SEO today is a completely different game than it was 2–3 years ago....
Keval Shah | Ecom SEO + AEO@SEOKeval
25 views
Jan 09, 2026
~3 min read
Advertisement
1
SEO today is a completely different game than it was 2–3 years ago.
A lot strategies that once worked for brands no longer deliver results.
So, I had to rebuild my SEO process from the ground up.
And after a lot of testing, here are the strategies I've found to work best now:
1. You used to be able to slam Product Category pages with links and they'd eventually rank. Now, that will get your pages penalized.
You need to slow it down a bit.
Build a few links to Product Category pages, but focus most of your link building on your homepage and relevant blog content that internally links to your Product Category pages.
2. Speaking of blog content, clicks to top-of-funnel topics got nuked by AI models and AI overviews. They still get some clicks, but nothing like before.
But that doesn't matter. Top-of-funnel content never drove a ton of revenue anyways.
You need to focus most of your blog content around "best" and "vs" keywords.
> Best [Product]
> [Your Brand] vs [Competitor Brand].
> [Competitor Brand] vs [Competitor Brand]
Those topics get clicks, drive revenue and influence AI models.
3. In 2026 and beyond, you need to build links like your life depends on it. Because it does.
Not only has the competition within Google increased, but Google's algorithm has been tuned up to care about authority more than anything.
With every algorithm update, low-authority sites get nuked, and high authority sites rise to the top. Build a ton of high-quality links, and be a high-authority site.
4. Google now evaluates sites more holistically than on a page-by-page basis. That means, if you want to rank for a keyword, you need to cover the overall topic.
Let's say you want to rank for the keyword "pillow covers."
You need to build pages to rank for other relevant keywords within that topic like "linen pillow covers", "white pillow covers", "striped pillow covers", etc.
5. SEO alone is not enough to rank anymore. Google wants to rank actual brands... meaning, you need to have good customer reviews, returning customers, people searching your brand name, etc.
You need to build up that reputation and authority within your industry through other channels like paid ads, email marketing, and influencer marketing before starting SEO.
Only then will Google will give you the green light to start ranking in their search results.
6. AI models are slowly becoming to new default search engine. You need to start optimizing for them.
Fortunately, optimizing for AI models is 50-60% traditional SEO. The remaining 40-50% is getting your brand and products featured in relevant Reddit threads and "best of" listicles on authoritative site.
AI models LOVE to cite Reddit and listicles. So, make sure you're appearing in them, and you'll get cited pretty quickly.
Those are the main changes I've seen.
If you do all of this, you have a good chance of thriving in this new search landscape.
As always, SEO continues to be a game of adaptation.
Those who adapt will do well.
Those who don't will just have to pay more and more money to Google and Meta.
A lot strategies that once worked for brands no longer deliver results.
So, I had to rebuild my SEO process from the ground up.
And after a lot of testing, here are the strategies I've found to work best now:
1. You used to be able to slam Product Category pages with links and they'd eventually rank. Now, that will get your pages penalized.
You need to slow it down a bit.
Build a few links to Product Category pages, but focus most of your link building on your homepage and relevant blog content that internally links to your Product Category pages.
2. Speaking of blog content, clicks to top-of-funnel topics got nuked by AI models and AI overviews. They still get some clicks, but nothing like before.
But that doesn't matter. Top-of-funnel content never drove a ton of revenue anyways.
You need to focus most of your blog content around "best" and "vs" keywords.
> Best [Product]
> [Your Brand] vs [Competitor Brand].
> [Competitor Brand] vs [Competitor Brand]
Those topics get clicks, drive revenue and influence AI models.
3. In 2026 and beyond, you need to build links like your life depends on it. Because it does.
Not only has the competition within Google increased, but Google's algorithm has been tuned up to care about authority more than anything.
With every algorithm update, low-authority sites get nuked, and high authority sites rise to the top. Build a ton of high-quality links, and be a high-authority site.
4. Google now evaluates sites more holistically than on a page-by-page basis. That means, if you want to rank for a keyword, you need to cover the overall topic.
Let's say you want to rank for the keyword "pillow covers."
You need to build pages to rank for other relevant keywords within that topic like "linen pillow covers", "white pillow covers", "striped pillow covers", etc.
5. SEO alone is not enough to rank anymore. Google wants to rank actual brands... meaning, you need to have good customer reviews, returning customers, people searching your brand name, etc.
You need to build up that reputation and authority within your industry through other channels like paid ads, email marketing, and influencer marketing before starting SEO.
Only then will Google will give you the green light to start ranking in their search results.
6. AI models are slowly becoming to new default search engine. You need to start optimizing for them.
Fortunately, optimizing for AI models is 50-60% traditional SEO. The remaining 40-50% is getting your brand and products featured in relevant Reddit threads and "best of" listicles on authoritative site.
AI models LOVE to cite Reddit and listicles. So, make sure you're appearing in them, and you'll get cited pretty quickly.
Those are the main changes I've seen.
If you do all of this, you have a good chance of thriving in this new search landscape.
As always, SEO continues to be a game of adaptation.
Those who adapt will do well.
Those who don't will just have to pay more and more money to Google and Meta.