Matteo Barletta's USA-based local SEO SaaS needed to rank for the very thing it sells. Google Business Profile optimization, local Maps visibility, multi-location SEO. We built the full content engine. 56+ articles. Seven topical clusters. One prompt system that doesn't break.
Build me an engine like this →no retainer trap. just truth.
Matteo had the product. He needed the compounding visibility only deep topical content delivers.
GMB Crush is a local SEO SaaS. The cruelest irony in B2B: pitching local rankings to a buyer who Googled "best GBP tool" and didn't see you. Credibility was on the line every time someone typed a query.
A few scattered posts. No cluster strategy. No internal linking architecture. No publishing rhythm. The kind of blog that looks like an afterthought because that's exactly what it was.
Their customers come from legal, dental, home services, and multi-location brands. Each vertical has its own search language. Generic SEO content was leaving every one of those audiences on the table.
The market is drowning in ChatGPT-spat articles. Generic prompts producing generic prose producing generic rankings (i.e. none). The bar to actually rank now is higher than ever, and most agencies are still pretending it isn't.
Seven clusters (Legal, Dental, GBP, Maps, and three more) each mapped to real buyer paths. Pillar pieces anchor every cluster. Supporting articles funnel internal link equity upward. The architecture compounds; one-offs don't.
We built the ZimmWriter v3 prompt stack Matteo runs on. Surfer + Neuron Writer for SERP analysis. Custom prompts tuned per cluster so output reads like a practitioner wrote it. Output people read. Not slop.
Not a deck. Not a strategy doc. The actual files that live on their site.
Full topical authority map for seven verticals. Cluster centers, pillar pieces, supporting articles, search-intent classification per query.
Custom prompt system built inside ZimmWriter v3. Cluster-aware. Vertical-aware. Tuned to GMB Crush's actual brand voice.
Pillar pieces and supporting articles shipped across seven clusters. Refresh series in active rotation. The catalog keeps growing.
Internal linking isn't an afterthought, it's a system. Every published article slots into the cluster's link graph on a documented rule set.
JSON-LD schema for every published article (Article, FAQPage, BreadcrumbList, Organization, Person) entity-stacked per Casey Keith principles.
Content isn't ship-and-forget. The refresh series keeps the catalog ranking, flagging stale data, missing entities, weak sections, new SERP competitors.
Two calls with Matteo. Verticals scoped, customer language captured, existing site audited, competitor SERPs torn apart.
Cluster centers defined. Keyword universe per cluster. Pillar + supporting hierarchy. Internal linking rule book written.
ZimmWriter v3 stack tuned. Brand voice captured in prompts. Surfer + Neuron Writer wired into the brief generation loop.
Articles shipped on a fixed cadence. Every piece QC'd, schema'd, internal-linked, monitored in GSC from day one.
Quarterly audit. Stale flags. SERP-shift response. The catalog isn't a graveyard, it's a living system that keeps earning.
Every one of these is something we run daily, not something we read a tutorial about.
Different industry. Different problem. Same practitioner approach.
Most of the granular performance data is Matteo's, his platform, his account, his decision what to publicize. We won't post screenshots of someone else's GSC without permission. What we'll say: the engine is in active rotation, the catalog continues to grow, and the relationship is ongoing. If you want specifics, we'll request permission and walk you through on a call.
Because Matteo and the team had to be able to run the engine without us in the loop every day. ZimmWriter is one of the few SEO-content tools designed for serious prompt engineering at production scale, not a chat UI someone copies queries into. We built the prompt stack inside it so the system is theirs to operate. That's what white-label delivery should mean: we hand over a machine, not a dependency.
Google's actual position (March 2024 guidance onward) is that content quality matters, source of generation doesn't. What gets penalized is generic, low-effort, no-original-insight content, and most AI output is exactly that. Our approach: prompt systems tuned to brand voice, vertical-specific knowledge, Surfer-scored optimization, human editorial pass, and a fact-check loop. Same hard work, different production speed.
Honestly? The smaller you are, the more important it is. Big sites can brute-force rankings with domain authority. Small sites have to be smart, and topical authority is how small sites punch up. We'd rather build one tight cluster of 15 pieces that ranks than scatter 60 disconnected posts that don't.
It varies by domain age, topical authority, competition, and a lot of other variables nobody can promise around. What we tell every client: the first 60 days are setup + early publishing. Months 2–4 you see indexation, impressions, and long-tail wins. Months 4–8 the cluster effect compounds and primary terms start moving. Anyone promising you'll rank in week three is lying.
Yes, that's what File №01 exists to prove. The architecture (topical mapping, ZimmWriter prompt stack, internal linking system, refresh loop) is replicable across SaaS, B2B, e-commerce, local. Each industry needs its own cluster map and prompt-system tuning, but the engine pattern holds. Email us with what you sell and we'll tell you honestly if it's a fit.
We'll audit your current content, map the cluster opportunity, and tell you (honestly) what a real engine looks like for your vertical. Before you commit to anything.
no retainer trap. just truth.
info@themarketingmaniacs.com · themarketingmaniacs.com · SECP A827871 · Texas LLC