Most agencies in India take 8–12 weeks to build a custom website. We average 14–21 days. Same scope, same quality, same React + Next.js stack. The difference isn't speed of typing — it's the process.
Here's exactly how it works.
The 21-day timeline
Day 1 — free 15-minute scoping call
Almost always on WhatsApp or Google Meet. We ask three things: what does the business do, who is the visitor, and what one action do you want them to take. We turn that into a one-pager listing every page on the site and what each page is for.
If the founder can't answer the third question in one sentence, that's the first problem we solve — and usually the most valuable part of the engagement. A site without a clear primary CTA gets built and then doesn't convert. We refuse to start without one.
Day 2–3 — free 48-hour mockup
We design the homepage in Figma. Real copy (extracted from the call), real layout, real visual hierarchy. No template, no placeholder text. The mockup is sent over WhatsApp / email with a one-line summary of the design decisions ("we put the main CTA above the fold because…").
The founder either approves or asks for changes. Usually 1–2 rounds. If we're still iterating after 4 rounds, something deeper is off — usually the brief itself — and we go back to day 1.
Critically: nothing gets coded until the design is approved. The biggest source of agency delays is "let's start coding and figure out design as we go" — that produces months of rework.
Day 4–14 — build phase with daily updates
The site is built in our staging environment on Vercel with a public URL the founder can open on their phone. Every day we post a 30-second WhatsApp video walking through what was built that day. The founder either signs off or sends a screenshot with arrows on it.
The exact build sequence:
- Day 4–5: homepage in code (1:1 with Figma).
- Day 6–7: remaining pages — About, Services, Contact, etc.
- Day 8–9: mobile responsiveness across every breakpoint (360, 414, 768, 1024, 1440, 1920).
- Day 10–11: integrations (forms, payments, analytics, email).
- Day 12–13: animations + polish (the 5% of the work that creates 30% of the perceived quality).
- Day 14: SEO setup, structured data, sitemap, Google Search Console submission.
Day 15–18 — final review + content load
Founder reviews the staging site against the original spec. We populate any final copy / images they've been gathering during the build. Run Lighthouse — must hit 90+ on mobile. Run Rich Results Test — must validate. Run accessibility scan — must clear basic WCAG.
Day 19 — launch
We point the founder's domain at Vercel, deploy production, verify analytics fires, submit sitemap to Search Console, send the launch announcement.
Day 20–26 — free post-launch support
Bug fixes, last-mile tweaks, "can we move this section here?" requests — all included in the first 7 days post-launch.
Three principles that make this timeline real
1. "No" is a feature
The single biggest cause of website delays is scope creep — features added mid-build that nobody scoped on day 1. We refuse those in the strongest possible way: anything not on the day-1 scope doc is a change request, with its own quote and its own delivery date. This isn't a billing trick; it's a focus mechanism. It forces the founder to decide what actually matters before we burn time on it.
2. Tight feedback loops > long planning
The most expensive part of any project isn't the work — it's the rework. We catch rework in days, not weeks, by showing work-in-progress every single day. If the founder is going to hate a section, we want them to hate it on Day 4 with 30 minutes invested, not Day 30 with 80 hours invested.
Daily WhatsApp updates aren't a status report. They're a feedback solicitation. Every video ends with: "tell me if this is wrong before we move on."
3. Same stack, every project
We use Next.js + Tailwind + Vercel on every site. Every. Single. One. We've built this stack 50+ times. We know its quirks cold. We have internal helper components, integration recipes, SEO templates ready to copy-paste. The marginal cost of building a new site on a familiar stack is a fraction of building on an unfamiliar one.
Some agencies pride themselves on "we pick the right tech for the job". For under-₹5-lakh projects, this is usually marketing-speak for "we want to bill 2 extra weeks of figuring it out". The right tech for a marketing website in 2026 is Next.js. Period. Decisions made in advance ship faster than decisions made during.
What we don't do (and why)
- Long pre-engagement consulting. A 15-minute scoping call is enough to write a good brief. Multi-meeting workshops at ₹10k/hour are theatre.
- Account managers. Founders talk directly to the people writing the code. Adds 0 overhead, removes a layer of telephone-game.
- Daily standups with the client. A 30-second WhatsApp video is more honest than a Zoom meeting.
- Project tracking software for the client to log in to. Jira / Asana / Monday adds friction for everyone. WhatsApp + a Google Doc is enough.
- Custom Figma "design systems" for ₹15k brochure sites. If the site has 5 pages, a Figma file is enough. Design systems are for products with multiple teams and 100+ screens.
When 3 weeks is the wrong target
Some projects shouldn't be rushed. We explicitly extend the timeline for:
- Custom e-commerce with multi-vendor logic (5–8 weeks).
- Marketplaces with two-sided flows (6–10 weeks).
- SaaS products with billing, accounts, dashboards (8–16 weeks).
- Sites where the founder hasn't decided the brand yet (we pause for a week while they figure it out).
For these, we run the same process — daily updates, tight loops, fixed scope — just over a longer timeline.
If you're scoping a website project — send us a two-line description. We'll come back in 48 hours with a free mockup, a fixed quote, and a delivery date. Start here.
Want this process applied to your website?
Tell us what you need and we'll show you a free mockup within 48 hours. No commitment, no upfront payment.