SEO Estonia 2026: Is SEO dead?

SEO isn't dead. You won't find it buried at Rahumäe Cemetery. ChatGPT answers questions, Google serves up AI Overviews, and here we go again: is investing in your website and SEO still worth it?

Oh yes, absolutely. The Ballers team attended SEO Estonia 2026, and the talks reinforced something we've been saying for a while: even in the age of AI search, you still need a website that works, content that makes sense, and a good reason for customers to trust you.

We've put together the biggest takeaways from the conference and, more importantly, what you can actually do with them on your own website.

1. Make sure your important pages can be found

Dinesh Sivapragsam talked about websites with millions of pages, but his main point applies just as much to a ten-page business website. Creating more pages won't get you very far if Google can't find them or add them to its search results.

Kristina Frunze highlighted the importance of internal linking. Think of your website's pages as islands. Unfortunately, a bunch of isolated islands like Kihnu and Ruhnu won't get us very far. What we need is a direct connection to the mainland. A tunnel from Tallinn to Helsinki, if you will. On a website, those connections come from links in the navigation, service pages, and related articles. They help both visitors and search engines find the next relevant page.

In the project Kristina presented, the team improved content, added internal links, and fixed technical issues. They also earned backlinks from trustworthy websites and expanded pages with FAQs and structured data, giving search engines additional information they could understand.

According to her results, organic traffic increased by 107% in six months, while the number of visitors coming through AI grew by 5,300%. That last number sounds like the return on some meme coin, but the starting point was around six users per month. The growth came from several improvements working together. Adding an FAQ section alone won't magically produce the same results.

A good internal link works like a road sign at the right intersection. It shows people where to go next. If someone is reading about e-commerce website development, they should be able to find relevant projects, pricing information, and a contact form from there. Finding the next step shouldn't feel like an orienteering competition.

Google's AI search guidelines are also built on traditional SEO fundamentals. You don't need special AI markup, and adding FAQs or structured data doesn't guarantee a spot in AI-generated answers.

2. Answer the questions your customers are already asking

Aleyda Solis emphasized that e-commerce content should help customers make decisions. Keywords alone aren't enough. Buyers have budgets, specific needs, and plenty of questions they want answered before spending their money.

In her example, a runner was looking for cushioned running shoes for marathon training. An online store should explain which models are suitable and why. Accurate product information and useful filters help narrow down the options, so customers don't have to figure out the suitability of every pair themselves.

Michał Suski's analysis showed that pages referenced by AI answered the user's question and included the facts needed to support that answer. His recommendation was to ask AI the same questions your customers ask before making a purchase. Then look at which pages AI references and what information those pages provide. This can help you identify what your own website should explain more clearly.

Then check whether your website answers those questions.

• Who is this product or service suitable for?
• What's included in the price, and what affects the final cost?
• What's the difference between two similar models or packages?
• Is the product compatible with the customer's existing devices?
• What should customers know about delivery, installation, or getting started?

For e-commerce websites, also make sure prices, availability, and product details match across product pages, category pages, and structured data. Customers shouldn't reach the checkout only to discover that the advertised price no longer applies. Building a business around Turkish bazaar-style pricing tricks probably isn't the greatest long-term strategy.

3. Give AI something worth writing about

Kyle Roof demonstrated that simply telling AI to "write SEO-optimized content" didn't automatically improve the text in his tests. In some cases, important concepts actually disappeared. A little reminder to ourselves as well: always review and edit AI-generated content.

In his team's study of 10,937 pages, the target keyword was missing from the H1 heading on 63.3% of pages and from the HTML title tag on 59.5% of pages. These figures describe the pages included in the study. Our practical takeaway? Before experimenting with your next clever AI trick, check whether your page titles actually make it clear what the page is about.

Georg Richard Aare recommended collecting your company's knowledge in one place and building a knowledge base. Completed projects, customer questions, the reasoning behind decisions, and actual results give AI something useful to work with when creating content drafts. Missing information can be collected from your team and saved for future use.

• What was the customer's problem? Describe the situation so that even someone outside your industry can understand it.
• What did you do, and why? Highlight the decisions that could be useful to the reader.
• What changed? Include the results if they're known and verifiable.
• What can others learn from it? Turn the experience into practical advice readers can apply to their own situation.

AI can help you write. You give it something worth writing about.

Google's guidelines on AI-generated content also emphasize accuracy, quality, and usefulness. Mass-producing pages that offer little or no value can violate Google's policies.

4. Make customer experiences visible

Chris Meabe talked about B2B companies and Reddit. His advice was to understand the community first and focus on helping people. Promoting your product in every discussion can do more harm to your reputation than good.

Greg Gifford focused on reviews. Ask for feedback consistently, make leaving a review easy, and respond to what customers say. Especially when something went wrong. Remember, your next potential customer might be reading that response too.

Choose your review platforms based on where your customers actually compare businesses.

Start with one or two relevant platforms. When responding to a review, address the customer's actual experience. Thank them for their feedback, explain what happened if necessary, and let them know how the issue can be resolved.

Copying and pasting "Thank you for your feedback!" under every single review makes you sound like a robot. And the internet already has more than enough robots and mediocre content. Fact.

5. Look beyond visibility. Track leads and sales.

Karl Kangur emphasized that there's no permanent first-place ranking in AI-generated answers. The companies recommended by AI, and the order in which they appear, can change.

The response may depend on how a question is phrased, previous conversation context, the AI model being used, and information found online. When personalization is enabled, factors such as the user's location and preferences can also play a role. You can read more about these factors in OpenAI's model optimization guide, ChatGPT Search documentation, and Google's AI search guidelines.

If AI recommends your company first, that's a good sign. But one lucky shot doesn't win the game. To understand how often your company appears in AI recommendations, check the answers at different times using several questions that matter to your customers.

Michał Suski also pointed out that AI monitoring tools and actual customers may receive different answers. It's like watching the same game from different camera angles. You won't necessarily see the same thing. That's why it's worth checking which questions were asked and how the answers were collected before drawing conclusions from a report.

Aleyda Solis recommended looking at visibility, website readiness, and business results together. In practice, this comes down to three questions:

Check whether AI mentions your company or links to your website. Record the question, date, AI platform, and result. Repeat the same questions over time to see whether your company appears in recommendations more frequently. At the same time, keep an eye on whether your website is generating more leads and sales.

Also, ask new customers how they heard about you. Kristina Frunze pointed out that someone might first discover a company through an AI-generated answer but visit its website later through Google. Your analytics might never capture that first interaction.

6. Prioritize improvements that help customers move forward

Maja Jovancevic recommended figuring out which problems need to be solved first and which tasks can wait. In her example, a confusing website structure and poor internal linking prevented the site from getting the full benefit of its content and external backlinks.

It's like spending money advertising your store, only for customers to walk in and have no idea where to find what they're looking for. Maybe fix the store layout first.

She also showed two Google Trends graphs side by side: "AI SEO" and "SEO recovery". The curves looked almost like identical twins. As interest in using AI for SEO increased, so did interest in recovering lost SEO performance. One graph makes promises. The other sends the invoice.

Which brings us, quite naturally, to doping.

AI in SEO is a bit like doping. It's a tempting shortcut to better results, but it doesn't replace training or proper technique. Just ask the Estonian national ski team.

If you understand SEO fundamentals, AI can help you get a lot of work done faster. But if you're hoping AI will make up for what you don't know, you might just end up producing mistakes faster and at a much larger scale.

Kalin Karakehayov warned against excessive auditing. It's easy to spend so much time checking and planning that you could have already fixed something useful instead.

When deciding what to tackle first, focus on what's causing the biggest problems in your customer's journey.

• Your contact form isn't working? Fix it and send a test inquiry. Make sure the message actually reaches the right inbox.
• An important service page is difficult to find? Add links from relevant pages and test the journey yourself.
• Customers don't understand what's included in the price? Improve your pricing information before publishing another generic blog post.

Assign an owner, a deadline, and a simple success check to every task. How will you know when it's actually done?

An SEO audit is most valuable when something on your website actually improves afterward.

Where should you start?

Listening to the talks at SEO Estonia gave us a serious sense of déjà vu. A technically sound website, well-planned internal links, useful content, and trustworthy backlinks. Yep, sounds like a regular day at Ballers.

The presentations from international SEO experts reinforced the same principles we follow in our everyday work.

AI is changing how customers discover your business. But once they land on your website, they still have the same questions. What do you offer? Is it right for them? And what should they do next?

A good website gives them clear answers.

Pick one important service or product page to improve this week.

• Open it on your phone and go through the customer journey yourself.
• Add answers to the three questions your customers ask most often.
• Include a relevant project example or customer review.
• Make sure related pages link to it.

You can take that first step without attending another conference or commissioning a hundred-page SEO audit.

If your website or online store could use some improvements, get in touch with Ballers. We'll take a look, agree on what needs fixing first, and get to work.