Should we change existing blogs to rank in AI results?

That's a question we've been debating a lot with clients here in Dubai, whether it's worth going back into blogs that already exist and reworking them for AI search, rather than just writing fresh content going forward.

I ran into this properly with one of our clients earlier this year, a professional services firm with a decent library of blog content that ranked reasonably well on Google but simply never came up when we tested the same topics on ChatGPT or Perplexity. We pulled together SEO, content, technical SEO, web development, UX and analytics to look at it properly rather than guessing. Technical SEO checked whether AI crawlers could actually read the pages in raw HTML, since a lot of the site's content was loaded via JavaScript and effectively invisible to bots like GPTBot. Content and UX restructured the highest traffic articles around direct questions with clear answers up front, and analytics helped us prioritise which posts had strong existing traffic but zero AI citations, so we weren't wasting effort on pages nobody was reading anyway.

The core issue is that ranking on Google and being cited by AI are increasingly separate outcomes. AI systems favour clear structure, fact density and content they can extract without needing to interpret clever headlines or buried answers. The most common mistakes I see are pages with vague titles, answers hidden several paragraphs down, and content genuinely blocked from AI crawlers without anyone realising.

In practice, updating existing blogs makes more sense than starting over. Restructure key sections as direct question and answer pairs, add sourced statistics, check robots.txt isn't blocking AI bots, and confirm your best content is actually crawlable server side. It's incremental work, but it's realistic and it compounds.

Yes, updating existing blogs is generally the smarter move than starting fresh, and honestly it lines up with what I've been doing anyway. Here's my actual thinking on it.

Old content decays fast now, faster than it used to. Some research floating around this year found that content older than four years makes up only about 3.6% of AI-referred traffic, and roughly 80% of AI-driven traffic goes to pages that were updated within the last two years. That tells me freshness isn't just a Google thing anymore, it's an AI citation thing too. 

The reason a lot of well ranked blogs still get ignored by tools like ChatGPT, Gemini, and AI Overviews comes down to structure, not quality. These systems aren't reading your post the way a person does, they're scanning for a clear, extractable answer. If your main point is buried under three paragraphs of scene setting or storytelling, the AI just skips past you and grabs whatever competitor put the answer in the first hundred words. So when I go back into an old post now, I'm not rewriting the whole thing from scratch, I'm restructuring it. Answer first, context after.

A few things I've started prioritizing when refreshing older posts:

Putting the direct answer to the target query right up top, before any storytelling or setup. Adding FAQ sections even where there's no rich result guarantee, because it still gives AI tools a clean chunk to pull from. Tightening internal links so newer posts point back to the older ones and vice versa, since a lot of good content just gets buried and forgotten inside a site. Swapping out old stats and examples for current ones, because outdated numbers quietly kill trust signals even if nobody complains about them directly.

Not every post deserves the same effort though. I've settled into basically a tiered approach. High traffic or high value pages get a full rewrite with new sections, updated data, and schema added properly. Mid tier pages that are ranking somewhere on page two just need the intro and conclusion sharpened and broken links fixed, that's often enough to nudge them up. And then there's a bottom tier of thin, overlapping, or genuinely useless posts that I'd rather merge into one stronger page or just retire with a redirect than keep patching forever.

One thing I'd push back on is the idea that a rewrite always beats a fresh page. If a post was thin from the start, targeted the wrong intent entirely, or reads in a tone that just doesn't hold up to today's standards, patching it can actually make it worse, a messy update on a bad foundation. In that case building a new page and redirecting the old URL keeps the link equity without dragging forward a weak structure.

So my honest answer is yes, update your existing blogs, but treat it as restructuring for how AI tools actually extract information, not just refreshing dates and swapping a stat or two. That distinction is where most people are still getting it wrong.

In my opinion, not every existing blog needs a complete rewrite in order for it to be ranked high in AI searches. Instead, I check out what are my top performing or most valuable pieces of content and will refresh those pieces with a focus on making them clearer, accurate and easier both for people and AI systems to consume.

When refreshing, I typically will:

1) Add direct answers to frequently asked questions

2) Improve existing headings

3) Update any outdated content

4) Include frequently asked question content as appropriate

5) Strengthen internal linking throughout the piece. This approach improves the overall quality of the content without changing its original intent.

As a result, my intention is not created just for AI consumption but rather provides genuine benefit to users. Consequently, if the structure, trustworthiness and completeness of the content is present in the blog post there will be increased likelihood of success both in organic and AI generated search results.

I do not believe every existing blog needs to be rewritten specifically for Ai results. Instead, I review high value blogs and update them to make the content clearer, more comprehensive, and easier for both users and Ai systems to understand.

When updating a blog, I focus on adding direct answers to common questions, improving headings, including FAQs, strengthening internal links, and ensuring the information is accurate and up to date. Well structured content is more likely to be referenced by Ai tools than content that is outdated or difficult to navigate.

I also avoid changing content that is already performing well unless I see a clear opportunity for improvement. A complete rewrite can sometimes hurt existing rankings if it removes important keywords and context.

From my perspective, the goal is not to rewrite blogs for Ai. The goal is to create content that provides clear, trustworthy, and useful information. When a blog genuinely helps users, it has a better chance of performing well in both traditional search results and Ai generated answers.

Before starting SpiderWorks in 2005, I had been working in the blogging and SEO industry since 1999. I have witnessed almost every major transformation in search technology. I have seen the era of primitive search directories, keyword stuffing, backlink manipulation, PageRank obsession, semantic search, mobile-first indexing, Core Web Vitals, and now the biggest disruption of all - AI-powered search experiences.

For more than two decades, SEO professionals learned how to optimise content for traditional search engines that ranked web pages primarily through links, keywords, structure, and authority signals. However, around two years ago, AI search engines and AI-powered answer systems fundamentally changed how users consume information online.

Today, users increasingly receive direct AI-generated answers instead of simply clicking blue links on search result pages. This has changed the way content needs to be structured and optimised. However, one thing has not changed: authoritative, trustworthy, experience-driven content still remains the foundation of search visibility.

The difference is that AI systems now evaluate content differently compared to traditional ranking algorithms.

AI Search Engines Prefer Clear, Structured Knowledge

Traditional SEO often focused heavily on keywords and ranking positions. AI search engines focus more on extracting accurate, reliable, and well-structured information.

One of the most important changes you should make to existing blog posts is improving content clarity and structure. AI systems prefer content that is easy to parse, summarise, quote, and understand.

This means you should update your old blog posts with:

- More descriptive headings and subheadings

- Break long paragraphs into smaller sections

- Add bullet points and numbered lists

- Include concise explanations for important concepts

- Add FAQ sections wherever relevant

- Use natural language instead of forced keyword repetition

AI engines often pull snippets directly from well-structured sections of content. Posts that clearly answer specific questions have a higher chance of being referenced in AI-generated answers. So, you may update your old blog posts with more clear, direct answers to problems. 

Demonstrate Real Expertise and Experience

One of the biggest shifts in modern SEO is the increasing importance of experience-based authority.

Generic AI-generated articles are flooding the internet. As a result, search engines and AI systems are increasingly prioritising content written from real-world expertise and firsthand experience.

This is where experienced bloggers and industry professionals have a huge advantage. If you have strong expertise in your area, you may update the old blogs to reflect your subject knowledge. 

When updating older blog posts:

- Add personal insights and observations

- Include real examples and case studies

- Mention industry experience where relevant

- Add practical recommendations

- Share historical perspective and trend evolution

For someone like me who has observed search engine evolution since 1999, this historical context itself becomes a strong authority signal. AI systems increasingly value content that reflects deep subject understanding rather than surface-level summaries.

Improve Content Freshness

One of the biggest weaknesses of older blog posts is outdated references.

Despite all the technological disruption over the last 25 years, the fundamental principle of SEO has remained remarkably consistent.

Search engines, whether traditional or AI-driven, ultimately want to surface the most authoritative, trustworthy, and useful content for users.

What has changed is the method of evaluating authority.

In the early 2000s, backlinks and keywords dominated. Today, AI systems analyse expertise, structure, semantics, trust signals, user engagement, brand authority, and content quality at a much deeper level.

Conclusion:

So the conclusion is, you should go and update all of your old blog posts to improve their ranking in AI search results. Updating the content of old blogs with the fresh and more relevant content will help in better ranking on the normal search results as well. 

If you need any help to improve the ranking of your website on AI search, you can connect with us. We have helped several clients to improve their ranking on AI results. Also, if you need help with updating your old blog posts, our content writers can help with the same. 

Yes, we should update the existing blogs to improve their chances of appearing in AI search results and Google rankings. Updating outdated content, adding more useful information, expert insights, and improving the content structure can help the blogs perform better in search results.

AI search results usually prefer content that is updated, clear, and trustworthy. Regularly refreshing important blogs can also help maintain rankings and improve user engagement.