Name: Lia Golledge Company: MagicBlocks Function: Growth
Date: September 2024 – May 2026

TL;DR

When MagicBlocks set aggressive new traffic and revenue targets, the growth function was a team of one. Rather than hiring locally at scale, Lia Golledge built a lean 4-person remote team in Indonesia and paired them with a custom AI toolchain — Custom GPTs, Claude Projects, and Claude Skills — designed specifically for MagicBlocks' content workflows. The result: 20+ blog posts per month, 5 pieces of content per week, and a growth engine that scaled output without scaling headcount or cost.

Key Outcomes

• 20+ blog posts published per month (up from what one person could manage alone)
• 5 pieces of content produced per week across formats
• 4-person remote team hired (content and creatives, SEO/AEO specialist, research and outbound, and marketing assistant), onboarded, and operating at full capacity
• Custom AI toolchain built across GPTs, Claude Projects, and Claude Skills
• Compliance checking automated for regulated-industry content via Claude Skills
• Content engine supporting +938% traffic growth and +1,696% Google traffic increase over 6 months

The Design Thinking Journey

1. Empathise — Understanding the Problem

MagicBlocks set new targets: more traffic, more leads, more revenue. The content and growth function had one person running it — Lia. Everything that existed had been built by her: the content strategy, the SEO approach, the AEO experiments, the publishing workflow. It was working. But it wasn't scalable.

The core tension wasn't quality — it was throughput. A single operator can think strategically and execute tactically, but not both at volume simultaneously. To hit the new targets, output needed to multiply. The question was how to do that without a large local team budget and without sacrificing the quality and consistency that had driven early results.

There was also a second problem underneath the first: the workflows hadn't been systematised. They lived in Lia's head. Bringing anyone new in — remote or local — would require building those systems from scratch anyway.

2. Define — Framing the Right Problem

The temptation was to frame this as a hiring problem. It wasn't. The real problem was: how do you multiply output without multiplying cost or complexity?