How AI recommends you

The simplest and, at the same time, most effective approach is often this: don’t write about what you’d like to talk about, but about what your customers really want to know.

A very good starting point, for example, is:

• Questions from consultation meetings
• Objections raised during sales
• frequent follow-up enquiries by email
• Typical concerns people have before making an enquiry
• Misunderstandings surrounding your offer

It is precisely these that often give rise to the best content. Instead of writing generally about a topic, it is much more effective to answer specific questions such as:

• When is our offer really worth it?
• Who is it suitable for – and who is it not suitable for?
• What are the specific benefits?
• How much does it cost, how much does it save, and what does it improve?
• What mistakes do we see time and time again in practice?

Content like this is closer to the user’s reality — and therefore usually more relevant for AI.

AI favours content that is tangible. In concrete terms, this means: fewer empty claims, more substantiated statements.

A weaker example would be: ‘Our strategy improves your visibility.’

A stronger version is: ‘By revising the content, expanding helpful FAQ sections and creating clearer services pages, we were able to generate significantly more qualified enquiries within six months.’

It’s even better if you can use real figures, timeframes or specific results.

What you should check throughout:

• Where am I still being too general?
• Where am I making a claim without backing it up?
• Where can I add examples, results or scenarios?

Many companies have FAQs on their websites — but they are often too brief, too superficial or too sales-oriented.

Yet FAQs can be particularly powerful if they’re structured in a way that’s genuinely helpful.

Not like this: ‘How long does it take?’
“It depends on the individual case.”

But rather:How long does a relaunch usually take?
“Depending on the scope, a relaunch usually takes between eight and sixteen weeks. If the content, SEO structure and design are already well prepared, it goes faster. Delays often arise where approvals are missing or content is still being developed in parallel.”

That’s specific, clear and useful.

My tip: Take your most important FAQs and turn them into short mini-guides rather than brief, standard answers.

If you want to be recommended by AI, you shouldn’t just say you’re good — you should show how that can be seen.

The following are particularly effective:

• Case studies
• Before-and-after examples
• Typical client scenarios
• Real-world lessons learnt from projects
• Concrete problem solutions

For example:

• What was the client’s initial situation?
• What was the problem?
• What was changed?
• What led to the improved outcome?

Content like this is not only more credible, but also much easier to quote.

Many companies have strong indicators of trust — but they make far too little use of them. If you have good reviews, relevant client projects or media mentions, this information shouldn’t be tucked away on the very last subpage; it should be prominently featured in your communications.

In practical terms, this means:

• highlight good Google reviews on service pages
• incorporate relevant customer testimonials into blog posts
• Highlight references that are relevant to the topic
• Make press reports or mentions visible
• Organise awards and certifications clearly

It’s important not just to display logos, but to follow them up with substantive content that builds trust.

So don’t just say: “We’ve won an award.”

But rather: “Our work has been recognised because …”

It’s not just your own website that counts. It also helps if other trustworthy sources mention you, your business or your expertise. These include, for example:

• Specialist articles
• Interviews
• Press releases
• Guest posts
• Industry portals
• Mentions in specialist media
• well-researched posts on LinkedIn or Reddit

The point isn’t simply to be visible everywhere. The point is to be visible with substance. Helpful insights, clear opinions and professionally robust posts generate precisely the sort of external credibility that is also relevant to AI systems. Here’s another tip from me: it’s better to have a presence on fewer channels, but to do it properly. 

AI is a very useful tool for:

• structural suggestions
• initial drafts
• condensing content
• rewording
• Variations on headings
• Outlines and summaries

What I wouldn’t recommend: copying texts word for word.

Because that’s precisely when you often end up with the sort of text you see everywhere these days: bland, generic, interchangeable — in short: a run-of-the-mill, cookie-cutter piece. That’s why there should always be a human touch at the end:

• add your own opinion
• incorporate practical knowledge
• Give concrete examples
• Check statements
• verify sources
• Make the language more personal
   

The most powerful content is not created by AI alone, but through the right collaboration between AI and humans.

A point that is often underestimated: originality is no longer just a nice extra, but a genuine factor in relevance. If everything sounds like neatly polished, run-of-the-mill output, it lacks precisely what makes content powerful today:

• attitude
• Experience
• Recognisability
• Personality

That doesn’t mean texts have to be artificially ‘different’. But they should have their own distinctive style. Ask yourself this with every text:

• Does this really sound like us?
• Would a customer realise that there’s genuine experience behind this?
• Is this specific enough?
• Is this something any competitor could write in the same way?

If the answer is yes, it usually means there’s still a lack of substance.

What often works very well in practice is surprisingly simple: turning genuine customer queries into genuine content.

One of our clients did exactly that. Instead of writing generic blog posts, they collected common everyday questions and turned them into helpful articles. Simple FAQs were transformed into well-researched content with real practical value. The result: greater relevance, better visibility and even concrete enquiries.

Actually, this is nothing new. At its core, it’s exactly what good marketing has always been about:
staying close to the customer’s needs and creating content that really helps.

If you want to be recommended by AI systems, you shouldn’t try to produce as much content as possible. The key is to create the right content. Content that:

• answers specific questions
• offers genuine value
• builds trust
• is substantiated rather than merely claimed
• and has its own distinctive style

Because that is precisely what distinguishes run-of-the-mill content from content that is truly relevant today. AI can help to make content smarter, more efficient and better. But it should never lead to a loss of humanity, control and a genuine understanding of customers. Because, ultimately, true added value is still recognised by the fact that a person has understood what other people really need.

If you really want to use AI effectively in content marketing, you shouldn’t just try out individual tools; instead, set up a well-organised system. The standards for content have risen significantly. This makes it all the more important to use AI intelligently – with robust workflows and clear rules. Use sensible automation and agent-based processes where they can be applied safely and effectively. In my view, this is precisely what distinguishes the haphazard use of AI from a genuine improvement in quality. I myself use Make.com, amongst other tools, for my content marketing workflows. I’ll be sharing a more in-depth look at this shortly.

And if you need help integrating AI into your content marketing in a structured, efficient way without compromising on quality, then feel free to join one of my one-to-one workshops.

If you’re interested, just send me an email at inga.roser@aufwind-group.de.