How IT For Prof Deploys WordPress Case Studies in 90 Minutes

The Problem

One content manager. Ten WordPress sites on Elementor. Target: 15 posts per month. Reality: 8-10 posts, because each takes 8-12 hours.

Each case study post has 14 Elementor containers and 54 widgets: hero section, stats cards, problem description, comparison table, step-by-step accordion, counter metrics, FAQ with schema, two CTA blocks.

The manual process: write in Google Docs, open Elementor, build 14 sections from scratch, drag 54 widgets, paste text into each, find stock images, crop, convert to WebP, upload, configure Rank Math SEO, check mobile, fix, publish.

Hiring a second content manager: +80,000 RUB/month (~€750). Not in budget.

"54 widgets per post. Each one filled by hand. Drag, paste, upload, repeat. On 10 sites, it's a conveyor belt one person can't keep running."

— Konstantin Tyutyunnik, IT For Prof

The Solution

IT For Prof built an automated pipeline: Claude Code writes content, Respira deploys it to WordPress, AI Engine generates images.

The stack

• Claude Code — AI agent from Anthropic. Writes drafts, structures content, suggests keywords. Human reviews and edits.
• Respira for WordPress — MCP server that connects AI directly to WordPress. Reads page structure, updates Elementor widgets, runs SEO analysis. 88 tools, 11 builders supported.
• AI Engine — Free WordPress plugin with MCP server for image generation. Creates images from text prompts, verifies with AI Vision.
• Respira Skills — Post-publish optimization: SEO & AEO Amplifier, AI Image Optimizer, Internal Link Builder.

The key tool: wordpress_update_module

Instead of manually filling each widget, Respira extracts the template structure and fills each widget programmatically:

// Extract template structure
const structure = await extract_builder_content({
  builder: "elementor",
  pageId: 23020
});

// Fill each widget by path
await wordpress_update_module({
  builder: "elementor",
  pageId: 23020,
  editTarget: "live",
  moduleIdentifier: { path: "[6].elements[1]" },
  updates: {
    content: "<p>Solution section with <a href='...'>links</a></p>"
  }
});

One call per widget. 54 widgets filled in ~40 minutes. No manual Elementor editing.

Post-deployment analysis

// SEO check
await wordpress_analyze_seo({ pageId: 23020 });

// Readability score
await wordpress_analyze_readability({ pageId: 23020 });

// Schema verification
await wordpress_check_structured_data({ pageId: 23020 });

Before / After

The Workflow

Real Results

The first post created through the pipeline: TrueNAS + Proxmox cluster case study.

Rank Math 81/100 on first pass—something that rarely happens with manual publishing where forgotten meta tags and missing alt texts drop the score to 65-75.

Bug Report & Support

During their work, IT For Prof discovered two issues:

• apply_builder_patch returns success: true but doesn't persist changes to _elementor_data
• find_builder_targets with query parameter always returns 0 matches

They prepared a detailed bug report with minimal reproducible steps and sent it to me. I acknowledged the issue within hours.

"This level of responsiveness from a solo developer is remarkable—many enterprise SaaS products don't respond that fast."

— Konstantin Tyutyunnik

The workaround: wordpress_update_module with path-based identifiers works perfectly and became the foundation of their pipeline.

Note: These issues are being addressed in Respira 4.2.x.

Why Respira Works for Agencies

IT For Prof is now offering this pipeline as a service to digital agencies managing multiple WordPress sites.

The economics at Agency tier (€199/year for 20 sites):

• 1 content manager manually: 8-10 posts/month
• 1 content manager + Respira pipeline: 25-30 posts/month (1hr review/day)
• Dedicated reviewer + AI pipeline 24/7: 80-100+ posts/month

The 11-builder support means they're not locked into Elementor—clients on Divi, Gutenberg, or WPBakery can use the same pipeline.