lofty goal of migrating all the pages in one quarter.

	---

	## ✅ Work scope

	Migrate all marketing landing pages incrementally to Astro, starting with higher-traffic pages.

	Everything must be:

	- Faster: Minimal client bundles, near-instant server responses
	- More reliable: Statically rendered at build time whenever possible, no runtime failures
	- Better SEO: Improved Core Web Vitals scores, faster indexing

	Target pages include:

	- Homepage, Features, Pricing
	- Productivity method pages
	- Contentful custom landing pages
	- Integration pages
	- Template pages
	- The entire Inspiration hub
	- Critical upgrade workflows

	---

	## 🚢 Deployment plan

	We'll deploy incrementally using CloudFront routing rules to gradually shift traffic from Next.js to Astro.

	Phase 1: Deploy Astro build alongside Next.js (separate subdomain or path)
	Phase 2: Route high-traffic pages to Astro, monitor metrics and issues
	Phase 4: Switch primary traffic, keep Next.js as fallback
	Phase 5: Full cutover once metrics validate success

	Phase 6 (stretch): Wipe the Next.js project from the monorepo. 💣💥

	This approach minimizes risk and allows us to rollback at any point.

	---

	## 📊 Measures of success

	Hypothesis: Migrating to Astro improves performance and user experience.

	Metrics:

	- Lighthouse scores: All pages hit 90+ performance scores
	- Core Web Vitals: Pass all thresholds (LCP, FID, CLS)
	- Bundle size: Reduce from ~2MB to <100KB JavaScript
	- Page speed: Sub-1s First Contentful Paint on 4G

	We'll track these weekly during rollout and compare against baseline.

	---

	References:

	- [Solution spec](https://docs.google.com/document/d/1ewURPqXlyQpRaTBrAM7WWekrJtJKF-bnLqUtSujmAgo/edit?tab=t.0)
	- [Migration spreadsheet](https://docs.google.com/spreadsheets/d/11EGImfn0SsLXxrx3ItmzCeytPe0jj6V5y0hWjbchihQ/edit?gid=0#gid=0)

	Questions, concerns, or suggestions? Drop them below! 👇
No results found