Radyant

osapiens

Growth Partner (External)

Hilo by Aktiia

Planeco Building

Distart

radicant bank

Sirius

Enter

finally

Digitail

SEA & SEO for Startups and Scaleups | Founder @ Radyant | External Partner for Enter, ToolSense, Vamo, peers. and 30 more

Radyant is a growth agency specialized in SEO, and AI Search. We help startups and scaleups dominate Google, ChatGPT & Co. to unlock wasted growth potential and build a competitive moat.

References:

⚒️ osapiens (B2B SaaS, Series B)
🤝 Partnership since 09/25

🩸 Hilo (B2C HealthTech, Series B)
🤝 Partnership since 07/25

🌍 Enter (PropTech/ClimateTech, Series B)
🤝 Partnership since 09/22

🏗️ Planeco Building (PropTech, Bootstrapped)
🤝 Partnership since 03/25

🐶 Digitail (B2B vSaaS, Series A)
🤝 Partnership since 09/24

👩‍⚖️ Mineko (LegalTech, Bootstrapped)
🤝 Partnership since 12/23

🎧 Sirius (EdTech, Seed)
🤝 Partnership since 10/23

💳 Finally (FinTech, Series B)
🤝 Partnership from 03/25–09/25

🔥 Vamo (PropTech/ClimateTech, Seed)
🤝 Partnership since 01/23

⚙️ ToolSense (B2B SaaS, Series A)
🤝 Partnership 08/21–05/25

🏠 VARM (PropTech/ClimateTech, Seed)
🤝 Partnership 01/23–04/25

📱 adjoe (AdTech, €100M Bertelsmann funding)
🤝 Partnership 09/21–09/22

Grokipedia is one of the most concerning developments for information integrity I've seen in years. And most people haven't even noticed yet. Here's what's happening.

Grokipedia is Elon Musk's alternative to the (his words) "left-biased" Wikipedia. 

"The goal of Grok and Grokipedia is the truth, the whole truth and nothing but the truth."

As someone who works in AI and online information discovery, I need to show you what I'm seeing. Because this isn't about false information. It's about something more dangerous. Systematic mass manipulation through editorial framing.

I compared articles on Grokipedia vs Wikipedia. Same person. Completely different framing.


_________________________

Here's what I mean:

1) George Floyd

Grokipedia leads with: "American man with a lengthy criminal record including convictions for armed robbery, drug possession, and theft."

Wikipedia leads with: "African-American man who was murdered by a white police officer in Minneapolis during an arrest."

2) Joe Biden

Grokipedia leads with age concerns, cognitive questions, and controversies.

Wikipedia's leads with his career accomplishments, positions held, and legislative history.

Same facts available. Completely different editorial choices about what to lead with.

This is the manipulation technique: The facts aren't wrong. That's what makes it effective.

It's about what you choose to put first. What you emphasize. What you minimize.

Lead with someone's criminal record vs lead with how they've been murdered.

Lead with controversies vs lead with accomplishments.

That's how you shape perception.

_________________________

Why this is so concerning:

Wikipedia is one of the most cited sources by AI models, also used for training. If Grokipedia becomes an alternative citation source that AI tools trust, we're talking about Musk-framing at a massive scale.

Billions of AI responses. Shaped by systematic editorial choices, explicitly targeted against Wikipedia, which is supposedly "controlled by far-left activists" (quote Musk).

This isn't about left vs right. This is about information integrity.

My ask: Go compare for yourself.

Look up George Floyd on both platforms. Then Joe Biden. Then anyone else. See the pattern. See what they choose to lead with.

Form your own opinion before it is formed for you.

What do you see when you compare them?

Niklas Buschner

Founder @ Radyant | SEO & AI Search Growth Partner | Host @ Masters of Search

Pipedrive does (reportedly) $200M ARR competing against Salesforce and HubSpot. A key growth driver has always been Search. Jessica and I spent 10h+ analyzing their SEO & AI Search strategy, so you don't have to.

Key strategies behind their success:

▸ The blog is the dominant acquisition pillar: With 700+ indexed pages, it's their largest organic asset routing users into product, industry, and comparison pages. The blog does the heavy lifting, comparison pages close the deal.

▸ Outcome-first positioning: They sell the post-solution state before explaining mechanics. Not "Pipedrive has X feature that does Y" but "Want to close more deals? Here's how..." The product becomes a natural extension of existing desire.

▸ Conversion follows value: The insight that excited me most. Pipedrive doesn't push trial CTAs every 200 words like most B2B SaaS. They wait until the reader gained something useful (a template, a clarified concept) before offering a next step.

▸ Strong fundamentals, strategic gap: They rank exceptionally well on classic CRM queries (pipeline, sales CRM, integrations). But they have minimal presence on newer intent clusters like AI-led sales topics (lead scoring, predictive analytics, revenue intelligence). They're visible where demand is already defined, but not where new demand is forming.

▸ AI search: Solid foundation, clear opportunity: Pipedrive appears in ~14% of AI citations, mostly from deep how-to guides and comparison pages. These get lifted because they contain actionable steps and decision logic, not just surface definitions.

Full breakdown includes a Miro board with traffic patterns, content clusters, and AI search citation mapping.

Link in comments. 👇

Reddit is trashing ChatGPT Atlas after 10 minutes. So I used it as my standard browser for 72 hours. Here's my assessment.

A quick excerpt of what I tested Atlas for:
– Building target account lists (research → Google Sheets entry)
– Product comparison research (electric cars)
– Scrolling through LinkedIn and doing additional research on some posts
– Asking follow-ups to YouTube videos
– Checking pinned emails for things to delegate to the agent
– Scheduling focus time in my calendar for certain tasks

__________________________

My observations:

1) It feels unintuitive and lacks guidance

The experience is ChatGPT and a browser, but doesn't feel intuitive. It lacks the early Apple flair where you'd open and immediately understand how to use it. You get a few suggestions to try things, but you need a lot more guidance on how to get value from it.

2) ChatGPT in the sidebar is nice

Having ChatGPT in the sidebar is genuinely useful. Asking context or follow-up questions on whatever you're viewing without copy-pasting into a separate tab eliminates friction. This is one thing they got right.

3) Sensitive mode is good but incomplete

When I asked Atlas to scan my starred emails for actionable items, it automatically activated sensitive mode. It only works while I'm watching the tab. Shows they understand the risk profile of different tasks. But when I used agent mode to add things to my calendar, it didn't activate.

4) Agent mode doesn't activate itself intelligently

Even when you clearly articulate actions, it doesn't recognize when it should switch to agent mode. You end up in circular conversations where it tries to answer questions instead of taking action. You always have to manually activate the agent. This should be automatic based on intent.

5) Agent execution is slow and makes critical errors

It created a decent Google Sheets table, then messed it up when trying to add more data. Multiple back-and-forth attempts made it worse. I can see it working for simple routine background tasks in another tab, but it falls apart too easily.

6) Visual context doesn't work automatically

ChatGPT in the sidebar can't see your viewport, only the full page content. It only understands what you're focusing on if you manually activate agent mode. This makes interactions feel unnatural. You're looking at something, asking about it, and ChatGPT has no idea what you mean.

7) Cross-tab context is missing

Every sidebar conversation lives in the current tab, unless you manually select it elsewhere. I guess nobody is doing that. Research in one tab, open another, context disappears. Taking context with you between tabs should be table stakes for any browser trying to be your "agent."

__________________________

I see potential, but it still feels very experimental tbh.

Have you tried Atlas yet? What did you find?

Nov 12, we‘ll do an exclusive 20 people AI Search dinner at 893 Ryōtei, one of Berlin’s finest Asian restaurants (you wouldn’t guess from the exterior). Only 6 seats left. 👀 

We’re bringing together founders and marketing leaders actually executing on AI search. Not theory, real implementations.

14 seats are already confirmed from people at Peec AI, Pipedrive, HubSpot, 1Komma5, Personio, getpress and more!

Format: Dinner from 6-8 PM, then whoever’s still up joins for drinks in a Berliner Kneipe just around the corner.

Our goal for the group: to pass the airport test.

(If you have no idea what I’m talking about, check out my recent Masters of Search episode with Andy - link in comments)

If you’re a founder or marketing leader exploring AI search and want in, shoot me a DM.

No alternative text description for this image

This isn't your average "please sign up to my webinar" post. I'm so excited to join Josh and AirOps on Friday to talk AI-powered content workflows, because I haven't been this excited about a tool since playing with LEGO as a kid.

Let me explain.

In July, I was researching content automation platforms for Radyant, and stumbled on AirOps. Checked their case studies, feature pages. Quickly ended up in the documentation.

Before trying it. Before the demo. Just from reading what they built, I sent this to my team:

"Holy sh*t, this could actually be the platform I have dreamed of for the last 6 months."

That feeling is rare.

We had the demo the week after. Probably the easiest sales call AirOps ever had. I just wanted the account as fast as possible. Didn't really care about the pricing.

Two months later, it's our core content engineering platform.

Here's what changed:

Before AirOps, we were already doing interesting things with AI-powered content creation. Had Custom GPTs or dedicated Claude Projects, came up with the Superprompt concept, started building first agentic content workflows with Zapier.

You have to know, I love Zapier. But it felt like using a Swiss army knife when you need a specialized power tool.

Example: I wanted to integrate our client's sitemap, create a vectorized knowledge base, and automatically identify internal linking opportunities.

With Zapier, this got complex fast. Multiple steps, API calls, data transformations. PAIN!

With AirOps? Built-in knowledge base. Sitemap auto-syncs weekly. Vector search returns the best internal linking opportunities instantly.

That's the builder feeling I'm talking about. Not just using a tool. Actually creating with it.

That's why this webinar matters to me.

We're not doing generic "here's how AI content works" slides.

We're showing:

→ The exact workflow how we built 247 location-specific pages with AI

→ Our content agent infrastructure that ships 100+ premium editorial pieces monthly

→ The actual 10,000+ character system prompts we're using and how we're engineering the right context to go with them

If you've been wondering how to move from "using AI to speed things up" to "building actual content engineering systems," this is for you.

Friday, October 24, link in the comments. 👇

AI is expanding Search, not replacing it. Here’s a little anecdote my colleague Jessica (Head of Growth @ Radyant) told me when we had lunch to prove that point.

She goes to the drugstore.

Standing in the aisle.

Dozens of skincare products.

The classic paradox of choice.

She pulls out her phone.

Takes a photo of the shelf …

… and opens ChatGPT.

“Which of these products is best for me and why?”

ChatGPT gives her a recommendation.

Explains differences, checks reviews.

She buys what ChatGPT recommends.

The point of story?

She never did this with Google.

It’s a completely new thing.

We’re assuming people use AI search the same way they used Google. Just asking questions they would’ve typed into a search bar.

But the reality is, People are inventing entirely new use cases.

Which means: If you’re just converting your keyword list into prompt variations, you’re optimizing for the wrong thing.

Here’s what to take from that:

Go ask your team how they use ChatGPT.

Ask your customers.

Ask your users.

You’ll be surprised.

Don’t just ask: “what questions do you ask”

Ask:
- ”how are you using it?”
- ”what did you do yesterday?”
- ”if you check your last 5-10 conversations, what were they about?”

That’s where the real opportunities are.

What new AI behaviors are you discovering that didn’t exist before?

I was invited to the new podcast »narrative now« by the leading European PR agency getpress to chat with founder Maximilian Ziche about all things AI. [🇩🇪 only]

It was a lot of fun and we covered some pretty interesting and I‘d say also non-obvious things.

We spoke about:

- Google vs ChatGPT
- Why LLMs actually hallucinate
- How AI sees the world (aka biases)
- How AI is already used for propaganda
- Why local models are the actual threat
- How AI propaganda could be stopped
- Why we might need content trust seals

and a lot more!

Link to the episode in the comments 👇 

Let me know how you liked it!

"Talk to your customers" is the most ignored advice in SaaS marketing. It's also what Surfer (bootstrapped to €15M ARR) used as their unfair advantage. Here's how they did it (and at which scale) and how you can get started with small steps too.

Most marketing teams build content strategies based on keyword research tools and competitor analysis.

Surfer did something different.

Their founders did 50 calls per week, 30-60 minutes each.

That's already a full work week, just listening to prospects or customers talk about their problems, needs, and how they imagine a solution.

The result? Content that actually resonates because it uses customer language. SEO strategies that target real pain points, not just high-volume keywords. Product positioning that connects immediately because it came directly from user conversations.

This depth of customer understanding became impossible for competitors to replicate.

______________________

You don't need to start with 50 calls per week. Here are 3 simple steps to build customer understanding into your content process:

Step 1: Join at least 2 sales calls every month silently, just listening. Pay attention to how prospects describe their problems and what language they use.

Step 2: Schedule at least 2 user interviews every month. Ask about their challenges, and how they are actually using your product. Learn what they were doing before, what changed, and how the solution manifests in their daily life.

Step 3: Set up a regular sync between marketing, sales, and product (if you haven't already). Share insights, roadmaps, narratives, and customer language across teams.

The magic isn't in the number of calls. It's in the discipline to do it consistently and actually use what you learn in your content.

______________________

Full conversation with Tom Niezgoda is live on YouTube. Link in comments. 👇

How many customer conversations did you have last month?

PS. When we onboard new clients at Radyant, the first thing we do is dive deep into understanding their users. We join sales calls, analyze support tickets, review customer feedback, everything available. You can't create content that converts without truly understanding who you're speaking to.

Can you guess where we‘ll do our next event? 🍣 🍸 😁

Full details dropping soon.

Only so much for now:

- It will be about AI Search
- There will be awesome people
- We have 7 seats left (w/o even announcing)

If you’re in Berlin on November 12, and wanna join, hit me up!

STOP starting your content workflows with keyword research. If you don’t trust me, trust someone way smarter than me — and learn why it’s just one piece of the puzzle. New #MastersOfSearch episode with Margarita Loktionova (Content Marketing Lead @ Semrush) is LIVE now.

Here's what we covered: 
→ Why customer research should come before keyword research
→ How to use keywords to validate demand, not generate ideas
→ What “net-new value” content actually looks like
→ Why original research scales distribution across AI, PR & social
→ How research turns your brand into a source, not a commentator

🎧 Full episode in the comments 👇

This conversation will hopefully change how you think about generating ideas for your content.

What's a piece of research you feel like is missing in your category?

Taplio

Niklas Buschner's Linkedin Analytics

What is Niklas talking about?

's Best Posts (last 30 days)

Want to drive more opportunities from LinkedIn?

Taplio

.css-1jgreyo{background-image:linear-gradient(284deg, rgba(39,43,72,1) 0%, rgba(33,47,183,1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Niklas Buschner's Linkedin Analytics

What is Niklas talking about?

's Best Posts .css-1emoo84{font-size:var(--chakra-fontSizes-md);font-weight:400;color:var(--chakra-colors-gray-600);}(last 30 days)

Want to drive more opportunities from LinkedIn?

Niklas Buschner's Linkedin Analytics

's Best Posts (last 30 days)