Case Study

How Shulam Built an SEO Intelligence Layer for Ziv Health

796+ pages. DA 20 to 28. 6 API automations. 4 moats. Compounding weekly.

796+
Pages Published
Across 12 content hubs
DA 28
Domain Authority
Up from DA 20 in 5 weeks
132
Referring Domains
Target: 130+ by May 31
6
API Automations
Keyword to competitive matrix
28
Tests Passing
Full coverage on all services
17
API Endpoints
Admin dashboard + cron jobs

The problem

Ziv Health operates in one of the most competitive organic search landscapes in healthcare — GLP-1 weight loss, testosterone therapy, women's HRT, dermatology, and more. Competitors like Hims, Hers, and Ro spend millions on paid acquisition. Ziv needed to win on organic — but had no system for identifying keyword gaps, tracking rankings, or measuring competitive position.

The SEO work was manual and reactive. Someone would guess which keywords to target, write a page, publish it, and hope. No data on what competitors ranked for. No tracking of position changes. No automated audits. No backlink monitoring. Every decision was a guess.

Ziv needed a system that turns SEO from guesswork into a compounding machine — where every page published makes the next page rank faster, every backlink acquired lifts the entire domain, and every week the gap between Ziv and its competitors widens.

How Shulam built it

Week 1

SEMrush API integration and schema design

Connected SEMrush API (key f89330ec). Designed v097 migration with 6 new tables: keyword_research, ad_intelligence, page_analysis, position_tracking, site_audit, and competitive_matrix. Drizzle ORM, BigInt atomic units, PostgreSQL.

Week 2

Built 6 automated services

KeywordResearchService, AdIntelligenceService, PageAnalysisService, PositionTrackingService, SiteAuditService, and CompetitiveMatrixService. Each runs on its own cron phase (Phases 123-128), stores results in PostgreSQL, and exposes admin API endpoints.

Week 3

Wired the Content Factory flywheel

Keyword gaps feed content briefs. Content briefs become compare pages (6-8 per week in waves). Pages are published, indexed, and tracked. Position tracking monitors rankings. The loop compounds — every page makes the next page rank faster.

Week 4

Backlink Intelligence layer deployed

BacklinkIntelligenceService scans every Sunday at 4am ET (HOSEA auto-run). Toxic backlinks trigger GABRIEL dispatch for immediate disavow. /admin/backlinks/ dashboard shows live referring domain count, DA trend, and toxic link queue.

Week 5

Backlink sprint launched — DA 20 to 28

Targeted high-DA referring domains: Crunchbase (DA 91), LinkedIn (DA 98), and more. DA climbed from 20 to 28 in 5 weeks. Referring domains from 79 to 132. Sprint runs through May 31 with a target of 130+ RDs.

Ongoing

4 moats compounding weekly

M-ZIV-59: Keyword intelligence moat (gap data no competitor has). M-ZIV-60: Content velocity moat (6-8 pages/week). M-ZIV-61: Position tracking moat (daily rank snapshots). M-ZIV-62: Competitive matrix moat (real-time competitor ad spend visibility).

6 API automations

Each automation runs on its own cron schedule, stores results in PostgreSQL, and exposes admin API endpoints. No manual work. No spreadsheets. Data flows automatically.

1

Keyword Research

Phase 123

Pulls keyword gaps, search volumes, and difficulty scores from SEMrush. Identifies high-intent, low-competition terms that map to Ziv Health product lines. Feeds the Content Factory with prioritized briefs.

2

Ad Intelligence

Phase 124

Monitors competitor ad copy, spend estimates, and landing page strategies. Surfaces opportunities where organic content can capture paid traffic at zero marginal cost.

3

Page Analysis

Phase 125

Audits every published page for on-page SEO signals: title tags, meta descriptions, heading structure, internal links, text-to-HTML ratio. Flags pages that need optimization.

4

Position Tracking

Phase 126

Daily rank snapshots for all target keywords. Tracks movement over time. Alerts when a page drops out of the top 20 or enters the top 10 for the first time.

5

Site Audit

Phase 127

Crawls the site for technical SEO issues: broken links, duplicate content, missing alt tags, slow pages, redirect chains. Prioritizes fixes by traffic impact.

6

Competitive Matrix

Phase 128

Maps the competitive landscape across all 12 content hubs. Tracks competitor DA, keyword overlap, content gaps, and backlink profiles. Updated weekly.

The SEO flywheel

The intelligence layer does not just collect data — it drives a compounding loop. Every revolution of the flywheel makes the next revolution faster. Here is how it works for Ziv Health:

1

Keyword gaps

SEMrush identifies terms competitors rank for that Ziv does not

2

Content briefs

Gaps become structured briefs with target keywords, headings, and internal link maps

3

Pages published

Content Factory produces 6-8 compare pages per week in waves across 12 hubs

4

Backlinks acquired

High-DA outreach sprint adds referring domains — Crunchbase, LinkedIn, and more

5

DA climbs

More referring domains and better content push domain authority higher

6

More traffic

Higher DA means existing pages rank better, new pages index faster — the flywheel accelerates

The compounding effect

Ziv publishes 6-8 compare pages per week across 12 hubs. Each page targets a keyword gap identified by the intelligence layer. Each page adds internal links to existing pages, boosting their authority. Each week, the domain gets stronger. This is not linear growth — it is exponential.

The souls behind the system

H

Hosea — Acquisition Intelligence

Soul #38 of 46 in the Ziv Health NLP

Hosea orchestrates all SEO intelligence. Runs the 6 API automations on schedule, monitors keyword rankings daily, triggers backlink scans every Sunday at 4am ET, and dispatches Gabriel when toxic backlinks are detected. Hosea also runs the weekly SEMrush benchmark loop and competitive matrix updates.

KPIs: Organic traffic WoW growth, keyword positions in top 10, referring domain count, content velocity (pages/week), backlink toxicity score.

C

Caleb — Chief Growth Officer

Soul #12 of 46 in the Ziv Health NLP

Caleb owns the growth number. Takes Hosea's intelligence and translates it into growth strategy — which hubs to prioritize, which competitor gaps to exploit, when to shift from organic to paid. Caleb reviews the weekly SEO digest and adjusts the content calendar accordingly.

KPIs: WoW growth rate (target 5%+), CAC payback period, organic vs. paid mix, hub coverage (pages per hub), G-Stack score.

Results so far

796+ pages across 12 hubs

Weight loss (91), women's health (86), female HRT (93), hair loss (81), longevity (81), men's health (81), ED (79), dermatology (78), vaginal health (77), and more. Every page targets a specific keyword gap identified by the intelligence layer. Every page links to zivhealth.com product pages.

DA 20 to 28 in 5 weeks

Domain authority grew 40% through targeted high-DA backlink acquisition. Crunchbase (DA 91), LinkedIn (DA 98), and other authoritative sources. The backlink sprint runs through May 31 with a target of 130+ referring domains.

4 compounding moats

M-ZIV-59 (keyword intelligence), M-ZIV-60 (content velocity), M-ZIV-61 (position tracking), M-ZIV-62 (competitive matrix). Each moat deepens every week. Competitors cannot replicate the data advantage without building the same infrastructure — and by the time they do, Ziv is 6 months ahead.

Zero manual SEO work

Everything runs on cron. Keyword research, position tracking, site audits, competitive analysis, backlink monitoring — all automated. The admin dashboard shows real-time status. Hosea and Caleb handle the rest.

Technical architecture

// SEO Intelligence Layer — stack
SEMrush API → 6 automated services (Phases 123-128)
PostgreSQL → 6 tables (v097 migration, Drizzle ORM)
17 API endpoints → /api/v1/admin/semrush/*
Backlink Intelligence → HOSEA Sunday 4am ET scan
Toxic detection → GABRIEL auto-dispatch for disavow
Content Factory → keyword gaps to published pages
Position Tracking → daily rank snapshots, delta alerts
// Output: 796+ pages, DA 28, 132 RDs, compounding

The intelligence layer runs on Node 22, Express 5, and PostgreSQL — the same stack as the rest of the Ziv Health facilitator. Each of the 6 services is a standalone class with its own cron phase, database table, and API endpoints. All 28 tests pass on every deploy.

The backlink intelligence layer adds a second loop: HOSEA scans referring domains every Sunday, scores them for toxicity, and dispatches GABRIEL to handle disavow files when toxic links are detected. The /admin/backlinks/ dashboard shows live DA trend, referring domain count, and the toxic link queue.

Want an SEO intelligence layer for your platform?

Keyword research, position tracking, backlink monitoring, competitive analysis — all automated, all compounding. Included in every Lighthouse plan.

Apply for Lighthouse →