Content marketing that gets read AND cited.
SERP-driven briefs. AI-drafted, human-edited. Structured for AI search citation in ChatGPT, Perplexity, and AI Overviews. Linked into your existing site for topical authority compounding. From $2,000/mo.
Content marketing in 2026, what changed.
The job of content marketing changed in 2024-2025 and most agencies have not caught up. Three shifts: AI Overviews now sit above the 10 blue links for 40-60 percent of informational queries in B2B and home-services categories; ChatGPT, Perplexity, and Claude pull from a different citation graph than Google's classic ranking signals; and helpful-content scoring keeps de-ranking generic templated work regardless of whether a human or a model wrote it. The playbook that still works for Houston SMBs: SERP analysis on every brief, AI drafting for speed, human SME editing for trust signals, pillar-and-cluster architecture for topical authority, passage-level structure for AI citation, FAQPage + Article schema on every post, IndexNow on publish, and ruthless pruning at 90 days. Cadence: 4 posts/mo floor, 8/mo sweet spot, 12-16/mo for category leadership. Investment: $2K-$11K/mo. Results: brand and bottom-funnel terms in 30-45 days, category terms at 90-180 days, compounding at 6-12 months.
The full service.
1. Topical strategy + keyword mapping
Before any post gets written, we map the topical territory you can credibly own. Most SMB content programs fail not because the writing is bad, but because the strategy is unfocused. Forty posts spread across twelve unrelated topics produces zero authority signals. Twenty posts inside one topical cluster produces a dominant position.
- Topical territory definition: the 2-4 areas where you can credibly publish 20+ posts each
- Keyword universe build: 200-500 target terms grouped by topical cluster and intent (informational, commercial, navigational, transactional)
- Difficulty + opportunity scoring per term (search volume, KD, competitor strength, intent match)
- Pillar + cluster architecture: 1 pillar page + 6-10 cluster posts per territory
- Cannibalization audit: existing pages competing for the same intent get merged or redirected
- 12-month editorial calendar with phased rollout (foundation posts first, expansion posts as authority builds)
2. SERP-driven brief production
Every single post starts with a live SERP analysis. The 10 ranking results plus the AI Overview tell us what Google considers the right shape for that intent. We do not write from a template; we write to fit the gap in the current SERP.
- Top 10 organic results pulled live, content depth and structure analyzed
- AI Overview composition checked: which sources got cited, which passages were lifted
- People Also Ask + related searches captured for FAQ targeting
- Word count target derived from competitive median (not arbitrary "1,500 words" rules)
- Heading structure recommendation: H2s mapped to dominant search intents
- Internal link map: which existing pages should link in, which should link out
- Schema recommendation per post (Article + FAQPage minimum, HowTo / Product / Recipe / Course as appropriate)
3. AI drafting in your voice
Drafting is where AI earns its keep. A capable model with a tight brief and a locked voice produces first drafts in minutes. The bottleneck moves to editing, which is the right place for the bottleneck.
- Voice samples locked during onboarding (10-15 of your best existing posts plus 2-3 SME interviews transcribed)
- Brief + voice profile feed into Claude or GPT-5 for first draft
- Tone, sentence rhythm, jargon level, and POV all controlled at prompt level
- Brand-banned-words list enforced (no em dashes, no "in today's fast-paced world," no "delve")
- Voice similarity score against samples before draft moves to editing (0-100, target 85+)
- Pull quotes and data callouts pre-marked for the editor
4. Human SME editing (the trust layer)
This is what separates our content from a content mill. Every draft passes through a subject-matter editor who adds original data, real examples, contrarian POV where appropriate, and the lived-experience details that AI cannot fabricate.
- SME editing pass: original data, internal numbers, customer-story anecdotes, contrarian POV where defensible
- Generic-AI-residue cut: hedged phrasing, "it is important to note," empty transitions, unnecessary qualifiers
- Reader-finish check: would a reader of this post be smarter at the end than the start?
- Fact-check pass on any stat, percentage, or named source
- Linking pass: every post gets 3-6 internal links to relevant existing pages plus 1-3 outbound citations to authoritative sources
- Author byline with real bio, link to author page, knowsAbout schema for E-E-A-T
5. AI-search-ready structure (GEO)
Generative Engine Optimization is the new layer. ChatGPT, Perplexity, Claude, and Google AI Overviews pull short passages, definitions, and lists into their answers. Posts structured for passage-level citation get cited; posts structured as long-flowing-prose do not.
- Definition sentences at the top of each section (one declarative sentence answering the implied question)
- Numbered lists for procedural content (AI Overviews love numbered steps)
- Comparison tables for vs-style content (AI extractors lift tables wholesale)
- Stat callouts with source attribution inline (citable units)
- TL;DR or summary block at top (often lifted as the answer paragraph)
- FAQPage schema on every post (the most reliably-cited shape in AI Overviews)
- llms.txt at apex listing canonical URLs for AI crawlers
6. Schema, on-page, and publishing
The technical layer most agencies skip. Without schema and on-page polish, even great content underperforms in search and AI Overviews.
- Article schema on every post (headline, author, datePublished, dateModified, image, publisher)
- FAQPage schema for the post's FAQ block
- BreadcrumbList schema for in-section posts
- HowTo schema for procedural content (when appropriate)
- Title tag and meta description optimized against the live SERP (not generic templates)
- Featured image generated or selected, alt text written, OG image rendered
- Internal link insertion in the right anchors (not "click here," real keyword anchors)
- IndexNow submission on publish (Bing, Yandex, Seznam, Naver) for fast indexation
- GSC submission backup for Google indexation
7. Performance review + pruning at 90 days
Most agencies publish and forget. We measure at 90 days and prune. The bottom 20 percent of any cohort either gets rewritten with better intent match or de-indexed. Bloat is the enemy of topical authority.
- 90-day cohort review: impressions, clicks, average position, AI Overview citation status per post
- Top 20 percent: identify what worked and replicate the pattern in upcoming posts
- Middle 60 percent: minor edits, schema fixes, internal link additions, image refresh
- Bottom 20 percent: full rewrite or noindex + 301 to a stronger sibling
- Internal link audit: posts gaining authority get linked to from older posts
- Quarterly topical-cluster review: is the territory consolidating or fragmenting?
Cadence tiers + investment.
| Tier | Posts / month | Right for | Investment |
|---|---|---|---|
| Foundation | 4 | SMBs with a thin blog (under 30 posts) or no consistent publishing rhythm | $2,000-$3,000/mo |
| Growth | 8 | SMBs in competitive niches actively building topical authority, 6-12 mo horizon | $4,000-$6,000/mo |
| Dominance | 12-16 | SMBs going for category leadership in one vertical, 18 mo+ horizon, budget to sustain | $7,000-$11,000/mo |
| Pillar pages | Add-on | Definitive 3,000-6,000w pages, 1-3 per quarter | $1,500-$2,500 each |
| Existing content audit | One-time | 30+ existing posts that need triage (rewrite / merge / prune) | $2,500-$5,000 |
Pricing includes SERP analysis, briefs, AI drafting, human SME editing, schema markup, on-page optimization, publishing, and IndexNow submission. CMS access required (WordPress, Webflow, Shopify, custom). Featured image generation included; commissioned illustration billed separately.
Real outcomes (sanitized).
Houston home-services SMB, 8 posts/mo for 9 months
- Starting state: 12 thin blog posts, ranking for 24 keywords, 80 organic clicks/mo
- Strategy: 2 topical clusters (one core service, one adjacent service), pillar + 16 cluster posts
- 9-month outcome: ranking for 1,840 keywords (49 in top 10), 4,200 organic clicks/mo, 18 cited references in ChatGPT/Perplexity answers
- Leads attributable to organic blog content: 31/mo at month 9 (up from 0)
Houston B2B services firm, 4 posts/mo for 12 months
- Starting state: 8 generic posts, no rankings of value, 0 attributable leads
- Strategy: 1 topical cluster around a specific buyer pain (not a service category)
- 12-month outcome: 5 first-page rankings on $500-$2,000 CPL commercial terms, 14 inbound demo requests directly from blog content
- AI Overview citation on 3 target terms (queries where the prospect's research starts)
Texas-wide e-commerce SMB, 12 posts/mo for 6 months
- Starting state: 60 product-comparison posts, all written by content mill, mostly thin, 1,200 organic clicks/mo
- Strategy: bottom 35 posts noindex+301'd, top 25 rewritten with original data + product photography, 72 new posts in dominance pattern
- 6-month outcome: 8,400 organic clicks/mo, 2.1x organic revenue, 11 of the new posts ranking in AI Overviews
Common mistakes (avoid).
- Publishing volume without topical focus. 40 posts spread across 12 topics never builds authority. 20 posts in 1 cluster does.
- Using AI to ship faster but skipping the SME edit. Generic AI content gets de-ranked. The edit is non-negotiable.
- Writing for word count instead of for the SERP. 1,500 words is not a magic number. Match the depth the SERP rewards for that intent.
- Ignoring AI search. If your content is not structured for passage-level citation, you cede ground to competitors who structure for it.
- Never pruning. Bloat compounds. Posts that drag the average pull the whole site's rankings down.
- Treating the blog as separate from the site. Posts that do not link to commercial pages do not produce leads. Link discipline matters.
- Outsourcing voice to the writer. Voice is the brand asset. Lock it during onboarding, enforce it in editing, do not let it drift.
- Measuring vanity metrics. Sessions and page views do not pay bills. Track keyword positions, AI citation, internal link equity, and attributable leads.
Tools we use.
- DataForSEO for SERP analysis, keyword research, AI optimization API (live, paid per call)
- Claude + GPT-5 for drafting, voice-tuning, and editing assistance
- Mastodon SEO Pipeline (internal): SERP → brief → draft → publish, one command
- Google Search Console for impression/click/position monitoring
- RustySEO for desktop crawling and content auditing
- IndexNow API for instant indexation across Bing/Yandex/Seznam/Naver
- Schema validator + Rich Results Test on every published post
- WordPress, Webflow, Shopify, custom CMS publishing layer (we adapt to your stack)
30 / 60 / 90 day milestones.
| Window | Milestones | What good looks like |
|---|---|---|
| 0-30 days | Topical strategy locked, voice samples loaded, 12-month editorial calendar shipped, first 4-8 posts published | SME-edited drafts hitting voice score 85+, schema validating clean, IndexNow firing |
| 30-60 days | Bottom-funnel + brand terms ranking, first AI Overview citations on long-tail terms | 3-8 first-page rankings on low-difficulty terms, 1-2 AI citations |
| 60-90 days | Mid-funnel category terms moving, first cohort review + pruning of weakest 20 percent | Visible position improvement on tracked terms, 15-30 percent of cohort moved into top 20 |
| 6-12 months | Topical authority compounding, ranking for unscoped terms, AI Overview presence on commercial intent queries | 10x impressions vs starting state, attributable leads from organic content |
FAQ.
- How is your content different from a content mill?
- SERP-driven briefs, human SME editing, original data, low-volume / high-quality cadence. Voice locked to yours, not generic agency voice.
- Will AI content rank in 2026?
- AI content with real editing, data, and POV ranks identically to fully human content. Generic AI content does not.
- Do you write for AI search engines?
- Yes. Passage-level citability, FAQPage schema, llms.txt at apex, definition sentences, citable stats.
- How much does content marketing cost?
- Foundation $2K-$3K/mo for 4 posts, Growth $4K-$6K/mo for 8 posts, Dominance $7K-$11K/mo for 12-16 posts.
- How long until results?
- Brand and bottom-funnel 30-45 days, category terms 90-180 days, compounding at 6-12 months.
- What cadence is right?
- 4/mo is the floor, 8/mo is the sweet spot, 12-16/mo only for category leadership plays with 18+ month horizon.
- Do you do pillar pages?
- Yes. 1 pillar + 6-10 clusters per topical territory. Pillars priced as add-on at $1,500-$2,500 each.
- Will you write in our voice?
- Yours, locked during onboarding from your best existing posts + SME interviews. Target voice-similarity score 85+ before publish.
Related reading.
- SEO services for the technical and authority layer that content stacks on
- Social media management for distribution and trust on top of content
- AI for marketing teams for the underlying pipeline
- Our process for how engagements actually run
- Pricing for full transparency across all services