PL / EN Free consultation
Case study

MOL case study: from zero to #2 in ChatGPT in 7 months

In March 2026, MOL, a pergola and awning manufacturer from Poznań, didn’t appear in AI answers about its industry. By October, ChatGPT and Gemini list it #2 among patio awning manufacturers in Poland, and for an English export query it ranks #1 in Perplexity and Google. Here’s what we did and what came of it.

Author: Published: Updated: 5 min read

Starting point: a brand AI didn’t know

MOL is a manufacturer of pergolas, awnings, roller blinds and insect screens from Poznań, selling in Poland and abroad (mol.net.pl). We started working together in March 2026. For questions customers and distributors ask, such as “patio awning manufacturer in Poland” or “pergola manufacturer for export”, AI models answered with a list of competitors. MOL wasn’t in those answers.

The goal was simple: get ChatGPT, Gemini, Perplexity and Google to recommend MOL when someone looks for a manufacturer in this niche – both to customers in Poland and to B2B partners abroad.

What we did

We worked in three areas that decide whether AI knows and recommends a brand:

1. A website AI understands

We restructured MOL’s website so AI models can easily read who the company is, what it makes and for whom – from content structure to a clear, consistent company description AI can take facts from.

2. Content that answers customer questions

We created expert content answering the questions customers and distributors ask in the buying process – with the specifics AI likes to quote.

3. Presence beyond the website

AI models trust what other sources say about a company. So we made sure knowledge about MOL is available beyond the company’s own website too.

Results in AI answers

Example 1: “Patio awning manufacturer in Poland”

WhenMOL’s presence
March 2026Absent
July 2026Only occasionally
October 2026#2 in ChatGPT (of 6 companies) and #2 in Gemini (of 5 companies)

ChatGPT describes MOL as “a very interesting choice for B2B” and cites specifics from the company’s website: 23 models, over 120 fabric patterns and an export department. For balcony awnings, MOL is #1 in ChatGPT.

ChatGPT answer (in Polish) to “Patio awning manufacturer in Poland”: MOL – pergola and awning manufacturer from Poznań in 2nd place, described as a very interesting choice for B2B, with MOL’s website among the sources.
ChatGPT, 8 October 2026 (Polish query): MOL #2, MOL’s website among the cited sources.

Example 2: export and an English query

For “pergola awning manufacturer from Poland export”, MOL was absent in June. In October:

  • Perplexity: #1,
  • Google: #1,
  • ChatGPT: #1–#3 (depending on the test),
  • mol.net.pl is cited in every engine.
ChatGPT answer to “pergola awning manufacturer from Poland export”: MOL from Poznań in 1st place as a supplier of pergolas, awnings, ZIP blinds and insect screens with an export department.
ChatGPT in English: MOL #1 among Polish pergola and awning manufacturers for export.

Traffic from AI: from zero to about 300 visits

Before we started, nobody came to MOL’s website from AI tools. Within a few months there were about 300 visits, mainly from ChatGPT. It’s high-quality traffic: over half of these visitors stay and read.

The real impact of AI is bigger, because visits from AI answers in Google Search (AI Overviews) are counted as regular Google traffic.

Lessons for your company

  1. Results come in months, not years. The first AI traffic appeared within weeks, and #2 in ChatGPT after 7 months of work.
  2. Content must answer questions from the buying process. Guides with specifics work best: models, parameters, tables, FAQ.
  3. The brand must exist beyond its website. Independent sources give AI confirmation that the company can be trusted.
  4. The company description matters. AI quotes facts that are easy to find on the website: specialization, number of models, export.
  5. Measure regularly. A fixed set of questions checked every month shows progress before it turns into traffic.

How we measured

We checked visibility regularly with the same set of questions in ChatGPT, Gemini, Perplexity and Google. Results come from checks between July and 8 October 2026. AI answers vary with the account, location and timing, so we report results from specific dates. Learn how to run such a test yourself in How to check if AI recommends your business.

Frequently asked questions

How long did it take for MOL to appear in AI answers?

The first visits from AI came within weeks. Seven months after we started, MOL was #2 in ChatGPT and Gemini for patio awning manufacturers in Poland.

Which actions had the biggest impact?

Combining a website prepared for AI with regular content that answers customer questions and brand presence in external sources.

Are such results possible in other industries?

Yes. The mechanism is the same in every industry: AI recommends companies that are clearly described and present in trustworthy sources. The pace depends on competition and content quality.

Kacper Bartoszak
Entrepreneur, owner of the PEOPLE HATE ADS* agency and SEO for AI. Has worked with brands such as Lech Poznań, Binance and Danone. About the author →