Skip to content

How we work

No portfolio theatre. Just how we operate.

This studio was founded in 2026 and is building its client list in public. What it is not starting from is zero: Kristian Bansil worked with the 12 companies below before founding it: retail groups, hotels, law firms and distributors, most of them across fourteen years in the UAE. They are his prior engagements, not clients of this studio. What you won't find below is stock mockups or invented metrics. Here is that track record, then the process, the standards we hold and the numbers we agree to be judged on.

Prior work

Who the founder has worked with.

These are Kristian's engagements from before React Web Development Services was founded in 2026. They are not clients of this studio, and nothing here is offered as its case study. Engagement details are available on request, including reference calls where the company has agreed to them. We publish no metrics we cannot attribute.

Founding clients

The offer we make while we're building our case-study shelf

  • A success metric agreed in writing before we start: cost per lead, conversion rate, bookings, load time. Your number, not a vanity one.
  • Weekly demos from week one, with a live staging link you can open on your phone.
  • Fixed price after paid discovery. If we scoped it wrong, that's our problem, not a change order.
  • Founding-client rate in exchange for a case study and a reference call. The three tiers are published, first three clients only: Presence ₱18,000, Starter ₱32,000, Ignite ₱58,000.
  • 30 days of post-launch optimisation included, because launch day is when the learning starts.
  • Full handover, always: repo, domain, ad accounts, CRM. You could fire us and lose nothing but us.

The engagement

Six phases. You see working software every Friday.

  1. 01 / Week 1

    Paid discovery: ₱25,000, credited to your build

    Stakeholder interviews, audit of what exists, technical requirements, integration map and a fixed quote. We charge for this deliberately: it filters people who aren't serious, and it means you get consulting instead of a pitch. If you don't proceed, the document is still yours.

  2. 02 / Week 2

    Architecture, decided on paper

    Information architecture, data model, third-party integrations, hosting, security posture and the measurement plan. Changing a decision here costs an hour; changing it in week eight costs a fortnight.

  3. 03 / Weeks 3–5

    Design in the browser

    Real interfaces you can click, on your own device, with your own content, not a static mockup that hides every hard interaction. Approved screens become a design system your future work inherits for free.

  4. 04 / Weeks 5–11

    Build, with weekly demos

    Typed, reviewed, tested code on a staging link that exists from day one. Every Friday you see what moved. You will never wait until the end to discover we misunderstood you.

  5. 05 / Launch week

    QA, accessibility, speed, handover

    Cross-browser and device QA, WCAG 2.2 AA pass, Core Web Vitals in the green, analytics verified against real events, SEO redirects mapped. Then credentials and code transfer to you with a recorded walkthrough.

  6. 06 / 30 days after

    Optimise against reality

    Heatmaps, funnels and search data replace our assumptions. We fix what real users struggle with, then either hand you a documented playbook or run it monthly as your growth team.

Standards

What “done” means here.

These are pass/fail gates on every launch, not aspirations. If we miss one, we fix it before we invoice the final 10%.

  • < 2.5s

    Largest Contentful Paint on a mid-range Android over Philippine mobile data, not on a MacBook over fibre.

  • 90+

    Lighthouse performance, accessibility, best practices and SEO on the templates that matter.

  • WCAG 2.2 AA

    Keyboard navigable, screen-reader tested, contrast verified, motion respectful of prefers-reduced-motion.

  • 100%

    Conversion events verified end-to-end before launch. Untracked launches are how agencies hide poor results.

Engineering

Boring choices, on purpose.

We pick technology you can hire for in the Philippines in three years' time, not whatever trended last quarter. Every choice below is one we'd defend in front of your CTO.

  • Next.js / React
  • TypeScript
  • Node
  • PostgreSQL
  • Tailwind
  • Sanity / Payload
  • Vercel / AWS
  • Shopify
  • React Native
  • WebGL
  • Claude API
  • HubSpot

Graceful degradation, everywhere

No WebGL? The hero falls back to a CSS gradient. No JavaScript? Content and navigation still work. Payment gateway not configured? Checkout says so plainly instead of faking a receipt. Software that lies about its state is worse than software that fails.

Built for Philippine conditions

Tested on mid-range Android over mobile data. Redundant internet and power on our side, offsite backups, fully remote-capable operations, because in Central Luzon a typhoon is a certainty, not a contingency.

Judge us on the first sprint, not the sales call.

Start with paid discovery. One week, ₱25,000, credited in full against the build. You'll know exactly how we work before you commit to anything.