The verdict in three sentences
When a trade finance banker, an industrial buyer or a graduate recruit asks AI assistants like ChatGPT about a Frankfurt-based trading house, the answer draws on what the public web says about you: governance pages, published reports and third-party mentions. A trading house that publishes little, or whose details contradict each other across its website, its Impressum and directories, is often described vaguely, out of date, or not at all. In the 2026 market, AI visibility work typically starts with a diagnostic in the 3,000 to 8,000 EUR range and is measured by a dated record of answers to the same questions, never by a promise of being recommended.
What do your counterparties actually ask AI assistants?
A commodity trading house does not sell to consumers. Its readers are counterparties, banks financing the flows, NGOs following supply chains, authorities and recruits. Their questions to AI assistants are verification questions, asked before a call or a due diligence process.
| Who asks | Example question put to an AI assistant | What the answer needs to be able to cite |
|---|---|---|
| Trade finance bank | "Which metals trading houses are headquartered in Frankfurt and publish a sanctions policy?" | Compliance page, published sanctions policy, legal entity |
| Industrial buyer | "Agricultural commodities trader based in Germany with offices in Switzerland and Singapore?" | Up-to-date offices page, consistent addresses |
| ESG analyst or NGO | "Does this company publish a sustainability report or a responsible sourcing policy?" | Report actually published, date, scope |
| Potential counterparty | "Who runs company X in Frankfurt and how long has it existed?" | Leadership page, factual history |
| Trader or analyst candidate | "Commodity trading employers in Frankfurt hiring juniors in 2026?" | Careers page, programmes, working language |
| Group procurement team | "How do I check KYC on a German trading supplier?" | Onboarding process described in general terms |
| Business journalist | "Main trading houses present in the Rhine-Main region?" | Press mentions and industry associations |
None of these questions is about price. They are about credibility. That is the main difference from a law firm or a clinic: for a trading house, the AI assistant acts as a first reputation filter.
Why some trading houses appear in answers and others do not
AI assistants assemble answers from public sources they find consistent. A company mentioned in an association report, whose website clearly describes its activities, and whose address is identical across the Impressum, the commercial register and industry directories, is more likely to be described accurately.
The factors that matter, with no guaranteed recipe:
- Consistent public information: the same legal name, offices and executives everywhere. A closed Geneva office still listed on one page creates confusion.
- Clear factual pages: a "Compliance" page that states what the company actually publishes (sanctions policy, code of conduct, ESG report) rather than slogans.
- Third-party mentions: trade press, industry associations, conferences, public partnerships. Never invented volumes or counterparties.
- Structured data: markup for the organisation, addresses and open roles, helping search engines connect the facts.
- Freshness: a 2021 sustainability report presented as the latest one sends a negative signal.
- Languages: aligned German and English versions stop the assistant from blending two different descriptions.
The German context adds its own constraints. The Impressum is a legal requirement for most commercial websites, data protection falls under the GDPR and the BDSG, supervised by the state data protection authorities. If a group entity carries out a regulated financial activity, BaFin comes into play and every public claim needs review. This is not legal advice.
What the work covers and what it costs on the market
The AI visibility market is young. Offers overlap with SEO, press relations and corporate content. In broad terms, serious work covers a diagnostic of current answers, aligning public information, clarifying governance and compliance pages, structured markup and tracking over time.
| Service | 2026 ballpark (EUR) | Note |
|---|---|---|
| One-off AI answers diagnostic | 3,000 to 8,000 | Depends on number of questions and languages |
| Aligning public information | 2,000 to 6,000 | Website, Impressum, directories, profiles |
| Rebuilding compliance and governance pages | 4,000 to 12,000 | Legal review on top, client side |
| Monthly retainer (content + tracking) | 2,000 to 6,000 per month | Typical German agency quotes |
| Measurement only | 500 to 1,500 per month | Dated record, quarterly report |
| Specialist financial PR | 3,000 to 10,000 per month | Financial and corporate agencies |
These figures come from typical agency quotes and public rate pages, and they vary widely. At Kolonell this work is on quote, after an initial diagnostic.
How to measure without fooling yourself
Honest measurement relies on a dated record. You fix a list of 20 to 40 representative questions, in German and English, ask them of AI assistants at regular intervals and archive the answers. Three indicators are enough: the share of answers mentioning the company, the accuracy of what is said (offices, executives, activities) and whether the company's own official sources are among those cited.
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
| Indicator | Typical starting point | What you track |
|---|---|---|
| Share of answers citing the company | Often 0 to 20% | Quarterly trend |
| Accuracy of information | Frequent errors on offices | Number of errors logged |
| Official sources cited | Rarely the website itself | Presence of site and reports |
| DE / EN consistency | Diverging descriptions | Gaps between the two languages |
| Freshness of facts cited | Data 2 or 3 years old | Date of the facts used |
Answers vary from one session to another: only a series over time means anything.
Mini case study
Illustrative example: Rheinmark Commodities, a 60-person metals trading house in Frankfurt with offices in Zug and Singapore. The head of communications logs 30 questions in German and English. Out of 60 answers, 9 mention the company, which is 15%, and 4 of them still cite an office closed in 2024.
She commissions a diagnostic at 5,000 EUR followed by six months of tracking at 1,000 EUR per month, 11,000 EUR in total. If this saves the credit team just one day of extra due diligence questions per quarter, valued at 1,500 EUR, the saving over a year reaches 6,000 EUR. The real gain is qualitative: accurate information when a bank runs its checks.
FAQ
Can anyone guarantee that AI assistants will recommend our trading house?
No. Nobody controls the answers of AI assistants. You can improve the consistency and clarity of public information, then measure the trend over 6 to 12 months.
Do we need to publish volumes or counterparties to be cited?
No, and that is often impossible for commercial reasons. AI assistants rely mostly on governance, office and compliance facts, which cover about 80% of verification questions.
How long before we see an effect?
Usually 3 to 6 months before corrected pages are picked up, because assistants refresh their sources at different paces. A quarterly record shows it.
Do we need a website in German and English?
For a Frankfurt house dealing with international banks, yes in most cases. Two aligned versions reduce description errors between the languages.
What budget should we plan to start?
A diagnostic sits in a 2026 ballpark of 3,000 to 8,000 EUR on the market. It lets you decide afterwards whether a monthly retainer is useful.
Find out how AI assistants describe your firm. We record what assistants say today about your trading house, in German and English, and scope the corrections worth making. Request a diagnostic quote: reply within 48 hours.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.
