MAKO

The problem

MAKO is the two-person studio I co-run with Manhit Kohli - he runs marketing and campaigns, I build the websites and systems - and until this build it had no site of its own. The work was real (Reportcards' site health climbing 69 to 88, Arun Silver's 303-product bilingual catalogue, a client's Instagram reels pulling millions of views) but none of it lived anywhere a prospect could see it, and the domain itself just showed a placeholder "coming soon" page.

What it does

The public site for MAKO, the two-person build-and-performance studio I co-run with Manhit Kohli. Five plain pages - home, work, services, studio, contact - sharing one CSS/JS bundle, deployed to Hostinger with no framework and no build step, matching the zero-marginal-cost approach the studio sells to clients.

Fixed website pricing on the page, marketing scoped on a call, and a case-study section built from a client's real Instagram and Meta Ads exports rather than agency-speak. Every CTA is one WhatsApp link, prefilled, no form and no bot in the loop.

Why it is built this way

Built as a plain multi-page static site - one shared CSS/JS bundle across home, work, services, studio and contact, no framework, matching the studio's own zero-marginal-cost philosophy. Services carry fixed prices printed on the page instead of "get a quote," and every call-to-action is a WhatsApp deep link prefilled with a "My business:" starter message, because the studio's whole pitch is no account managers and no handoffs. Deployed to a hidden `/preview/` path behind HTTP Basic Auth first, checked there, then promoted to the live root once it held up - so the real client-facing domain never showed a half-finished page.

What was hard

The studio's own case-study numbers did not exist as a clean dataset; they were scattered across a client's Instagram and Meta Ads exports in one portfolio zip. Mining that turned an "agency claims" page into one with attributable receipts: six reels for one client totalling roughly 12M views and 8.6M accounts reached, and four August-2026 Meta ad campaigns spending about ₹36.5k for 297 leads - one of them landing 53% under the Jaipur benchmark cost-per-lead. And going from a placeholder straight to a live public domain meant the boring infrastructure had to be right the first time: forced HTTPS and www redirects, a real 404 page, and Privacy/Terms pages rewritten for the studio's actual stack (no analytics, no form, Google Fonts disclosed) instead of template boilerplate.

Outcomes

• ~12M reel views cited
• 297 leads, ~₹36.5k spend
• Fixed prices, no quote form

Stack