Your B2B marketing is failing due to wrong KPIs and a lack of systems thinking. Focus on qualified sales pipelines, Generative Engine Optimization, and clear ICPs for revenue.

By Edin Cerimagic, CEO iGrow | Revenue Marketing, AI Visibility, and Qualified Pipeline

I am giving this presentation because I see the exact same problems over and over again. In the last two weeks, I had conversations with ten different companies. Eight out of ten had the exact same issue: performance slumped, no qualified inquiries, budget is increased, and still nothing happens. No revenue. No pipeline. No clarity.

This is not an isolated case. This is a systemic problem in German-speaking B2B marketing, and in this article, I want to point out where the mistakes lie, why they keep happening, and how to do it better.

Revenue is not a coincidence, but many treat it that way

For years, I have been dealing with a single question: How do you actually generate revenue? Not just somehow, not with a gut feeling, not with the hope that someone will eventually inquire, but with a system that works reproducibly.

And what I repeatedly realize is that the biggest hurdle is not the budget. The biggest hurdle is that the wrong customer comes along, or no customer at all. That you generate leads that show during the discovery call that the person does not even belong to the target group. That marketing and sales talk past each other. That someone invests in Google Ads without knowing what inquiries result from it.

"I looked at what we have on Google Analytics. Where is the traffic coming from? Google Search Console branded versus non-branded. Am I actually getting traffic for what people know me for or not?"

That sounds trivial. But most companies cannot answer this question. And as long as you cannot answer this, you are driving in the fog.

The problem with the sales pipeline and why most have too little of it

In many conversations, I first ask: Do you have a pipeline? What do you currently have in the pipeline? How much proposal volume do you have out there?

The answers are sobering.

The definition of a sales pipeline is clearer than "a channel that is supposed to bring revenue": It describes the internal sales process with clear, measurable phases and concrete steps that make the progress of a deal visible, thus enabling predictable revenue and more efficiency. The sales funnel is often confused with this, but it represents the customer's perspective from interest to purchase decision, while the pipeline manages internal sales activities. How many proposals do I have out there? How high is the volume? If my revenue target is 1 million euros, then I need a pipeline coverage of at least factor 3.

"I need proposals worth at least 3 million euros out there. Why? Sales takes too long. Decision-makers need time, complex questions must be asked, stakeholders must be onboarded. At some point, it just happens that things get watered down and deals fall through. That's why I always have to make sure that I have three times as many proposals out there."

This is not an optional thought. This is the foundation of any serious revenue planning. And yet, most conversations with me start with someone saying they need more budget without knowing how many of their current inquiries actually get converted into real opportunities.

Where I come from and why I take this so seriously

I was at Geizhals. I was at the Ritz-Carlton. I was at the Kronenzeitung. I was at Infonika. All of these companies had one thing in common: the relationship between marketing and sales did not work. They were fighting. Sales said we are not getting inquiries. Marketing said we generated 100 inquiries.

Then, in the revenue meetings, it was discovered that the defined target customers and the incoming inquiries did not match. Gmail addresses, fake bots, competitors who just filled out forms to generate costs.

"If you have a form: block freemails, only allow business accounts. When you run campaigns, you will actually get targeted inquiries and not fake bots."

At the time, in high-ranking roles, I had C-level managers who did not want to implement my suggestions. That was the moment I went independent. Not out of defiance, but because I knew there was a better way.

The four pain points that I see over and over again

When I talk to B2B companies that have a revenue problem, the causes can almost always be traced back to four key problems.

First: AI confusion. Organic rankings are dropping. Direct traffic is slumping. Google rankings are losing relevance while AI systems dominate. At the same time, user behavior is shifting: more and more people start their search via AI assistants and social networks instead of just classic search engines. Many business owners feel this but do not know what to do about it. Classic search engine optimization still aims for visibility in search results, whereas Generative Engine Optimization (GEO) optimizes for mentions in AI answers, and this very importance is still underestimated in many teams.

"Visibility is buckling in organic rankings and also in traffic. Direct traffic is collapsing, Google rankings are dropping, AI is dominating, this is increasing significantly."

And at the same time, this very AI is also the most powerful tool we have if used correctly.

Second: Rising advertising costs without a corresponding return. We also have iGrow in Croatia. In Croatia, a CPC for our industry costs us 20 cents. In Vienna, we pay 10 to 60 euros, depending on the keyword. An inquiry costs us between 200 and 800 euros here.

"With a daily budget of 100 euros, I have to run it for eight days just to get a single inquiry in Vienna, and then hope that the customer, if I have an ACV of sum X, also yields a return on investment of factor 2.5. With these high costs, that is virtually impossible."

Third: More of the wrong thing. Eight out of ten companies I talked to are doing more of what doesn't work. Writing more content, running more ads, buying more tools.

"More of the wrong thing actually amplifies the problem and doesn't solve it."

The boss starts posting on LinkedIn, but without a strategy behind it. New channels are opened, Bing ads are imported because they can easily be copied from Google. A tool orgy is staged: SEO tool, GEO tool, ChatGPT, Gemini, Perplexity. €4,000 in tool costs per capita in a medium-sized B2B company, without anyone knowing what that actually brings in.

Fourth: "We have always done it this way." This sentence is the most dangerous. Anyone who hears it must question themselves three times why it is not working, because somewhere in between lies the truth.

The three crucial questions everyone must ask themselves

In my presentations, I always show three questions that make a difference. They sound simple, but very few people answer them honestly.

The first question: What is being optimized? Many companies optimize for impressions and clicks. Those are vanity metrics. Data-driven companies optimize for leads, booked demos, decision moments. That feeds the pipeline.

The second question: What is considered success internally? If a marketer is rewarded internally for traffic and impressions, then they optimize for traffic and impressions, not for revenue. For valuation, concrete metrics like Customer Lifetime Value (CLV) and Return on Investment (ROI) count, because they make real advantages visible and make the collaboration with a B2B Growth Partner as an external revenue engine measurable.

"If I have marketers internally who try to optimize for impressions and traffic, that does not feed the pipeline. Then I have a marketing team that is busy and generates no revenue."

The third question: Where is the budget being used and why there? Anyone who cannot answer this question is making marketing decisions based on gut feeling instead of reliable information.

What AI can do and what it cannot do

AI is not a savior. It is an amplifier. That is the most important sentence I can say on this topic.

"First of all, AI cannot replace humans, but it can replace those who do not know how to use AI. An AI cannot somehow replace a bad strategy either. You need a good database and a good system."

Once I have clearly defined my value proposition, when I know what I stand for and what I promise, when I have invested time and resources in my authority and my brand as an entity, then AI brings the right people to my platform, especially when I format my content for AI Search Optimization. This requires a clear content strategy so that content is prepared in a machine-understandable, extractable, and citable way for GEO. This works incredibly fast.

We had a client from the US with whom we made a bet: 100 percent AI visibility across seven channels by Easter. If it doesn't work, the collaboration ends. It worked, using a structured AI Search & GEO Strategy that relies on LLM-optimized content and clear entities.

"When an inquiry comes in, speed-to-lead time: how long does it take during business hours until the prospect is contacted? What happens as a follow-up? Do I have SDRs? What is an SQL, what is an SAL? If I haven't defined these things beforehand and my system isn't built to generate revenue, then none of this is of any use."

The difference between linear thinking and systems thinking

Anyone who thinks like this: budget equals reach equals revenue, hasn't understood how it works in 2026 yet. That is linear thinking. It ignores everything that happens between reach and the purchase decision.

Systems thinking means: I recognize the demand. I analyze the market. I look at where the competition is positioned. I sharpen my offer until I am the logical choice, not an alternative.

"I started with €200 deals a month and switched to €10,000. Selling €10,000 is easier than €200 if you are positioned correctly."

I know the book "Key Person of Influence" by Daniel Priestley. He says exactly that:

"Sharpen your offer so strongly that you are the logical explanation for your customer, not the alternative. As soon as you are an alternative, you are replaceable."

This applies to employees. This applies to agencies. This applies to products. Whoever is replaceable always loses on price.

ICP, why the ideal target customer changes everything

I take a flipchart. I write down: industry, account size, number of employees, who the decision-maker is, what their buying persona is, what security concerns they have, what they need as a GDPR prerequisite. Everything I know about this person. From this, a clear list of potential customers emerges; in the next step, this becomes concrete as account-based marketing for individual high-value target customers with personalized campaigns, embedded in a well-thought-out inbound marketing strategy.

Based on this specialization, I build my niche and train the AI with my content, with digital PR, and everything available, within the framework of clearly defined inbound marketing strategies. So strongly that I am the logical answer for this person, no longer just one choice among many.

"As a marketing agency, I could also compare myself to 1,000 agencies in Vienna. Then I would always lose on price, and I don't want that."

Nobody wants to be compared. Whoever is compared is replaceable. Anyone who has defined their ICP so clearly that they are specifically positioned for exactly these people is not compared. They are recommended.

AI visibility is not luck, it is a system

I hear again and again: "We once had an inquiry via ChatGPT. That was more of a lucky coincidence." In truth, targeted Prompt Optimization for AI Visibility allows very precise control over which search queries trigger AI systems to recommend a brand.

No. That was not luck. That was an unused opportunity to build this channel systematically.

"Nine out of ten companies come to us via ChatGPT now. That is massive. And you have to question that. If you don't track it with UTM parameters, then you can ask in the discovery call: How did you actually find us?"

If I do not exist in the AI, I lose the customer's trust before they have even seen my website. I know an optician in Vienna who has been offering top-quality, handmade glasses for 20 years, whom almost everyone in the area knows, but the AI doesn't know him. Then someone else wins who knew their way around this better. Most people don't understand that.

AI acts as a trusted authority. Anyone recommended by ChatGPT has a trust advantage before a single word has even been directly communicated.

The three decisions that drive growth

I have observed over many years what actually works. In the end, it is always three decisions that make the difference.

First: Focus on quality over quantity. 1,000 qualified visitors a month bring in more money than 10,000 random clicks. Precise targeting and close coordination with the sales team increase close rates and relieve the team up to the deal closure. This is not a difference of opinion. This is pipeline math.

"1,000 qualified visitors a month bring me more money than 10,000 random clicks. That is a typical funnel thought."

If you think in campaigns, like what Easter bunny do I post on LinkedIn to get a few likes—90 percent of which come from my own employees—you are not thinking in funnels. Thinking in funnels means: What goes in at the top, what comes out at the bottom?

Second: Cash clarity over creativity. There are millions of websites that offer everything to everyone at the same time. This leads to all sorts of things coming in, and you have to prove yourself, fight, and explain in discovery calls.

Solve a clear problem, one clear landing page, one clear call to action. If I support a software company that wants demos, then every relevant page says: Book a demo. Not three different CTAs, not a free initial consultation here and a contact form there. One goal. One action.

Third: Measurability over gut feeling. We connect all data from all systems with tools like Claude and Windsor AI. Google Analytics, Google Search Console, Google Ads, LinkedIn, everything together. Then I analyze where I am losing potential customers, why traffic might be sluggish, and what actually works.

"I look into my systems every day, what can I improve? 99 percent of inquiries come either via ChatGPT or organic paths."

Money prompts, why most measure the wrong traffic

I define so-called money prompts for myself. These are transactional search queries in AI that signal a high willingness to buy and behave similarly to high-intent B2B leads via Google.

If someone asks: "Who is the best shoe manufacturer in Vienna?" that is a transactional prompt. Extremely high intent to buy. That is a money prompt I want to track.

If someone asks: "Which shoe for flat feet is good for walking along the Danube?" that is informational. It doesn't bring me a customer. I don't track that prompt, just as in lead generation in online marketing, I have to distinguish between pure informational clicks and true purchasing intent.

"This is also a huge problem. People don't understand what actually brings them money. It's just like with keywords. If I take a transactional search term now, I get a customer. If I take an informational one, I get traffic, but no customer."

You have to be able to measure that. It has to be in the database. And those who don't do it are optimizing for the wrong thing.

The system I trust

There are many systems. I trust this one because it addresses the typical pipeline instead of lead problem in B2B:

I recognize demand, look at the market, and analyze the competitive positioning. Then I sharpen my offer until I am the logical alternative. Then I build trust before the purchase happens. Then I force conversion by aligning marketing, sales, and customer success on shared revenue goals in a revenue marketing system.

All combined with SEO and GEO, i.e., Generative Engine Optimization: LLM-optimized content that can be classified and recommended by AI, supported by a specialized GEO agency for AI search results. Google Ads campaigns only where demand already exists and the readiness to buy is high. A CRM system connects these marketing touchpoints directly with subsequent sales closures. The CRM makes the sales pipeline measurable and manageable. And content that builds trust, not to gather impressions, but to prepare real decisions.

"If I am not present in the AI with this entire system, I lose trust and the customer before they have even seen my website."

What drives me personally every day

I don't look at how many likes my last LinkedIn post got. I count the inquiries that resulted from it. I calculate how much revenue will come out of it. That is my KPI.

I haven't invested a single cent in advertising in Austria. I haven't bought any digital PR packages, even though someone calls me every day explaining that this is how I get visibility. 99 percent of my inquiries come through organic paths and via ChatGPT.

"If you train the algorithm on LinkedIn, write about a certain topic, and network with the right people, they will watch you. 14 touchpoints. Something has to happen about 14 times, across three or four different channels."

Therefore: YouTube, podcasting, LinkedIn, GEO. Everything that is relevant is used. But with clarity on what it is supposed to bring, not just to be busy.

My KPI is the AI Visibility Score. My KPI is a Cost per Lead of €0. Because if I don't run ads, that is the goal. For others, the goal is: what is my current CPL, and how do I reduce it by 40 percent in the next three to four months? That is doable. With just a few mechanisms, if the system is right.

What my sister teaches me

My sister Sabine is a psychologist and our Head of Revenue Psychology. She guides our clients, and I actively ask her to criticize me.

"I also say every time: criticize me if I talk nonsense. Because it is damn important to also look at where you make mistakes yourself."

Receptivity to criticism is not a sign of weakness. It is the prerequisite for growth. Anyone who doesn't question what they are doing wrong will keep making the same mistakes and wonder why revenue is stagnating.

I transcribe my discovery calls. I analyze them. I ask the AI what I could have done better. What did I do poorly? Where was I not good? Because that's how you learn.

Conclusion: Pattern recognition and clear priorities

In the end, it's always about the same thing. Pattern recognition. Setting priorities. Making clear decisions.

"It's always about pattern recognition. I have to recognize: can I even make clear decisions? I have to set priorities. My top priority for this year is to invest twice as much in AI, time-wise, not financially."

Focusing on the essentials means: a clear decision, a clean structure, and measurable results derived from targeted marketing ideas and concepts. Anyone who systematically implements this doesn't need a large budget. Those who don't implement this will just burn more and more budget and generate less and less revenue.

*Edin Cerimagic is the CEO and founder of iGrow, a revenue marketing agency for SaaS, tech, and B2B companies in the DACH region. This article is based on a presentation on qualified pipeline, AI visibility, and building revenue systems. Learn more on igrow.at.