How the Double Atari Website Keeps Evolving

How did this project start?

The first version of the Double Atari website came from a pretty simple spark.

I wrote about this in my post on building a fast, SEO-friendly website with AI, but the short version is that a friend of mine built a Scrabble website using AI. He had not built a website before, and the site was fast, clean, and surprisingly effective. I ran it through Lighthouse, saw how well it performed, and thought: I should try this for Double Atari.

At first, the goal was mostly practical. I wanted a site that loaded quickly, reflected my work better than the old version, and gave me more control than a hosted website builder. I also wanted to see what would happen if I combined a lightweight static site with the SEO, AEO, GEO, schema, analytics, and content strategy practices I use professionally.

That first build was not meant to be perfect. It was meant to give me a better foundation.

What changed after the site launched?

Once the site was live, the project shifted from “build the website” to “listen to what the website is telling me.”

That is where tools like GA4, Google Search Console, and SEMrush became more important. Instead of guessing what to write or which pages to improve, I started looking at the signals:

• Which pages were getting impressions but not clicks?
• Which blog posts were ranking on page two or page three?
• Which queries were Google already testing the site for?
• Which pages had engagement but not enough visibility?
• Which topics were close enough to deserve more content?
• Which pages needed stronger internal links?
• Which pages needed better titles, descriptions, headings, and schema?

I wrote about this process in How to Use GA4, Search Console, and SEMrush to Find AI Search Opportunities. That article was really about changing the way I think about content planning. It is not just keyword research anymore. It is a loop between visibility, engagement, authority, and whether the page is structured well enough for both traditional search and AI search.

What data did I use to decide what to improve?

In one recent review, the site had 155 referring domains, 4 organic keywords, and 78 GA4 organic sessions over the last 60 days. Those numbers are not huge, but they are useful because they give me a baseline.

The homepage was also showing a clear click-through-rate issue. It had 205 impressions, an average position around 4.75, and only 4 clicks. That is roughly a 1.9% CTR, which is low for a page showing that high in search results. That told me the page was appearing, but the title, description, or search intent match probably needed work.

The About page had a different problem. It was showing for “Andrew Charon” and “Andrew Charon Double Atari,” but it was not earning clicks. That suggested the page needed stronger brand-search signals: clearer title tag, better meta description, an H1 with my name, a stronger opening paragraph, better credentials, and Person schema tied to the author signals on the blog.

Some blog posts were in the most interesting zone: visible, but not high enough. Posts around Answer Engine Optimization, Squarespace vs. WordPress SEO, building the site with AI, and GEO vs. AEO vs. SEO were sitting in positions where Google seemed to understand the relevance but was not fully ready to reward the pages yet.

That is a very different problem than starting from zero. It means the next step is not random content production. It means improving the pages that already have traction.

How has Perplexity helped with the process?

I have been using Perplexity less like a writing shortcut and more like a strategy assistant.

The most helpful part has been connecting the dots across tools. GA4 shows what people do after they arrive. Google Search Console shows what queries are already creating impressions. SEMrush helps me understand keywords, competitors, and position tracking. Perplexity helps synthesize those inputs into a practical plan.

For example, if Search Console shows that a blog post is ranking in position 17, that does not automatically mean I need a new post. It might mean I need:

• A stronger definition section near the top.
• More internal links from related service pages.
• A better FAQ block.
• Updated Article schema.
• A clearer author bio.
• A comparison table.
• A more direct answer to the exact query Google is testing.

That is the kind of work Perplexity has been useful for. It helps turn scattered SEO data into a list of edits I can actually make.

What have I been changing on the site?

The work has been a mix of technical SEO, content cleanup, and authority building.

On the technical side, I have been improving title tags, meta descriptions, H1s, heading structure, FAQ schema, Article schema, Person schema, LocalBusiness schema, internal links, sitemap structure, and canonical consistency. I have also been careful not to change URLs unnecessarily because the site is live and starting to be indexed.

On the content side, I have been expanding pages where there is already some visibility. That includes adding deeper explanations, better examples, more practical FAQs, comparison tables, and more direct answers for search engines and AI answer engines.

On the credibility side, I have been trying to make it clearer that the site is written by a real person with actual experience. That means stronger About page content, author cards on blog posts, links back to the About page, and more consistent references to the work I actually do: SEO, GEO, AEO, website audits, CMS strategy, analytics, and content planning.

The site is still simple. That is intentional. But it is becoming more structured.

What is still the biggest challenge?

The biggest challenge is website authority.

The site can be fast. It can have clean HTML. It can have good schema. It can have well-written content. But if the domain does not have enough authority, it is still harder to rank for competitive topics.

That is the part I have to be honest about. The authority has not meaningfully increased yet.

That does not mean the work is not helping. It means the site is entering the stage where content and technical cleanup are not enough by themselves. The next phase needs to focus on signals outside the site too.

What did Perplexity recommend for building authority?

Perplexity’s recommendations were not mysterious, which is actually helpful. They were the kinds of things that are easy to understand but take consistency to execute.

First, build a focused backlink outreach list. That means identifying podcasts, newsletters, local business sites, marketing communities, WordPress resources, SEO blogs, Minneapolis business organizations, and partner sites where a Double Atari article, quote, interview, or resource would actually make sense.

Second, create linkable content that is more useful than a normal blog post. Examples include checklists, comparison tables, audit templates, keyword map examples, schema examples, and practical guides that marketers can use.

Third, use relationships more intentionally. If I have worked with people, contributed to a project, been part of a community, or have a useful perspective for a newsletter, that is a more realistic authority path than cold outreach at scale.

Fourth, keep building topical depth. One post about AEO is helpful. A cluster of posts about AEO, GEO, schema, AI search measurement, entity signals, and content structure is much stronger.

Fifth, improve internal links so that important pages are not isolated. If a page matters, the rest of the site should point to it clearly and naturally.

None of that is a hack. It is closer to reputation-building than technical optimization.

What content should I create next to build authority?

If I were using Double Atari as the example, these are five blog posts I would prioritize next:

1. How to Measure AI Search Visibility Without Overcomplicating It
   A practical post about tracking AI search mentions, referral traffic, branded searches, and content visibility across tools like GA4, Search Console, SEMrush, Perplexity, and ChatGPT.
2. The SEO Audit Checklist I Use Before Rewriting a Website
   A practitioner-focused checklist covering crawlability, metadata, headings, schema, internal links, page speed, analytics, indexation, and content intent.
3. How to Build a Keyword Map for SEO, AEO, and GEO
   A deeper version of the keyword map process, with examples of primary keywords, secondary keywords, questions, intent, page recommendations, and measurement notes.
4. What Marketers Should Know About Schema Markup Before AI Search Gets More Competitive
   A non-developer explanation of Organization, Person, Article, FAQ, Breadcrumb, LocalBusiness, and Service schema.
5. Why Local SEO Still Matters When People Search with AI Tools
   A post explaining how location signals, entity clarity, business profiles, local mentions, service-area content, and reviews still matter when AI tools summarize options.

Those topics support the service pages, answer real questions, and build around the direction search is moving.

What is the main lesson so far?

The biggest lesson is that a website is never really done.

That used to sound annoying to me. Now I think it is the point. The first version gives you something to measure. The next version responds to what you learn. The version after that gets more focused.

The Double Atari site started because I saw a fast Scrabble website and wanted to try building something similar for my own business. Now it is becoming a working example of how I think about websites: build something useful, measure it, improve it, and keep sharpening the structure until the site explains itself clearly to people, search engines, and AI systems.

That is the process I am trying to keep following.

FAQ

How often should a website be updated for SEO?

A website should be reviewed monthly for performance signals and updated when the data shows a clear opportunity. That does not mean rewriting everything constantly. It means checking Search Console, GA4, and keyword tracking to find pages with impressions, weak CTR, slipping rankings, or strong engagement that deserves more visibility.

What is the difference between optimizing content and building authority?

Optimizing content improves what is already on the site: titles, headings, internal links, schema, examples, and page structure. Building authority is broader. It includes earning links, mentions, citations, relationships, and external signals that help search engines and AI systems trust the site as a useful source.

Can AI tools help with SEO strategy?

Yes, but they work best when they are grounded in real data. AI can help analyze GA4, Google Search Console, SEMrush, crawl exports, and competitor patterns. The value is not just generating copy. The value is turning messy performance signals into decisions about what to improve next.

Why has authority been harder to improve than content?

Authority depends on how the rest of the web responds to the site. Content and technical fixes can be done directly, but backlinks, mentions, referrals, and reputation take outreach, relationships, usefulness, and time. That is why authority usually moves more slowly than on-page optimization.