Project 15 / 28
Migrations & Re-platforming
Move to a faster, cheaper, safer stack — without losing what works.
Legacy platforms, sprawling SaaS bills and fragile databases quietly tax speed, cost and security. We migrate you to a modern, owned stack — standard to headless, onto Cloudflare's edge, off per-task automation, and onto clean governed data — as a controlled, reversible cutover, not a risky rebuild.
Sentry's Migrations and Re-platforming service moves websites, hosting, automation and data onto a modern owned stack — standard CMS to headless, hosting to Cloudflare's edge, Zapier or Make to self-hosted n8n, legacy databases to governed data — as a controlled, reversible cutover rather than a rebuild.
WordPress
Drupal
Joomla
Sitecore 
Astro
Cloudflare 
Sanity, if you edit What actually changes
Your design is carried across exactly as it is. A redesign is a separate project, quoted separately, and only if you want one.
Website replatforming
Off an ageing CMS. Onto Astro, served from Cloudflare's edge.
WordPressDrupalJoomlaSitecore AstroCloudflare Sanity, if you edit An ageing content management system costs you three ways at once: maintenance and licences you keep paying, a break-in risk you carry, and slow pages that lose the visitors your ads and search already paid for. Replatforming removes all three, keeps your design, content and addresses, and leaves a site ready for what comes next.
Why replatformWhat staying put costs you
The valueWhat replatforming returns
Break-in risk you carry
Every plugin, module or extension runs on your server with reach into your database, maintained on someone else's schedule. Attackers use AI to find unpatched sites within hours of disclosure. Staying put means out-patching that, forever.
Attack surface removed
No plugins. No database. No code running on request. The category of exposure disappears instead of being patched.
Debt that blocks upgrades
Years of add-ons, themes and templates each pin a version and block the next upgrade, until nobody is willing to press the button.
Nothing to carry forward
Components to keep compatible go from dozens to zero. Content, design and web addresses come across. The debt is written off at migration.
A bill that repeats every month
Licence renewals, hosting that climbs with traffic, monthly patching hours, out-of-hours emergencies. You pay it whether or not anything improved that month.
Maintenance cost removed
Zero add-on licences. Zero patching hours. Static hosting that does not climb with traffic. The line is removed, not trimmed.
Visitors lost before the page loads
Pages are assembled on request through a stack of add-ons. Traffic your advertising and search budget already bought leaves before the page finishes.
Measurably faster pages
At least a 20 per cent improvement in measured performance where the current measurement leaves room for it. Proven on the same test, before and after.
No record, and slow recovery
Changes made in a CMS admin leave no record of who changed what, and recovery is a restore from a backup of unknown age by whoever still remembers the setup.
Versioned, with one-step rollback
Every change versioned, reviewed and tested before release. Rollback to any previous version in one step. Recovery in minutes.
AI cannot reach your content
Your content is locked in a database behind a login, in markup built for a browser and not for an agent. An AI tool cannot act on what it cannot read.
AI and agentic foundations
Content becomes machine-readable and deployment becomes something software can run. Module 01 of the roadmap below, and the prerequisite for every module after it.
Our own site runs on it. Every page sits in Google’s “good” band for speed, on mobile and desktop, measured 17 September 2026.
5 hours
Median time from a WordPress vulnerability being made public to it being exploited at scale, because attackers now use AI to find and exploit them automatically. Of the WordPress vulnerabilities disclosed in 2025, 91 per cent were in plugins and 9 per cent in themes. Only six were in WordPress itself. Patchstack · State of WordPress Security 2026
How we replatform
01
Discovery
We map the current site, its content, integrations and traffic, and agree what must be preserved.
02
Plan
A written migration plan: scope, page-by-page mapping, redirects, risks, and the measured baseline.
03
Approval
You approve the plan and the agreed targets. No build starts without it.
04
Replatform
Build, test and launch, then re-run the same tests and hand you the before-and-after.
Typical outcome: at least a 20 per cent improvement in measured performance where the current measurement leaves room for it, and the plugin and theme attack surface removed entirely. Proven on the same test, before and after.
Where this sits — the nine-module Growth Engine
- 01Websites
- 02Discovery
- 03Digital funnel
- 04Market sizing
- 05Data & KPIs
- 06Digital voice
- 07Automation & Integration
- 08AI
- 09SAOS
Replatforming is module 01, deliberately first: a fast, secure, agent-readable website is the asset every later module reads from and acts on. Each module stands alone, and there is no obligation to take the next.
What Migrations & Re-platforming delivers
Controlled & reversible
Map current → required, migrate in parallel, then cut over staged with rollback ready — never a big-bang gamble.
Faithful by default
Content, data, integrations, SEO and redirects carried across with care — you keep what works.
Cheaper to run
Edge hosting, self-hosted automation and a clean data layer cut licence, hosting and per-task costs.
Representative stack
n8n
BigQuery
SnowflakeSanityAstro
Next.jsCloudflare
SupabaseKPIs we move — measured, owned, reported
Every KPI is tied to a named owner and a target agreed up front — so this service is accountable to outcomes, not activity, for every stakeholder who consumes the value.
Capability Maturity — we walk you up the five levels
Operated at Level 4 · Quantitatively Managed — measured, predictable and owned. We track this rank and raise it over time. See the five levels →
Why this matters
Functional value, value-stream value — and why it can't be ignored.
Re-platforms websites, hosting, automation and data onto a modern, owned stack — standard→headless, →Cloudflare edge, Zapier/Make→n8n, and legacy→governed databases — via a controlled, reversible cutover.
Removes structural drag from the whole value stream at once: faster surfaces, lower run-cost, fewer dependencies and trustworthy data — so every downstream service compounds on a cleaner foundation.
Legacy stacks compound their cost every year — licence fees, slow pages, per-task automation bills and fragile data. Staying put means paying more each quarter for less speed, less security and less control.
Why we migrate
The same reasons we choose technology — applied to what you already run.
We select technology last, against process readiness and a commercial objective — never tech for its own sake. Migration is that same discipline applied to your current stack: we move you only where it measurably protects value, cuts cost or unlocks speed. How we select technology →
Speed
Modern, edge-served stacks are sub-second. Slow surfaces cost conversion and SEO every single day.
Cost
Cut licence, hosting and per-task SaaS fees — pay for capability, not lock-in or idle enterprise tiers.
Security
Flat, edge architecture with almost nothing left to attack — fewer moving parts, smaller surface.
Ownership & no lock-in
Composable and self-hosted where it counts — you own the system, not rent someone else's.
Measurability
Clean data and KPIs baked in — you cannot improve, automate or apply AI to what you cannot measure.
Fit to process
The right tool for a designed system — selected against process readiness, so it amplifies, not adds friction.
Four migrations we run
A controlled cutover — not a risky rebuild.
Every migration follows the same disciplined path: map the current state, define the required state, carry content, data, integrations and SEO across faithfully, then cut over with rollback ready.
Standard → Headless
WordPress, Sitecore or Drupal rebuilt on Astro + Sanity — faithful, sub-second and far less to attack.
→Hosting → Cloudflare edge
Legacy hosting, AWS or Azure moved to Cloudflare Pages, Workers & R2 — global edge, lower cost, fewer moving parts.
→Zapier / Make → n8n
Per-task SaaS automations moved to self-hosted n8n — own your workflows, cut per-operation fees, no vendor lock-in.
→Logical database migration
Legacy or fragmented databases moved to a clean, governed schema — integrity checked, lineage kept, zero-downtime cutover.
Common migrations we run
Linked migrations have a dedicated playbook — open it for the full method.
How a migration runs
Map it, move it faithfully, cut over with rollback.
A controlled, evidence-based sequence — so you migrate with confidence, and can always step back if needed.
Discovery & audit
Current state, dependencies, content/data inventory and the real risks — before anything moves.
Map current → required
Content, data, integrations, SEO and redirects mapped target-to-source, with success defined up front.
Build & migrate faithfully
Rebuilt in parallel, pixel- and schema-faithful — the live site keeps running throughout.
Controlled cutover
Staged switch with redirects in place and rollback ready — no big-bang gamble.
Verify & measure
Fidelity checks, performance and KPIs confirmed against the model — proven, not assumed.
One data layer · measured & owned
Every service creates data — captured, measured, owned.
Migrations & Re-platforming doesn’t just run — it generates signal. We land that data in one warehouse and turn it into live KPIs tied to a named owner, so performance is attributable and acted on, never trapped inside a single tool.
Data warehouse
Supabase Snowflake
Keboola BigQuery KPIs, dashboards & apps
Plotly
Retool
Apache Superset Platform-agnostic — we connect what you already use and add a warehouse only where it earns its place.
How we deliver · DMAIC
Every Migrations & Re-platforming engagement runs on DMAIC.
Define the goal and its value, measure the baseline, analyse the real constraint, improve with a proven build, then control the gains — so results are predictable, repeatable and defensible, not down to luck.
Agree the goal, value, budget & timescale up front.
Baseline the KPIs above — current state, not guesswork.
Diagnose the real constraint and the solution needed.
Build the chosen solution; prove the uplift.
Lock in the gains; monitor and sustain them.
FAQ
Common questions
How do you migrate without losing our search rankings?
Content, data, integrations, SEO and redirects are carried across deliberately rather than recreated. The current state is mapped to the required state first, so the redirect map is built from what actually exists — the single most common way a migration loses rankings is discovering the old URLs after the switch, not before.
What happens if the migration goes wrong on the day?
It is designed so that it can be undone. Sentry migrates in parallel and cuts over in stages with rollback ready, rather than running a big-bang switch. A migration that cannot be reversed is a gamble regardless of how well it was planned.
What exactly gets migrated?
Four things, together or separately: a standard CMS to headless; hosting to Cloudflare's edge; per-task automation on Zapier or Make to self-hosted n8n; and legacy or fragile databases to a clean governed data layer. Most re-platforming projects fail because only one of the four moves and the others keep taxing the result.
Why not just stay on what we have?
Because legacy stacks compound their cost every year — licence fees, slow pages, per-task automation bills and fragile data. Staying put means paying more each quarter for less speed, less security and less control. The decision is not migrate versus nothing; it is migrate versus an increasing bill.
Will migrating actually reduce what we spend?
That is a large part of the point. Edge hosting, self-hosted automation and a clean data layer cut licence, hosting and per-task costs, which is why the saving is modelled in the Financial Value Model before the work is committed to rather than claimed afterwards.
How do we know a migration is worth doing at all?
The free Discovery session and Financial Value Model produce fourteen structured commercial outputs, including what the current stack costs to keep and what the migration would return. If the numbers do not support moving, that is a legitimate outcome of Discovery.
Free service sheet
WordPress → Astro on Cloudflare
How a WordPress site becomes a static Astro site served from Cloudflare's edge: what changes, what it costs, what it removes, and how the migration is sequenced without losing rankings.
Start here · the free first step
Migrations & Re-platforming starts the same way every engagement does: with Discovery.
Two distinct moves — a free, no-obligation Discovery Session to find your value at stake, then the Discovery & Blueprint: 14 structured outputs — value model, roadmap and business case — before a pound of delivery is committed.
Engaged alone or as one engine