The right tools don't write content for you. But the wrong tools, or the wrong combination of tools, can slow your team down, create quality problems, and make it impossible to measure what's working. Here's what we've settled on after three years of iteration.
The Four Layers of a Content Stack
A complete content stack has four functional layers: research, production, distribution, and measurement. Most teams have tools for production and distribution. Far fewer have systematized research and measurement, which is exactly where the performance gap comes from.
Layer 1: Research
Keyword research: Ahrefs remains the gold standard for B2B SaaS keyword research. Its keyword difficulty scores are the most calibrated for realistic competitive assessment, and its SERP analysis tools make it easy to understand what's ranking and why. For teams on a tighter budget, Semrush covers the essentials.
Competitor analysis: We use Ahrefs' Content Gap tool to identify keywords competitors rank for that our clients don't. This surfaces quick-win opportunities that are often missed by teams doing keyword research from scratch.
Subject matter expertise: For technical content, the research layer has to go deeper than keyword tools. We use a combination of primary SME interviews (your own engineers and customers), reading primary sources (documentation, release notes, research papers), and hands-on product use. AI tools are useful for accelerating background research, but never as a primary source for technical claims.
Layer 2: Production
CMS: For most SaaS marketing sites, we recommend a headless CMS. Sanity is our preferred choice for new builds. Its flexible schema, real-time collaboration, and developer-friendly API make it well-suited to technical teams. Contentful is the enterprise alternative.
Writing environment: We use Notion for drafts and editorial workflow. The combination of collaborative editing, comments, and database-driven publishing queues works well for multi-writer teams. Linear is becoming popular for teams that want to manage content production alongside product work.
AI-assisted writing: We use Claude (Anthropic) for specific tasks in the writing process: outline generation, summarizing research, generating initial drafts for review-heavy content types, and editing for clarity. We do not use AI to write final copy without heavy human editing. The quality ceiling for AI-generated technical content is below the standard our clients require.
Layer 3: Distribution
SEO infrastructure: Technical SEO is non-negotiable. This means proper sitemaps, canonical tags, structured data (Article schema for blog posts, FAQ schema for appropriate content), hreflang if you're international, and Core Web Vitals optimization. We recommend Screaming Frog for technical audits.
Email: If your content strategy doesn't include email nurturing, you're leaving significant value on the table. Your best content should be repurposed into email sequences that stay in front of prospective buyers between visits.
Social distribution: LinkedIn is the dominant channel for B2B SaaS content distribution. The algorithm currently rewards native documents and carousels more than link posts. Adapt your distribution strategy accordingly.
Layer 4: Measurement
This is where most teams are underinvested. Standard GA4 traffic metrics don't tell you whether content is generating pipeline. You need:
- UTM tracking on all CTA links from content
- Multi-touch attribution in your CRM (HubSpot or Salesforce)
- Scroll depth and engagement time to diagnose content quality
- Content-influenced pipeline reporting: what closed deals had touchpoints on content?
The best content teams review a dashboard that connects organic sessions to trials started to deals influenced. If you can't build that, start with UTMs and first-touch attribution and work backward.
The Honest Assessment
No stack replaces judgment. Tools surface data; your team has to interpret it and make decisions. The companies that produce consistently excellent technical content have one thing in common: they invest in the people and processes before the tools.
Buy the tools that remove friction from the parts of your workflow that are already working. Don't buy tools hoping they'll fix a broken process.