Transforming SEO for an ISP: How CMAX Boosted Organic Traffic by 86% and Reduced CPA by 37%

An Agentic SEO platform autonomously creates, deploys, tests and improves content based on performance.

Our keyword research first maps where customers are searching across Google and ChatGPT. CMAX then turns those opportunities into customer-targeted copy, dynamically adds it to new or existing pages and continually optimises it using real performance data.

Unlike programmatic SEO, which typically publishes templated pages at scale, CMAX learns, adapts and improves its content over time. It uses only approved brand content, works with any CMS and runs largely on autopilot.

Traditional SEO and GEO programmes tend to target a narrow set of high-volume keywords. But this does not reflect how people or AI engines search.

At least 90% of search demand sits in the long tail. These searches are longer, more specific and individually lower in volume. When AI engines search the web to answer a question, they first use query fan-out, breaking one prompt into 8 to 12 related long-tail searches.

AI engines therefore rely heavily on the same long-tail content found through conventional search. If your brand is not visible for these fan-out searches, it is far less likely to appear in AI-generated answers.

CMAX is an Agentic SEO platform built to capture this opportunity at scale. It autonomously generates, deploys and continually optimises customer-targeted copy to build and maintain visibility across Google and AI search by targeting the long-tail and query fan-out terms customers use.

CMAX typically takes four to six weeks to implement and can start attracting new customers within 60 days of implementation. We offer both server-side and JavaScript options, making CMAX easy to deploy and maintain across new or existing pages.

Our copywriters first create custom content based on your brand, business, legal and regulatory guidelines. Once approved, this content becomes the training data for a custom LLM built specifically for your business.

We do not rely on off-the-shelf models from OpenAI, Anthropic, Grok or Gemini. We use our own models built specifically for SEO to autonomously generate, deploy and optimise customer-targeted copy across long-tail and query fan-out searches.

No. CMAX is designed to avoid the generic, templated content commonly associated with scaled AI publishing.

Our copywriters first create custom content using your brand, business, legal and regulatory guidelines. Once approved, this content becomes the training data for a custom LLM built specifically for your business.

Because we do not rely on off-the-shelf models to generate copy, we reduce the repeated language, predictable patterns and content duplication that can occur when generic models are used at scale.

CMAX uses your approved material to produce customer-targeted copy for specific search intent. The content and its internal linking are refreshed at least every six to eight weeks, helping it remain relevant rather than becoming stale.

During a consultation, we can demonstrate how leading brands successfully use CMAX and project the potential impact of this strategy on your brand.

No. CMAX can operate with or without an existing SEO programme or internal SEO team.

We complete the foundational keyword research, identify the long-tail and query fan-out opportunities, create the initial copy and support implementation. Your team provides the relevant brand, business, legal and regulatory guidelines and approves the content used to train your custom LLM.

If you already have an SEO team or agency, CMAX complements their work by targeting opportunities that are difficult to reach manually at scale. Once deployed, it autonomously generates, adds and continually optimises customer-targeted copy across new or existing pages.

The Missing Customer Map uncovers what is often a brand’s biggest marketing blind spot: the customers already searching for what it sells but never seeing it.

It shows where most search demand exists across your key products and services, and how effectively you and your competitors are capturing it across Google and AI search.

Where data access allows, we analyse your Google Analytics and Google Search Console data to model how often your brand appears across the long-tail searches used during AI query fan-out. This provides a stronger indication of whether your brand is being considered for AI-generated answers than prompt trackers, which can be inconsistent and offer limited actionable insight.

We also identify “silent” AI traffic that is often misclassified as direct traffic because of known attribution gaps in analytics platforms. While AI traffic is widely considered small, our analysis has found cases where it exceeded organic traffic as early as October 2025.

Most importantly, the Missing Customer Map provides a clear framework for systematically capturing this previously untapped market. CMAX can then execute that framework with little ongoing effort from your team, continually building visibility through its self-learning and self-optimising capabilities.

CMAX is already used globally by brands across a broad range of industries and generates more than 60 million visitors per month in aggregate.

CMAX can be implemented through either a server-side integration or an asynchronous JavaScript option, depending on your website and technical requirements. It works with any CMS and can add customer-targeted copy to new or existing pages without requiring a website rebuild.

Our JavaScript implementation is designed to load extremely quickly and minimise any effect on site performance. We can provide examples of similar client implementations that your team can independently test for speed. We also use Google Search Console indexing tests to confirm that the dynamically added content is being discovered and indexed correctly.

Implementation typically takes four to six weeks. We provide straightforward implementation guides and can work directly with your technical team whenever assistance is needed.

Once deployed, CMAX autonomously adds, tests and optimises copy.