Skip to content

Content Marketing Works For Us

Technical content about edge computing and agent infrastructure will generate qualified leads for our products.

The Assumption

Our PLG strategy depends on content marketing driving top-of-funnel awareness. We’re betting that:

  1. We can write compelling technical content
  2. There’s an audience for edge computing and agent infrastructure content
  3. That audience overlaps with potential buyers
  4. Content → signups → paid conversion funnel works

None of these are guaranteed. Many technical founders underestimate content effort or overestimate their audience.

Evidence

Developer content success stories:

  • Vercel, Netlify built massive audiences through technical blogs
  • Supabase grew through content + community
  • Cloudflare’s blog drives significant developer awareness
  • Individual developers (Tanner Linsley, Theo) built audiences that drive product adoption

Content-market fit signals:

  • “Edge computing” and “AI agents” are trending topics
  • Developer Twitter/LinkedIn engagement on these topics high
  • Podcasts, newsletters emerging in the space
  • Conference talks on these topics well-attended

Counter-signals:

  • Content creation takes time away from product development
  • Audience building is slow (months, not weeks)
  • Most content efforts fail quietly
  • Competition for attention is fierce

Counter-Evidence

What would prove this wrong:

  • 6 months of publishing with fewer than 50 followers
  • Zero conversions from content
  • Engagement metrics consistently poor (under 10 likes/post)
  • Time spent on content not translating to signups

Warning signs:

  • Posts get no engagement
  • Writing feels like a chore, quality suffers
  • Audience grows but doesn’t convert
  • Time spent on content delays product shipping

Impact If Wrong

Products affected: All products (content drives awareness for everything)

GTM impact:

  • Would need paid acquisition (higher CAC)
  • Would need partnerships/integrations for distribution
  • Would need to find alternative channels (open source, communities)

Runway impact:

  • Paid acquisition burns cash faster
  • Longer time to meaningful revenue
  • May need to raise earlier

Alternative strategies:

  • Product Hunt launches
  • Integration partnerships (Vercel, Supabase, etc.)
  • Open source components that drive adoption
  • Community building in existing spaces (Discord, Reddit)

Testing Plan

Content experiment (3 months):

  1. Publish 1 substantive technical post per week
  2. Track followers, engagement, website traffic
  3. Track signups attributed to content
  4. Measure time spent vs. results

Metrics to track:

  • Followers: Target 100+ in 3 months
  • Engagement: Target 25+ meaningful interactions per post
  • Website traffic: Track content → product page flow
  • Signups: Track referral source

Content types to test:

  • Tutorial content (how to build X with Cloudflare Workers)
  • Opinion content (why agents need sandboxes)
  • Technical deep dives (edge computing patterns)
  • Build-in-public updates (progress on SmartBoxes)

Timeline: 3 months to initial signal, 6 months to clear verdict

Kill criteria: If 6 months of publishing with fewer than 50 followers and fewer than 5 signups from content, deprioritise content marketing.

Depends on:

Supports:

Competes for time with:

Addresses constraint:

  • No Audience — content is the solution to this constraint

Proxy metrics:

Assumption

Technical content about edge computing and agent infrastructure will generate qualified leads for our products.

Depends On

This assumption only matters if these are true:

How To Test

Publish content; track engagement, followers, and conversion to signups.

Validation Criteria

This assumption is validated if:

  • 100+ engaged followers from content
  • 10%+ of signups cite content as discovery source
  • Content generates inbound inquiries

Invalidation Criteria

This assumption is invalidated if:

  • 6 months of publishing with fewer than 50 followers
  • Zero conversions from content
  • Engagement metrics consistently poor