Case study · Construction and hotel refurbishment
Elite Smart Build.
Elite Smart Build, a Spanish hotel refurbishment contractor with more than 200 projects in over twenty countries, runs its works on a platform of its own. Polargate built eleven modules on Supabase and React: a daily Gantt, budgets by role, alerts per department and material reception with automatic incidents.

At a glance
- Client
- Elite Smart Build
- Sector
- Construction and hotel refurbishment
- Year
- 2026
- Live site
- elitesmartbuild.com
- Stack
- ViteReactTypeScriptTailwind CSSshadcn/uiReact RouterSupabasePostgreSQLDenoResendVercelGitHubClaude
01 · Elite Smart Build
The challenge.
Elite Smart Build refurbishes hotels in more than twenty countries, with thirteen years in the market and more than 200 projects delivered. As the company kept growing, it wanted its seven departments, from purchasing and logistics to every site team, to work from the same live information, on a phone on site or at the office, with a tool that belongs to Elite.
For example: 100 / 100 / 100 Lighthouse mobile: accessibility, best practices, SEO. KD.tech
02 · Elite Smart Build
What we built.
Polargate started from the way Elite already works, its phases, departments and roles, turned it into a clear data model and agreed every modelling question with the client in writing before building anything. Scope was fixed in a signed specification: eleven modules for Phase 1. We then built a mobile-first web application on Vite, React, TypeScript and Tailwind, with Supabase Postgres, Storage, Auth and Edge Functions behind it, deployed on Vercel and running on the client's own domain.

For example: 2 languages live on the site, Spanish and English. Global Dynamic Security Group
03 · Elite Smart Build
The result.
The first version went live in production about a month after development started, with ten role-based accounts and the alert engine running every business day. Row-level security is forced on all 23 tables, and a database trigger writes every change to an immutable audit log. Before the first user logged in, the platform passed an adversarial multi-agent audit and all 59 confirmed findings were fixed. Elite's team and Polargate review it every week and keep adding improvements as the company grows.
- modules delivered in Phase 1
- 11
- Postgres tables with row-level security forced
- 23
- from development start to first testable version in production
- ~1 month
- role-based accounts live on the client's own domain
- 10
- audit findings confirmed and fixed before client testing
- 59
For example: 536/536 translation keys present in each of the three languages. Estepona Golf
04 · Elite Smart Build
How we did it.
For example: 4 days from diagnosis to the new backend running. Concert Music Festival
The context
Elite Smart Build delivers turnkey hotel refurbishments without structural works. It has thirteen years in the market and more than 200 projects across twenty countries. Its seven departments, from purchasing and logistics to the site teams, run dozens of works at the same time, and the company wanted a platform of its own where everyone sees the same live information: the daily plan, budgets, tasks and every delivery that reaches a site.
What we found in discovery
We started from the way Elite already works, not from a blank page: its works calendar, its quotations, its site procedures and its phases by department. From those we derived the entities that matter: works, phases, departments, people, suppliers, quotations, shipments, receptions and incidents. Twelve modelling questions were agreed with the client in writing before a line of application code existed, and the scope was fixed in a signed specification: eleven modules for Phase 1.
How it runs now
Access control lives in the database, not in the interface: row-level security is enabled and forced on all 23 tables, roles come from a ten value enum, and a trigger writes every change to an immutable audit log that only management can read. A Supabase Edge Function on Deno runs Monday to Friday and emails each department its open items from the client's own domain, verified for DKIM and SPF through Resend.
The application was scaffolded in Lovable and then fully detached: its own GitHub repository, its own Supabase project and its own Vercel deployment, with no dependency on a third-party runtime. Before the first user logged in the platform passed an adversarial multi-agent audit and all 59 confirmed findings were fixed. The first version reached production about a month after development started, with ten role-based accounts and a weekly review cadence with the client team.
FAQ
Questions about this project.
How long does it take to replace our spreadsheets with a custom system?
Can we keep working on site while the system is being built?
Who can see costs and margins inside the system?
Does it work on a phone on the building site?
Do we own the code and the database?
For example: 22 public routes prerendered as static HTML. Hotel Marien
Same sector
More work in this sector.
Other projects for companies that face the same problems.
For example: 16 indexable URLs, Spanish and English. Hotel Talavera
Next project
Global Dynamic Security Group.
Private security services, Dominican Republic
Brand and bilingual site for Global Dynamic Security Group
For example: 105,212 characters of body text readable by AI crawlers across 65 URLs, up from zero. Malacca Tables
Get started
Tell us what you want to build.
Websites, apps, custom systems and AI for companies that need them in production in weeks. Fill in the form or book a 30 minute call: a senior engineer answers within 48 business hours with a first take on scope, timeline and price.
First step
Two weeks to close scope, architecture and price before anything is built.
No price list. You get a private quote, in writing, after a short call.
Then, to build, always at a fixed price
Not ready yet?
- AI visibility auditOne-off, fixed price.

