Challenges: Too Smart for Google, Invisible to AI

Codewars was stuck in traditional content formats. Here’s what we uncovered:

1. LLMs couldn’t find their content

Kata content was buried in dynamic JS
No schema or semantic HTML. So, it was hard for LLMs to extract meaning
Code examples had no structured steps or explanations

2. Missing visibility in zero-click AI answers

ChatGPT, Gemini, and Copilot consistently showed results from Reddit, GeeksforGeeks, and Medium.
Even branded queries like “solve Codewars recursion challenge in Python” didn’t surface official pages.

3. No content targeting prompt-style queries

Content was dev-centric, not prompt-centric
No pages formatted like typical AI queries (“How to solve [concept] in [language] with edge cases?”)
No answer-first, structured walkthroughs that LLMs favor

4. No AI traffic measurement or strategy

No referrer tracking from ChatGPT or Copilot
The sitemap was flat. High-performing kata pages weren’t prioritized
Internal linking was unstructured, so topic authority wasn't established

Strategy: The Growth Plan by ResultFirst

Building Content That AI Could Actually Find

We built an innovative, flexible system to help Codewars show up more often when people search on AI tools like ChatGPT, Copilot, and Gemini.

Here’s how we did it:

Step 1: Matched Kata Topics to Real Searches

We went through Codewars’ library of coding challenges and sorted them by things like programming language, skill level, and topic. Then we looked at how developers were actually asking for help on AI tools. For example:

Best website for practicing programming
Websites to practice coding
Best Coding practice website
Best Coding website for Practice
Best websites to practice coding

We used those real questions to guide our content and title choices. Ultimately, we created over 400+ focused topics that answered specific, common developer questions, just as they’d ask them in AI tools.

The result?

Pages that actually answered what devs (and AI tools) were looking for.

Step 2: Made Pages LLM-Ready (Not Just SEO-Friendly)

We didn’t stop at writing better content. We structured it specifically for how large language models read the web. Here's how:

Cleaned HTML Structure: We used clear, semantic tags so AI tools could follow the logic:
- ‹h2› for the main problem
- ‹h3› for each step in the solution

This made the flow easy to parse for tools like ChatGPT and Gemini.

Schema That Speaks AI: We added Schema.org markup, such as FAQPage and CodeSample, so search engines and AI platforms could understand the content type and context, not just crawl the words.
Smart Keyword Placement: We dynamically pulled in terms like language, concept, and difficulty level into the meta tags and H1s.

This helped us match not just SEO queries but also AI prompt formats.

Deployed LLM.txt: Updated the llm.txt file to allow large language models like ChatGPT, Copilot, Gemini, Perplexity, and others to crawl and prioritize key pages.

Why Our Strategies Worked

Targeted how devs actually ask questions.	Used actual prompts from ChatGPT/Gemini like “reverse a string in Python” instead of vague keywords.
Broke solutions into clear, step-by-step formats.	Broke answers into clean tutorials: simple steps, readable code, and logical flow—just like you'd want from ChatGPT.
SEO-friendly HTML & schema	Used proper tags (‹h3›, ‹code›) and structured data to help AI parse and surface content.
Repurposed old kata into AI-friendly answers	Instead of rewriting everything, we took existing challenges and reframed them as walkthroughs tailored to common dev queries, turning legacy pages into top answers on ChatGPT and Gemini.
Tracked LLM traffic	Monitored what AI tools picked up, then optimized based on what worked weekly.

Ready to Be the Next AI-Smart Brand?

ResultFirst helps fast-moving SaaS tools, dev platforms, and marketplaces get found by humans and LLMs.
Same playbook. Same results. No guesswork.
Your content may be good. We’ll make sure AI sees it.
Start the conversation today.

The Results (April → June 2025)
LLMs	Apr '25	Jun '25	Growth
ChatGPT	21,438	24,521	14%
Gemini	193	336	74%
Copilot	10	225	2150%
Total AI	21,880	25,348	16%

22x Spike in Copilot and 2x Gemini in 3 Months

Challenges: Too Smart for Google, Invisible to AI

1. LLMs couldn’t find their content

2. Missing visibility in zero-click AI answers

3. No content targeting prompt-style queries

4. No AI traffic measurement or strategy

Strategy: The Growth Plan by ResultFirst

Building Content That AI Could Actually Find

Step 1: Matched Kata Topics to Real Searches

Step 2: Made Pages LLM-Ready (Not Just SEO-Friendly)

The Measurable Wins for Codewars:
3-Month Growth

AI Overview

Copilot

ChatGPT

Perplexity

Why Our Strategies Worked

Ready to Be the Next AI-Smart Brand?

ResultFirst is the ONLY SEO agency
you will ever need.

Case studies

GMAT Club

Plushbeds

Baxterboo

22x Spike in Copilot and 2x Gemini in 3 Months

Challenges: Too Smart for Google, Invisible to AI

1. LLMs couldn’t find their content

2. Missing visibility in zero-click AI answers

3. No content targeting prompt-style queries

4. No AI traffic measurement or strategy

Strategy: The Growth Plan by ResultFirst

Building Content That AI Could Actually Find

Step 1: Matched Kata Topics to Real Searches

Step 2: Made Pages LLM-Ready (Not Just SEO-Friendly)

The Measurable Wins for Codewars: 3-Month Growth

AI Overview

Copilot

ChatGPT

Perplexity

Why Our Strategies Worked

Ready to Be the Next AI-Smart Brand?

ResultFirst is the ONLY SEO agency you will ever need.

Case studies

GMAT Club

Plushbeds

Baxterboo

The Measurable Wins for Codewars:
3-Month Growth

ResultFirst is the ONLY SEO agency
you will ever need.