How Velux, Fakro, Drutex and Eko-Okna Are Winning at AI Search (Mostly by Accident)

In 2025, an increasing share of B2B purchase decisions start with a question asked to artificial intelligence. And while most companies still act as if Google were the only source of truth, their prospective customers are increasingly getting their first answers from language models. In practice, that means one thing: if a brand doesn't show up in AI's answers, it simply doesn't exist at the point where a shortlist of suppliers gets built.

I'm writing this report as someone who has spent more than a year working with manufacturing companies on their AI Search visibility. And I see one thing very clearly: most brands aren't "fighting" for that visibility. They don't even know there's a fight happening at all. What shows up in AI today is largely a "leftover" from earlier SEO, PR, trade articles, and old case studies. In other words, their AI "results" are often an accidental byproduct of past strategies, not the outcome of a deliberate plan.

And at the same time, paradoxically, it's AI that's starting to decide who ends up on the radar of distributors, trade partners, and investors. That's exactly why I put this case study together. Visibility in AI Search is becoming a new export advantage for window and door manufacturers. This isn't a trend. It's a change in the rules of the game.

Quick definition

AI Search visibility means having your content, brand, or product show up in answers generated by AI-powered search engines. In practice, it means the AI algorithm is using data from your site as a credible source to construct a direct answer for the user.

Definition by SOLV, AI Search Visibility experts.

1. Technical Foundations (On-Site SEO and AI-Readiness)

Language models don't browse websites the way people do. They analyze code structure, content semantics, the presence of schema, the quality of product descriptions, how developed the FAQ section is, and how predictable the navigation is. In the window and door industry, where products come in countless variants, parameters and configurations, technical foundations are exactly what decides whether a brand is legible to LLMs.

VELUX and FAKRO's sites are good examples of an "AI-friendly" approach: clear content segmentation, transparent user paths, dedicated FAQ sections, readable guides, and materials for professionals. DRUTEX and Eko-Okna have a strong product section, but a less organized educational layer, which makes it harder for language models to extract specific information.

This analysis draws on publicly available FAQ sections, guides, and customer-support content, including:

• VELUX FAQ: https://www.velux.pl/faq
• VELUX customer support: https://www.velux.pl/wsparcie-klienta
• VELUX installation guide: https://www.velux.pl/magazine/montaz-okna-dachowego-kompletny-przewodnik
• Fakro FAQ: https://www.fakro.pl/serwis/faq
• Fakro roof window FAQ: https://www.fakro.pl/okna-dachowe-pytania
• Fakro Innoview FAQ: https://www.fakro.pl/innoview/faq

1.1. Core Web Vitals, UX, and Indexability

For language models, what matters most is:

• load speed,
• visual stability,
• no excess elements blocking rendering,
• predictable URL naming,
• visible category navigation.

VELUX and FAKRO use a structure where educational content is broken into separate subpages, which is a big advantage from an LLM's perspective, since these models interpret short, unambiguous knowledge modules far better than long, catch-all PDFs.

Table: technical foundations

1.2. Structured Data and Content Format

Sites optimized for LLMs have:

• step-by-step guides broken into modules,
• short definitions,
• readable headings,
• logically grouped content,
• Q&A sections,
• technical data in formats that are easy to cite.

VELUX is a good example of a site publishing content in a format AI finds highly usable:

• an installation guide as a standalone subpage,
• a dedicated professionals section,
• guides written in steps and definitions.

Fakro has an extensive FAQ, often dozens of short question-and-answer pairs on a single page, which models interpret very well.

Table: AI Search readiness (content format)

1.3. Takeaways: Who's Best Prepared for AI Search?

1. Fakro: highest legibility for language models, best knowledge architecture.
2. Velux: strong brand, large volume of educational content that AI can easily cite.
3. Drutex: potential, but less educational content and less logical grouping.
4. Eko-Okna: a strong PR layer, but missing the technical, educational depth LLMs need.

2. Content and AI-Friendly Formats

What language models are able to cite depends heavily on content format. LLMs don't pick brands "intuitively," they pull the fragments from the web that are most legible to them: definitions, answers to questions, step-by-step instructions, guides, and short knowledge modules. That's exactly why generic product descriptions aren't what drives visibility, educational and expert content structured in an AI-friendly way is.

VELUX and FAKRO's sites reflect that trend well. They publish content in formats language models interpret as valuable information sources: installation guides, step-by-step instructions, sections for professionals, extensive FAQs, modular subpages with answers, and precise product parameters. All of that makes it easier for AI to build synthetic answers to user questions.

DRUTEX and Eko-Okna still have an extensive base of product content, but relatively little educational content in LLM-friendly formats. Language models also factor in current media context about a company, including news about financial trouble and major workforce cuts in 2024-2025. Signals like that affect how AI assesses a brand's credibility, which makes Eko-Okna cited less often in expert-style content than Velux or Fakro.

Examples of AI-friendly content analyzed in this section:

• Complete VELUX roof-window installation guide: https://www.velux.pl/magazine/montaz-okna-dachowego-kompletny-przewodnik
• Portal for professionals (instructions, installation, specifications): https://www.velux.pl/profesjonalisci
• FAKRO FAQ (technical and service questions): https://www.fakro.pl/serwis/faq
• FAQ on roof windows: https://www.fakro.pl/okna-dachowe-pytania
• Innoview FAQ (windows, doors, parameters): https://www.fakro.pl/innoview/faq

2.1. What Content Do AI Models Like Best?

The analysis shows that in the window and door industry, the best-performing formats are:

• FAQs in a question-to-short-answer format
• step-by-step installation guides,
• "everything about the product" sections,
• dedicated subpages for technical parameters,
• instructions for professionals,
• modular educational content (one topic per page).

These are the formats LLMs most often draw on when generating answers to customer questions about installation, insulation, window parameters, product selection, and comparing solutions.

2.2. Scoring the AI-Friendly Content Layer

2.3. Takeaways: Who Builds the Best AI-Ready Content?

1. Fakro: the richest educational base, the best format for LLMs, high density of Q&A content.
2. Velux: a large volume of guides and how-tos, very good materials for professionals.
3. Drutex: potential in its product catalogs, but lacking modular educational content.
4. Eko-Okna: strong PR materials, but little of the structured knowledge models need.

2.4. Why AI-Friendly Content Matters for LLM Visibility

In the AI Search era, educational content has become the new foundation of visibility. Language models favor:

• short answers,
• unambiguous definitions,
• clearly segmented content,
• a complete heading structure,
• separate subpages for individual topics.

Brands already investing in modular expert content (like Velux and Fakro) have a much better shot at being cited in AI answers, while companies relying mainly on PR materials or generic product descriptions fade out of language models' field of view faster.

3. Trust Signals (Reviews, Certifications, Awards)

From a language model's perspective, trust signals, customer reviews, awards, certifications, industry recognitions, and media presence, carry real weight in assessing a brand's credibility. LLMs analyze not just the content on a company's own site, but also external information: reviews, forum discussions, trade articles, reports, and press materials. The more credible and consistent reputational signals there are, the more likely language models are to start recommending a brand in answers about products, categories, or market comparisons.

Among the largest window and door manufacturers, reputational visibility varies significantly. Velux and Fakro have a strong presence in trade media, reports, and national publications, building a positive image as long-standing innovators. Drutex has numerous business awards, alongside a wide spread of user opinions, which language models interpret as a "mixed" reputation. Eko-Okna still ranks high in sales reports and has won industry awards, but its reputation is currently weighed down by extensive media coverage of financial trouble, layoffs, and high staff turnover in 2024-2025. Language models factor those signals in, which is why AI more often assigns Eko-Okna a reputation tagged as unstable or risk-associated, making LLMs less inclined to present it as a "safe choice" in an expert context.

3.1. Customer Reviews and External Ratings

Language models treat reviews as an important reputational element, especially when they come from independent platforms. Drutex has one of the largest review bases on the Oknotest platform, where customers rate products and the installation process: https://oknotest.pl/okna-opinie/firma/drutex

Construction forum discussions include both positive comments about PVC window build quality and critical opinions about service and installation:

• ForumBudowlane: https://www.forumbudowlane.pl/okna-i-drzwi/jakiej-firmy-okna-oknoplast-petecki-dako-drutex-t60751/30
• Oknonet forum: https://forum.oknonet.pl/viewtopic.php?t=19912

AI models synthesize this varied experience as a complex reputation: large scale, a known brand, but with clearly mixed user experience. Velux and Fakro have fewer typical consumer reviews, but a much stronger educational and expert presence. AI perceives their quality as stable, and models more often cite these brands in the context of "proven solutions," "good reviews," and "high product quality."

3.2. Certifications and Industry Awards

Official signals, awards, rankings, and recognitions, are among the strongest reputational elements, because AI treats them as objective quality indicators.

VELUX, a strong expert and employer position

• "Market Leader in Windows and Doors 2024" (roof and wood windows): https://prasa.velux.pl/?h=1&t=nagrody
• Top Employer 2024 certification: https://prasa.velux.pl/top-employer-2024-dla-grupy-velux
• Top Employer 2025 certification: https://builder4future.pl/2025/02/12/35-lat-troski-o-pracownikow-grupa-velux-w-polsce-z-tytulem-top-employer-2025

In AI's eyes, this reads as a stable, high-quality, credible manufacturer.

FAKRO, innovation, design, and premium products

• Iconic Awards, Best of Best (DXW roof windows and the INNOVIEW line): https://www.fakro.pl/informacje-o-firmie/nagrody-i-wyroznienia
• Innovator award (WPROST 2022) for HST SKY terrace doors
• Marketing competition awards (like Builder Super Power)

AI more often presents Fakro as an innovative brand, technology-driven and focused on build quality.

DRUTEX, economic and market awards

• Forbes Diamonds
• European Medal (for example, for IGLO EDGE windows)
• Recognitions in Builder and Forbes rankings: https://www.drutex.eu/pl/aktualnosci

Language models interpret this data as a signal of "a large, stable manufacturer with a strong market position."

EKO-OKNA, a leader in growth and scale

• Market Leader in Windows and Doors 2022: https://ekookna.pl/ekopl/pressroom/eko-okna-sa-z-tytulem-lidera-rynku-stolarki-2022
• Market Leader in Windows and Doors 2023: https://ekookna.pl/ekopl/pressroom/eko-okna-sa-z-tytulem-lidera-rynku-stolarki-2023

AI interprets Eko-Okna's reputation as mixed. On one hand, the company appears in trade reports as a major manufacturer; on the other, language models pick up numerous press reports on group layoffs, financial trouble, sales declines, and workforce difficulties from 2024-2025. Signals like these lower the level of trust and can reduce how often the brand appears in language model answers.

3.3. Media Presence and Expert Positioning

In AI Search, it also matters how often a brand appears in industry, news, and expert contexts. Velux and Fakro perform strongest here:

• intensive coverage in trade press,
• a high volume of educational materials,
• active expert portals,
• press releases about technological innovation.

Drutex has extensive coverage on investment, exports, and company results, which builds an image of stability. Eko-Okna is often cited in the context of scale, growth, and market-leader position, which AI reads as a signal of "a dominant player."

3.4. Takeaways: Reputation Signals as the Foundation of AI Search Visibility

1. Velux: the strongest combination of awards, publications, certifications, and PR.
2. Fakro: a leader in quality and innovation, very strong product and technology signals.
3. Drutex: large scale of operations and numerous business awards, but mixed user opinions.
4. Eko-Okna: still a major market share and brand recognition, but current financial trouble and large-scale layoffs are clearly dragging down the company's reputation. Language models factor in these signals, which is why the brand appears less often in AI recommendations around quality and stability.

For AI models, these repeatable, external, multi-source signals are exactly what forms the basis for assessing the credibility of window and door manufacturers. AI-focused educational content works best when it's reinforced by a solid reputational foundation.

4. External Presence (Media, Reports, Rankings)

Window and door manufacturers' visibility in AI Search depends not just on the quality of content on their own sites, but also on how they're presented in trade media, reports, market analyses, and press releases. Language models draw on a broad range of sources, so presence in external publications has a direct impact on how a company is perceived and how often it appears in AI answers.

A strong, repeated media presence reinforces brand credibility, while a lack of consistent external narrative, or a presence that's mostly PR material, weakens expert-context recognition in AI Search. Analysis of the four largest brands shows the differences between them are significant.

4.1. Media Presence Analysis: Comparing the Brands

Velux

Velux has one of the strongest media presences in the industry. The brand regularly appears in national press, trade media covering construction, and expert reports on the roof-window market. Press releases cover both technological innovation and topics related to ESG, sustainability, and organizational culture.

Example sources:

• Awards and recognitions: https://prasa.velux.pl/?h=1&t=nagrody
• Top Employer 2024 and 2025: https://prasa.velux.pl/top-employer-2024-dla-grupy-velux, https://builder4future.pl/2025/02/12/35-lat-troski-o-pracownikow-grupa-velux-w-polsce-z-tytulem-top-employer-2025

AI interpretation: a strong, stable brand, high quality, high credibility.

Fakro

Fakro consistently builds media presence through coverage of product innovation and prestigious industry awards. Numerous recognitions, such as Iconic Awards, along with frequent trade press coverage, reinforce the perception of the company as a technology expert and premium manufacturer.

Example sources:

• Awards: https://www.fakro.pl/informacje-o-firmie/nagrody-i-wyroznienia
• Innovation recognitions (DXW, Innoview)

AI interpretation: an expert, innovative brand, high product and technology quality.

Drutex

Drutex dominates in business-focused coverage, frequently appearing in the context of exports, investment, plant expansion, and financial results. Its media image is thus strongly tied to scale of operations, and less to educational or expert content.

Example sources:

• News and awards: https://www.drutex.eu/pl/aktualnosci

AI interpretation: a large manufacturer with a stable market position, but a less developed layer of technical, expert publications.

Eko-Okna

For years, Eko-Okna focused on PR communication and a strong presence in sales reports, including reports from the Polish Industry Analysis Center (CAB). In recent years, though, numerous press materials have appeared covering financial trouble, layoffs, and revenue declines, affecting the brand's overall media image.

Example sources:

• Market Leader in Windows and Doors 2022 and 2023: https://ekookna.pl/ekopl/pressroom/eko-okna-sa-z-tytulem-lidera-rynku-stolarki-2022, https://ekookna.pl/ekopl/pressroom/eko-okna-sa-z-tytulem-lidera-rynku-stolarki-2023
• Reports on financial and workforce trouble (2024-2025): Business Insider, Onet, industry forums

AI interpretation: large scale, high brand recognition, but a reputation weighed down by recent negative media coverage.

4.2. Table: Media Presence 2023-2025

4.3. How Does AI Interpret External Publications?

Language models don't evaluate brands on a single parameter, they combine positive, neutral, and negative data. The interpretation typically looks like this:

• Velux: high stability, trust, long-standing position, frequent citations.
• Fakro: innovation, design, quality, high volume of expert citations.
• Drutex: large scale, stability, but less technical knowledge, moderate citation volume.
• Eko-Okna: mixed data, positive trade reports plus negative media coverage, lowered informational credibility.

4.4. External Authority Score (synthetic)

4.5. Final Takeaway

Media presence has a direct impact on AI Search visibility. Language models favor brands that are stable, frequently discussed in media, with a large volume of expert content and repeated quality signals. Velux and Fakro meet these criteria best; Drutex partially; Eko-Okna, despite its large scale of operations, is currently dealing with a reputation weighed down by media coverage, which affects how AI evaluates the brand.

5. Three Moves Any Window and Door Manufacturer Can Make in 30 Days to Improve AI Search Visibility

Visibility in AI in the window and door industry isn't a matter of chance. It's the result of the content, data, and signals language models find on the web. Below are the three fastest steps manufacturers of windows, doors, facades, and lift-and-slide systems can take, whether they operate domestically or in export markets.

1. Publish one solid case study, ideally with numbers and a real result

In the window and door industry, nearly every customer asks about two things: parameters and specific outcomes for other clients. AI models work the same way, they favor content that can be summarized in numbers and facts. One good case study can do more for a brand's visibility than twenty generic product descriptions.

What works best in this industry?

• installations for major developers or investors,
• custom lift-and-slide system rollouts,
• export projects (for example, into Scandinavian, German, or UK markets),
• optimization results, like cutting installation time by 27%,
• data on improved insulation, acoustics, or energy efficiency.

Example structure:

• Problem: for example, "the project required windows with a Uw below 0.8"
• Approach: technology, configuration, timeline
• Solution: systems, product lines, process
• Result: numbers. Always numbers.
• A client quote or takeaway for investors

For language models, that format sends a clear signal: "This company has proof, not just a catalog."

2. Earn reviews on external platforms: GoodFirms, LinkedIn, trade portals

In the window and door industry, reviews are an absolute foundation, especially for companies working with distributors, foreign partners, and large investors. AI sees this too.

Models treat reviews as measurable proof of credibility. If reviews like these show up online:

• "the company improved the airtightness of an existing structure by X,"
• "delivered windows in 12 weeks despite complex logistics,"
• "solved an acoustic issue in an office building,"

AI treats that as a valuable signal.

Where to earn reviews?

• Clutch (works great for export-focused manufacturers)
• GoodFirms (strong influence on LLMs)
• LinkedIn Recommendations (frequently cited in AI answers)
• trade portals

The format that works best:

1. project description,
2. problem,
3. result,
4. rating.

That structure is ideal for AI, since models can easily pick out the pattern and fold it into an answer.

3. Build a series of Q&A content in a question-to-answer format tailored to the window and door industry

This is the simplest way to improve AI Search visibility, and one of the most underrated. Language models often build answers from content written in a Q&A style. If a company has materials like that publicly available, AI is happy to draw on them.

Example questions that work great in this industry:

• "What does the Uw coefficient actually mean in practice?"
• "What should you look for when choosing windows for an energy-efficient building?"
• "Which manufacturers offer the best lift-and-slide systems for export markets?"
• "How do aluminum, PVC, and wood windows compare?"
• "How do you maintain windows with large glazed surfaces?"

Every answer should be:

• short (50 to 120 words),
• specific,
• grounded in experience,
• free of marketing language,
• with a clear, practical takeaway.

When an AI model sees a structure like that, it registers: "This is citable."

An added benefit is that Q&A content like this:

• works perfectly as FAQ sections,
• can be published as LinkedIn posts,
• performs well in export content (across language versions).

Summary

In the window and door industry, these three moves aren't just "quick wins." They're the foundation of brand visibility in a world where purchasing decisions start with a question asked to artificial intelligence.

Case study, then reviews, then Q&A is the sequence that lets manufacturers build a genuine edge over competitors, over imitators, and over companies that haven't yet noticed AI has become a new layer of search.

Sources and Further Reading

VELUX content (FAQ, support, guides, installation)

• FAQ and customer support: https://www.velux.pl/faq
• Customer support knowledge hub: https://www.velux.pl/wsparcie-klienta
• Complete roof-window installation guide: https://www.velux.pl/magazine/montaz-okna-dachowego-kompletny-przewodnik
• Portal for professionals (architects, roofers, installers): https://www.velux.pl/profesjonalisci
• Press releases, awards and recognitions: https://prasa.velux.pl/?h=1&t=nagrody

FAKRO content (FAQ, service, technical detail, awards)

• Service and FAQ: https://www.fakro.pl/serwis/faq
• FAQ on roof windows: https://www.fakro.pl/okna-dachowe-pytania
• Innoview FAQ: https://www.fakro.pl/innoview/faq
• Awards and recognitions: https://www.fakro.pl/informacje-o-firmie/nagrody-i-wyroznienia

EKO-OKNA content (events, industry awards, releases)

• Market Leader in Windows and Doors 2022: https://ekookna.pl/ekopl/pressroom/eko-okna-sa-z-tytulem-lidera-rynku-stolarki-2022
• Market Leader in Windows and Doors 2023: https://ekookna.pl/ekopl/pressroom/eko-okna-sa-z-tytulem-lidera-rynku-stolarki-2023
• News and press releases: https://ekookna.pl/ekopl/pressroom

DRUTEX content (awards, reviews, business releases)

• Customer reviews on Oknotest: https://oknotest.pl/okna-opinie/firma/drutex
• Customer discussions on construction forums: https://www.forumbudowlane.pl/okna-i-drzwi/jakiej-firmy-okna-oknoplast-petecki-dako-drutex-t60751/30
• Window discussions on the Oknonet forum: https://forum.oknonet.pl/viewtopic.php?t=19912
• Business awards and recognitions (Forbes, Builder, European Medal): https://www.drutex.eu/pl/aktualnosci

Industry reports and sources

• Polish Industry Analysis Center (window and door industry reports): https://cab-badania.pl
• Industry forum, window and door market: https://forumbranzowe.com
• Builder, awards, rankings, macro data: https://builderpolska.pl

Market context: AI Search and LLM visibility

• Installation guide as an example of AI-friendly content (VELUX): https://www.velux.pl/magazine/montaz-okna-dachowego-kompletny-przewodnik
• Example knowledge portal for professionals (AI readability): https://www.velux.pl/profesjonalisci
• Example of a well-structured FAQ (FAKRO): https://www.fakro.pl/serwis/faq

Further reading (AI, LLMs, brand visibility)

• Analysis of the role of content quality in AI Search (OpenAI Research Blog)
• Schema.org documentation: FAQPage, QAPage, WebPage, CreativeWork
• A guide to how LLMs build citability
• Industry reports on the growth of window and door market segments
• Press materials from the construction and installation industry

A Strategy Built on Market Standards

Our work isn't based on guesswork, it's grounded in hard data and official guidelines from industry leaders. We analyze key documentation to make sure our strategies aren't just effective, but aligned with best practice and upcoming regulation, so results hold up over time.

• Google's Search Quality Rater Guidelines — SOLV applies these guidelines to build content grounded in E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness), the principles that drive credibility in the eyes of AI algorithms.
• Google Search developer documentation — for SOLV, this is the foundation of technical optimization that lets AI engines unambiguously understand the structure and context of information on a page.
• Semrush's "State of Search" report — analysis of this report's data lets SOLV align strategy with real trends and user behavior in the age of conversational search.
• The EU AI Act — SOLV monitors this regulation to build strategies aligned with upcoming requirements around transparency and ethical use of data by AI systems.

Want to know where your own brand stands in AI Search before your competitors figure it out? Start with our AI Search Audit.