We rebuilt our entire website using AI coding agents. The team: a graphic designer to support, myself, Cursor and Antigravity.
What would have taken 4–6 months and a team of specialists… took us about 66 hours, spread across 7 weeks.
I'm not talking about a few landing pages, but a full site:
- 15+ core pages
- 4 interactive 3D rendering scenes
- a bilingual glossary with 148 terms, and longer 'explore' pages.
- a blog system with categories, tags and search functions
- SEO architecture (~300 indexed URLs) (38x impressions and 5X clicks in the first 3 weeks in Search console)
- analytics and integrations (Hubspot, Apollo, Genesy, LinkedIn)
No team of specialists and no formal project structure or management tools (ASANA, Jira, etc) to coordinate tasks.
Now it is done, I wanted to understand what this actually meant in terms of speed and cost savings.
So I asked an LLM to review all the code and answer the question:
“What would this have required 1-2 years ago, without AI?”
What This Would Have Required Before AI
Team Composition
| Role | Why it's needed |
|---|---|
| Project Manager | Coordination, timelines, stakeholder alignment |
| UI/UX Designer | Wireframes, design system |
| Frontend Developer | HTML, CSS, JS |
| WebGL / Three.js Specialist | 3D scenes (niche skill) |
| Backend / DevOps | Build and deployment |
| SEO Specialist | Sitemap, structured data |
| Content Writer / Translator | 148 glossary terms EN/ES |
| QA Tester | Testing and debugging |
Total: 6–8 specialists involved
Effort Breakdown
| Task | Hours |
|---|---|
| UI/UX design | 40-60 |
| Frontend (15+ pages) | 80-120 |
| JavaScript interactions | 30-40 |
| 3D (Three.js scenes) | 40-60 |
| Glossary system | 50-70 |
| Blog system | 20-30 |
| Dynamic pages | 15-20 |
| Forms and integrations | 10-15 |
| SEO setup | 10-15 |
| Build and deployment | 15-25 |
| QA and debugging | 30-45 |
| Maintenance | 10-15 |
Total: 350–500+ hours
Cost Estimate (Corrected)
| Scenario | Rate | Total |
|---|---|---|
| Freelance specialists (end-to-end) | $80-120/hr | $28K-$62K |
| Small team setup | $150-250/hr | $52K-$130K |
Key cost driver (included above):
| Component | Rate | Cost |
|---|---|---|
| 3D / WebGL specialist | $150-200/hr | $6K-$12K |
Realistic total: $40K – $80K
Timeline
| Phase | Duration |
|---|---|
| Discovery and scoping | 1-2 weeks |
| Design | 3-4 weeks |
| Frontend build | 4-6 weeks |
| 3D development | 3-4 weeks |
| Glossary and content | 4-6 weeks |
| Blog and dynamic pages | 1-2 weeks |
| SEO and integrations | 1-2 weeks |
| QA and testing | 2-3 weeks |
| Deployment and fixes | 2-4 weeks |
Total: 4–6 months (realistic)
My Actual Investment
Now let’s compare that with a realistic estimate of what we actually invested:
- Time invested: 2 hours/day × 33 days = 66 hours
- CMO opportunity cost: €100/hour → €6,600
- Graphic designer cost: €5,000
Total: €11,600
Direct Comparison
| Metric | Traditional model | This project | Delta |
|---|---|---|---|
| Time (hours) | 350-500 | 66 | -80% to -87% |
| Calendar time | 4-6 months | 1.5 months (part-time) | -70%+ |
| Cost | €37K-€74K | €11.6K | -69% to -84% |
Clear Summary: What Changed
This wasn’t just faster execution. It was a different system.
What disappeared:
- coordination across roles
- project management tools
- handoffs
- waiting time
- context loss
What replaced it:
- direct iteration
- continuous feedback
- full context in one loop
- CMO in direct charge of all messaging
What Still Matters (More Than Ever)
After removing execution friction, the leverage shifts entirely. What remains critical is not production, It’s your capability as a CMO:
- positioning
- differentiation
- messaging
- strategic SEO
Because AI will build what you ask it to build, and the quality of that output is directly tied to the clarity of your thinking.
My Take
Execution is no longer the bottleneck. AI doesn’t replace strategy, it accelerates it and amplifies it by removing friction between thinking and building.
But the deeper shift is structural.
What previously required multiple specialists across design, development, SEO, content and operations… can now be executed by one full-stack marketing expert supported by AI.
Profiles like a CMO — with strong foundations in:
- positioning
- messaging
- SEO
- growth strategy
can now move directly from idea to execution without relying on multiple layers of specialists.
This doesn’t mean expertise disappears.
But it does mean the way expertise is applied is changing.
Instead of fragmented roles executing isolated tasks, we are moving toward integrated operators who orchestrate outcomes end-to-end, with AI handling a large part of the production layer.
That has clear implications:
- Many execution-heavy specialist roles will need to evolve
- Some will become less central to how marketing operates
- Others will shift up the stack toward strategy, differentiation, and complex problem-solving
The distance between idea → execution has collapsed.
And when that happens:
- iteration speeds up
- learning speeds up
- outcomes improve faster
When execution collapses from months to hours, the constraint moves.
Not to tools. Not to budget. To judgment.
- What do we build?
- Why does it matter?
- How do we win?
In this new reality, advantage belongs to those who can think across disciplines, and turn that thinking into execution instantly.