Video summary

Comment Créer & Vendre Un Site Web à 10'000€ Avec Claude

Main summary

Key takeaways

Business

Business strategy: sell the “system,” not the website

  • Positioning: A €10,000 site becomes feasible only when you sell a repeatable acquisition + conversion system that gets clients to say “yes” before building.
  • Core idea: A free “homepage redesign” leads to a yes vote, then a prototype call, and finally a paid implementation of a full customer-generating stack (forms/automation/CRM/agent/booking), not “a one-off build.”

Playbook / step-by-step process (copyable method)

Step 1 — Define a target niche (pick 2–3, test immediately)

  • Avoid targeting “everyone” (generic outreach gets no response).

  • Choose niches using 4 criteria:

    1. Immediate “heroic effect”: the website is an acquisition channel (saves time/money or drives contracts), not just a business card.
    2. High LTV potential: the niche’s new customer value is high enough to justify a €10k web offer.
    3. Digitized market: prospects/services exist online; email/prospecting works.
      • Example contrast: “plumber/locksmith style” doesn’t prospect well via email.
    4. Low regulatory constraints + fewer internal validation bottlenecks: avoid banking, insurance, healthcare, public sector, very large companies.

  • Example niche selected for the demo: high-end residential renovation (projects often represent tens of thousands of euros, so €10k is rational).

  • Tactic: start with three niches to reduce risk; if one fails, you have backups.

Step 2 — Build a lead list fast (automation, not manual scraping)

  • Lead sources by channel fit:

    • Google Maps = best entry point for physical businesses (fast list building).
    • LinkedIn = good for decision-makers + professional emails (but less presence for physical niches; treat as backup).
    • Instagram = quality discovery via hashtags; low volume, manual approach (backup).

  • Tool + workflow example (Google Maps scraper):

    • Use a scraper (explicitly mentions Appify) to extract from Google Maps:
      • Fields: business name, website, phone, and sometimes email (if found on website).
      • Deduplication handled automatically.
    • Budget note:
      • Appify includes ~$5 free credit/month, sufficient for up to ~1,000 leads with emails (per description).

  • Operational safeguards:

    • B2B only (prospecting professionals, not individuals).
    • Don’t scrape too big at once: start with ~100 prospects, run end-to-end, then scale.

Step 3 — Qualify + filter for “golden prospects”

  • Don’t email the entire raw list.

  • Define “golden” as: great business work but a bad website (outdated, broken, missing, or non-performing).

  • Exclude:

    • Off-target results (that slipped into search)
    • Prospects whose sites are already modern/efficient (no “wow” transformation)

  • Automated qualification (Claude-based script; described tooling):

    • Script checks:
      • Whether the site exists
      • Whether it’s modern or outdated
    • For borderline cases, it evaluates in:
      • desktop mode
      • mobile mode
    • Outputs:
      • prospect.csv (qualified prospects), categorized (e.g., outdated sites / broken sites / no site)
      • discarded.csv (out-of-target or already-good sites)
    • Adds fields for personalization, including:
      • opportunity score
      • normalized company name

Step 4 — Messaging: “offer contrast,” not “sell a website”

  • Avoid generic cold emails (example: one got deleted in 2 seconds due to bulk generic outreach).

  • Strategy: reverse the sales dynamic with a risk-free value test.

  • Primary cold email concept:

    • “I can redesign your homepage for free just to show you. You keep it; no price/commitment/link.”

  • Four-line email structure (must answer 3 questions in < 5 seconds):

    1. Specific observation (e.g., slow mobile, missing quote-request path)
    2. Legitimacy in one sentence (e.g., “I design websites and automation solutions for this profession.”)
    3. Clear free offer
    4. Frictionless CTA to accept (prepare/send a prototype, ask to reply/agree)

  • Personalization fields:

    • Use normalized company name, city, and profession/category (from qualified CSV).
  • Conditional variants:
    • If the site is broken: mention the error page
    • If no site: mention inability to find a website

Step 5 — Close the deal using a “Yes vote” prototype

  • Rule: Only build after the prospect says “yes.”

  • Sales flow:

    1. Free homepage redesign
    2. Propose a meeting
    3. Present prototype
    4. Upsell the full system (SEO, tracking, CRM integration, chatbot, automated email sequences, etc.)

  • Value framing:

    • The “€10k” is achieved via the system stack, not the free homepage.

Step 6 — Email operations to avoid deliverability damage

  • Don’t send from your main domain.

  • Two non-negotiable safeguards:

    1. Validate emails before sending (target <5% bounce rate)
    2. Never send from the main business domain
      • Use secondary prospecting domains; if one overheats, pause it so the main domain remains safe.

  • Example platform: Instantly

    • Configure campaign
    • Import qualified leads CSV
    • Map variables (company name, city, category, etc.)
    • Verify leads in platform
    • Create sequence emails + variables
    • No links in signature (per instructions)
    • Avoid spam flags

  • Sequence cadence:

    • 1 main email + two reminders
      • Reminder 1 after 3 days
      • Reminder 2 after 4 additional days (i.e., 7 days after the first email)

  • Throttling / ramp-up:

    • Start low (e.g., 5–10 emails/day/mailbox)
    • Ramp to ~30/day/mailbox
    • Beyond ~30/day increases blacklist risk
    • Tool manages gradual ramp and notifies on replies

Step 7 — Build the “wow” mockup in <15 minutes with Claude

  • Delivery promise:

    • Compared to prior multi-role workflows (designer + dev + weeks), Claude enables:
    • Before/after prototype in < 15 minutes, even without coding.

  • Example generation flow:

    • Extract current site assets (photos/colors/content) using Claude tool (mentions “refonte extraction”)

    • Correct hallucination mismatches (example: wrong primary color chosen initially; user corrects to site’s green)

    • Generate redesign direction (“refonte design”) and page output (“refonte bild”)

  • Critical quality rule:

    • Tool does not invent fake photos/reviews/numbers
    • Missing info creates placeholders (“holes”) that prompt what to ask the client for

Pricing model + revenue targets (explicit numbers)

Staircase model (entry → recurring revenue)

  1. Step 1 (Free model): free homepage redesign (entry / wow)
  2. Step 2: homepage offered / loss leader (includes key pages)

  3. Step 3 (System implementation):

    • Typical ticket: €4,000 (forms + automation + agent + appointment booking + CRM + tracking)

  4. Step 4 (Recurring):

    • Typically €500/month for hosting, monitoring, continuous improvement, follow-ups, adjustments

Year-1 math (given)

  • €4,000 setup fee + 12 × €500 = €10,000 in first year

Positioning outcome

  • One-off websites force “start from scratch” each time.
  • Recurring system builds ongoing acquisition + lead-handling improvements.

Concrete operational recommendations (actionable)

  • Choose niches where the website has measurable acquisition impact (“heroic effect”).
  • Use automation to avoid burnout:
    • Scrape → dedupe → CSV export → automated qualification.
  • Only outreach to qualified “golden prospects.”
  • Email should be a contrast offer with no risk (“free redesign keeps it”).
  • Protect deliverability:
    • Email validation to keep bounce rate <5%
    • Secondary domains for prospecting
    • Throttle sending to avoid blacklist risk.
  • Prototype rule:
    • Wait for “yes” before building (no free work for non-responders).
  • Deliver as an ongoing system with controlled hosting/ops (avoid “gift package” handoff that breaks relationship continuity).

Key metrics / KPIs mentioned

  • Lead volume: up to ~1,000 leads with email using the scraper/free credit.
  • Qualification start sample: ~100 prospects before scaling.
  • Email deliverability: bounce rate target <5%
  • Email follow-ups: after 3 days and 4 days later (7 days total)
  • Sending throttle: ramp to ~30 emails/day/mailbox as an upper limit
  • Time-to-mockup: <15 minutes for Claude-generated before/after prototype
  • Revenue targets:
    • Step 3: ~€4,000
    • Step 4: ~€500/month
    • First year: ~€10,000 (= €4,000 + 12 × €500)
  • LTV criterion: numeric target not clearly specified, but the decision rule is to ensure customer value supports the offer (example references €10,300 / €5,000+-type framing; exact formatting unclear).

Tools/platforms referenced

  • Appify (Google Maps scrapers)
  • Claude (site extraction, redesign mockup, qualification assistance)
  • Cloud Code / VS Code-like workflow (runs scripts/commands)
  • Instantly (email outreach sequences, ramp-up, validation/verification)
  • Versel (hosting for subscription model)
  • Email validation tools (generic mention; also dedicated validators like “email…” in the text)

Presenters / sources

  • Presenter: The unnamed creator of the video (speaking in first person; references their system, coaching program, and coaching member “Badre”).
  • No external written sources beyond tools/platforms: Claude, Appify, Instantly, Versel (mentioned as tools).

Original video