SwissOfferten

Reach & Rankings

Four months after launching in March 2026, SwissOfferten's organic footprint sits at 149,000+ Google impressions in the last 90 days alone, across 6,000+ unique search queries and 2,500+ indexed pages, and it is still climbing steeply. Ranking at national scale is the hard part of local visibility, and it is already done.

SwissOfferten ranks #1-2 in Switzerland for "stundenlohn putzfrau schweiz 2026", on page one for "reinigungsofferten vergleichen", and across a wide range of natural-language, question-style searches, the way real buyers (and AI assistants) actually phrase them.

When AI answers, it cites us

Search is collapsing into answers. People stop scrolling ten blue links and ask an assistant. Ask ChatGPT, Perplexity, Gemini, Claude or Copilot how to compare cleaning quotes in Switzerland, and SwissOfferten comes back as a named, cited source, standing next to the largest national portals. Not on one assistant, but across every major one, even Qwant. Most businesses are invisible to AI. We are inside the answer it gives.

It is the hardest position in modern marketing to win, and the hardest to copy. It rests on Schema.org structured data, a machine-readable OpenAPI endpoint assistants read directly, and page-one rankings on the natural-language questions they pull from. Pointed at a single business, the same engine makes that business the answer its customers get.

The Problem

Cleaning customers in Switzerland face a fragmented market. Dozens of one-off company websites, no central price comparison, slow callbacks losing leads to whoever responds first, and language barriers across DE / FR / IT regions. By the time a homeowner has compared three quotes, two of the firms have stopped responding.

Cleaning firms have the inverse problem. They pay for Google Ads they can't measure, get unqualified inbound from generic forms, and have no scalable way to expand into adjacent regions without spinning up new websites or paid campaigns per region.

The Solution

A two-sided marketplace built on Next.js 16 + Supabase. Customers submit a 4-step request (service type → region → property size → contact). The request is validated server-side, honeypot-filtered, rate-limited via Upstash Redis (5/min per IP), and instantly visible to qualified partner firms in that region.

Partners load CHF credit through Stripe and purchase the lead for CHF 30-60. Every transaction is Row Level Security (RLS) protected at the Postgres layer, with no direct-query bypass path, even from server code. Every form submission is server-action validated with Zod schemas. Every payment is webhook-reconciled.

The result: customers get multiple qualified quotes in minutes instead of days. Firms get leads that match their service area, pre-filtered for intent, with full audit trails of what they purchased and what closed.

Tech Architecture

• Next.js 16 App Router: Server Components by default, client-side rendering only where interactivity demands it
• Supabase: Postgres + Auth + RLS on every table. Zero unprotected query paths.
• Stripe (CHF): Credit top-ups via Checkout + webhook-driven reconciliation. No race conditions on credit balance.
• Upstash Redis: Distributed rate limiting across edge regions (5/min lead submit, 3/min checkout, 3/5min password reset)
• Resend + React Email: Type-safe transactional templates, multilingual
• Vercel: Edge hosting, automatic preview environments per PR
• Playwright E2E: End-to-end coverage on critical paths (lead submission, partner auth, checkout, navigation)

Coverage

Four languages live: DE, FR, EN, IT, the three federal languages of Switzerland plus English for expat customers. Regional landing pages span 26 cantons with proper i18n routing and locale-specific content.

Every page renders with Schema.org structured data (LocalBusiness, Service, BreadcrumbList) for traditional SEO. GA4 analytics is consent-gated through a cookie banner, no tracking before opt-in.

Hardening

Production-grade security baseline: CSP, HSTS, COOP headers on every response. Honeypot fields on all public forms. Server-side validation with Zod on every action. Rate limiting on every state-mutating endpoint. RLS policies tested in CI. Test mode versus live mode separation for Stripe with no cross-contamination paths.

Results

• 70+ verified cleaning firms onboarded across Swiss cantons, growing weekly
• 4-language coverage (DE / FR / EN / IT) live in production
• 203 canonical SEO landing pages across cantons and service types, curated for quality after a thorough content audit
• Live Stripe CHF payments with webhook-driven credit reconciliation, paying partner firms across Switzerland, zero unmatched transactions
• 100% RLS coverage on database tables, zero direct-query bypass paths
• Production-hardened: CSP, HSTS, COOP, honeypots, rate limits, server-side validation, automated E2E tests

Why this matters for a single business

Ranking a marketplace across 26 cantons is harder than getting one shop found in its own town. Pointed at a single local business, the same engine means one thing: you are the first result your neighbours find, and every inquiry reaches you in seconds instead of leaking to a faster competitor. Found first plus captured first equals more local jobs.