How AI and ChatGPT Are Changing B2B Sales Pipeline Development

Every few years, a new technology arrives and the marketing world loses its mind. Right now, that technology is AI — ChatGPT, Claude, Gemini, and a dozen others. And yes, the applications are genuinely incredible. Go and use them. But before you let your marketing team build a strategy around them, ask yourself one question: will any of this actually change your revenue?

Because I have watched this exact story play out before. New technology arrives. Marketing gets excited. Budgets shift. And business failure rates stay exactly where they were. That is not cynicism — it is pattern recognition built over thirty years of watching businesses spend money on things that do not work.

So let me show you the pattern.

1. Around 2010, B2B organisations piled into marketing automation. It was going to change everything. Businesses continued to fail at the same rate. No meaningful increase in revenue.
2. Pay-per-click to landing pages. No increase in revenue.
3. Gating content behind email forms to collect addresses. No increase in revenue.
4. The numbers tell the whole story. 500,000 businesses start in the UK every year. Roughly the same number close. That cycle has continued for over twenty years. 20% fail in year one. 30% by year two. Around 50% by year three. By year ten, 91% are gone. And yet marketing keeps selling the same solutions.
5. Marketing is cited as a contributing cause in roughly half of all business failures. Not the economy. Not the product. Marketing.

Most businesses just copy what they are told, by repetition, by big tech, by MarTech vendors with SaaS subscriptions to sell. Nobody stops to ask why profitability never improves. And that brings us to AI.

AI, Content and the Problem Nobody Is Talking About

The AI conversation is real. ChatGPT, Claude, Gemini, Midjourney, DALL-E, Higgsfield — these tools are extraordinary. I use them constantly. But being extraordinary does not make them a revenue strategy. And here is what the early enthusiasm is missing.

When AI tools first emerged as content creators, they confirmed what we had been saying on salesXchange since 2020: well-structured, long-form, detailed content — with proper headings, video, graphics and a logical architecture — is what gets found and read. We wrote about this years before AI made it mainstream. Have a look at our guide to writing great copy for your website if you want to see what that looks like in practice.

Now here is the problem that nobody in marketing wants to acknowledge.

• If every business uses AI to generate content, nobody stands out. Every brand now has access to the same tools, the same models, the same outputs. The result is what researchers are calling a "sea of sameness" — similar content appearing across every search result, identical structure, identical tone, nothing to differentiate one vendor from another. Google's own guidance now makes clear that content must be original, experience-led and genuinely useful to readers — not just well-formatted. AI-generated content without human expertise and real-world insight is being rated as low quality by Google's review teams.
• Marketing automation dies hard. Businesses that have spent years investing in MarTech SaaS are not going to abandon it. So what happens? They use AI to generate better content and then hide it behind an email form. Same broken model, better-looking output. The logic is absurd. If your content is now genuinely useful — detailed, well-structured, written to help a buyer understand their problem — why would you lock it away to collect a KPI based on email addresses? Or would you rather your prospects could actually read it?
• Your copywriters have questions to answer. If AI can now produce well-structured, long-form content that your team could not or would not produce before, what exactly were they doing? Three-paragraph blog posts that nobody reads, indexed by nobody, found by nobody. That is not a content strategy. Our copywriting articles go into this in considerable depth if you want to understand what content should actually look like.
• AI is replacing search for the 95% who are not yet buying. We know from our research that 95% of your market is not actively buying at any given moment. Those people are learning, researching, forming opinions. When they cannot find what they need on your site — because it is gated, thin, or buried — they now ask ChatGPT or Gemini instead. And Gemini does not recommend your business. It gives them a synthesised answer from whoever produced the clearest, most open, most authoritative content. If that is not you, you are invisible.
• 83% of B2B buyers research digitally before they ever speak to anyone in sales. 6sense's 2025 research puts this at 83% of buyers who fully define their purchase requirements before contacting a vendor. Gartner's data shows B2B buyers spend only 17% of their total buying time talking to potential suppliers. The buying decision is being made before your sales team even knows a prospect exists. The question is whether your content is shaping that decision — or whether a competitor's is.

This leads directly to the questions that will define the next few years of B2B selling.

• If buyers are researching independently and forming shortlists before contacting anyone, what does your digital presence say about you?
• What will differentiate your business from a competitor producing identical AI-generated content?
• If a prospect can get a detailed answer from an AI tool without visiting your site, why would they visit your site at all?
• And if they cannot find you, cannot distinguish you, and cannot trust you — will they ever call?

My answer to all of the above is the same. The most transparent businesses will win. Open access. Real expertise. Content that demonstrates you understand the buyer's problem better than anyone else in the market. That is what gets found. That is what builds trust before the first call. And that is what AI, used properly, can help you do at scale — if the underlying strategy is sound.

Which brings me to the point everyone avoids. AI amplifies the model you give it. If your go-to-market model is broken — random acts of content, gated assets, cold calling into a market where 83% of buyers have already decided before they speak to you — AI will not fix any of that. It will just produce broken output faster. We have written about why cold calling specifically needs to stop, and the same logic applies here. Fix the model first. Then let AI execute it.

The revenue decisions you make now around AI will expose every weakness in your existing approach to new business. That is not a threat — it is an opportunity. But only if you are willing to look at what is actually going wrong before you buy into the next round of marketing hype.