Site migrations
Pre-flight audit, redirect map, parity tests, post-launch 30-day watch. Zero-loss is the bar — we've never missed it on a managed migration. Ask for the case studies.
Service · 05 / 06
We ship code, not tickets.
Engineering-led technical SEO. Site migrations, Core Web Vitals, JS rendering, schema, log-file analysis. We open PRs against your repo, write the specs, and stay on the Slack channel the night the migration ships.
Stack fluency
Next.js (app router)Shopify · Liquid + HydrogenWebflowSanity / headless WPBigCommerceAstroRemixCustom Rails / Django
We've shipped technical work on each of the above this year. Yours isn't on the list? Tell us about it on the scoping call.
What we ship
Six engineering disciplines, in-house.
None of this is subcontracted. The person profiling your INP problem is also the person opening the PR. The person reading your logs is on the migration Slack channel.
Core Web Vitals + INP
Real-user data over Lighthouse theatre. We profile, identify the hotspots that move CrUX, and ship the changes against your repo. INP work is most of the calendar in 2026.
JavaScript SEO
Render-then-crawl vs. crawl-then-render, partial hydration footguns, dynamic content discoverability. We test against Googlebot's render and tell you exactly what's invisible.
Schema implementation
Beyond the basics. Schema patterns that feed AI Overviews, knowledge-panel hooks, product schema that survives Shopify variant bugs. We write the JSON-LD; you merge the PR.
Log-file analysis
Two-to-eight weeks of raw server logs analysed by hand. Where Googlebot is wasting crawl budget, what's getting served slowly, what's quietly being skipped. Underrated.
Internal linking architecture
Template-level link distribution analysis. The single highest-leverage on-page lever on most sites — and the one most agencies don't actually know how to fix at scale.
How a technical engagement runs
No Notion docs. No 'recommendations.'
The deliverable is what landed in your repo this sprint. Not a slide deck, not a backlog of suggestions for your team to ticket — actual merged work, or a pending PR with the reviewer tagged.
- Step 01
Repo access
We get pushed to a fork or branch. Read-only access for the first week while we explore; write access once the scope is signed.
- Step 02
Specs first
Every change has a one-pager spec before any code lands. Your engineers sign off. We don't surprise anyone.
- Step 03
PR-by-PR
Work merges in small, reviewable PRs. We tag the reviewer on your side. Description includes the SEO rationale, the test, and the rollback plan.
- Step 04
Post-merge watch
We monitor crawl, index, and Core Web Vitals for 14 days after each merge. If we caused a regression, we fix it before billing the next sprint.
FAQ
The questions engineering teams ask.
- Do you actually write PRs against our repo?
- Yes. We open PRs in a feature branch, write the description, link the issue (or open one), and tag your reviewer. We don't merge anything ourselves — your team has final approval. Most of our technical retainers run this way; we treat it like an external contributor relationship.
- What if our stack is something weird?
- We're stack-agnostic if the stack is rendering HTML over HTTP. The list of frameworks we've shipped on isn't exhaustive — anything modern, we'll figure out. The exception is legacy CMSes nobody supports anymore (early Magento, raw PHP from 2014). We'll tell you on the scoping call.
- How do you handle migrations?
- Pre-flight: full audit of the legacy site + a redirect-map review. During: parity tests on a staging environment, a launch checklist signed off by both teams, and we're on Slack the night of launch. After: 30-day watch with daily index-coverage and ranking deltas. We've shipped 47 managed migrations since 2022 with zero permanent traffic loss.
- Can you handle our migration if our agency built the new site?
- Yes — and this is often where we add the most value. We pre-flight what their team built, catch the gaps before launch, and run the post-launch watch. We don't need to have built the site to migrate it safely.
Got a migration on the roadmap? Bring it to a scoping call before the migration partner is locked in — we'll tell you what to watch for.
Scope a technical engagement