You receive an inquiry from a potential buyer. A sales rep gets in touch, asks a few questions, explains what the company does, and starts qualification.
It may seem that this is where the sales process begins.
Increasingly, it does not.
Before speaking to you, the buyer may already have researched the market, reviewed supplier websites, compared several approaches, read case studies, checked third-party sources, and asked AI to identify companies that could solve the problem.
Sometimes they arrive at the first meeting with a shortlist and a fairly clear idea of what they believe they need.
That idea may or may not be correct.
But it does not make the salesperson’s job any easier.
Formally, the lead has only just entered the CRM. In reality, the competition for that buyer started much earlier.
The latest research supports this shift. In its 2025 global Buyer Experience Report, based on nearly 4,000 B2B buyers, 6sense found that buyers first engaged sellers at around 61% of the way through the buying journey. Yet 94% had already ranked their shortlist before that conversation, and the vendor preferred before seller engagement still went on to win roughly four out of five deals.
So the important change is not that sales has disappeared from the buying process.
It is that sales increasingly enters a process that is already well underway.
THE SALESPERSON IS NO LONGER THE FIRST SOURCE OF INFORMATION
Not long ago, a significant part of the first conversation could be spent explaining the market itself.
What solutions exist? How does one approach differ from another? Which companies are worth considering? What should the client look at when choosing?
Today, buyers can do much more of that work themselves.
And traditional search is now supplemented by AI.
A potential buyer can describe the business problem, define the criteria, and ask:
“Find companies that work with businesses of our size, solve this type of problem, and have relevant industry experience.”
Within minutes, they can receive a list of suppliers, a comparison of approaches, possible limitations, and even a set of questions to ask during vendor meetings.
This is no longer marginal behavior. Gartner reported in 2026 that 45% of B2B buyers had used generative AI during a recent purchase. The same research found that 67% preferred an overall experience without a sales rep.
That changes the salesperson’s role.
In the past, a seller often helped the client build an initial understanding of the market. Now, the seller increasingly enters after that understanding has already begun to take shape.
The job is no longer simply to explain the product.
You need to understand what conclusions the buyer has already reached, where the information came from, which options are genuinely relevant, and where the preliminary analysis may have led them in the wrong direction.
AI does not necessarily reduce the value of a strong salesperson.
It reduces the value of a salesperson who can do little more than repeat the company presentation.
SOMETIMES YOU ARE COMPETING WITH MORE THAN ANOTHER VENDOR
Earlier this year, I had a meeting with a potential client that demonstrated this change very clearly.
Before speaking with me, the client had already used AI to research the consulting market. They had reviewed several possible approaches and formed an initial view of which solutions might fit their company.
So the meeting did not begin with the usual:
“What can you offer us?”
We moved almost immediately into comparing options they had already identified.
Which of them could actually work? What were the limitations? Which approaches were worth exploring further?
On one hand, the client came to the meeting much better prepared.
On the other, preliminary research did not mean the problem had already been solved.
AI can generate a list of possible tools and approaches. But that list does not answer the more difficult question:
What will actually work in this particular company, with its people, processes, constraints, and economics?
In some situations, the buyer’s initial assumptions have to be challenged or rebuilt.
And that creates a new difficulty.
Once someone has received a detailed and convincing answer from an AI system, they may treat it as authoritative. Correcting an already formed opinion can be harder than helping someone build the decision framework from scratch.
In that sense, the salesperson is no longer competing only with other suppliers.
They may also be competing with an AI-generated interpretation of the market for the right to be seen as the more credible source of expertise.
That does not mean buyers no longer value human input. Gartner found that 69% of B2B buyers prefer to validate AI-generated insights with sales reps. The same study found that 70% prefer a completely digital self-service experience. Those findings may appear contradictory, but they point to a different division of labor: buyers want to research independently, then involve a human when they need context, validation, or confidence in a decision.
That is a very different role from simply presenting a product.
THE COMMERCIAL FIGHT STARTS BEFORE THE INQUIRY
If potential customers increasingly begin with search, AI, peer recommendations, review platforms, analysts, or industry content, one question becomes unavoidable:
Will your company appear in that research at all?
A business may have twenty years of experience, a strong product, an excellent team, and successful projects behind it.
But if that expertise is barely visible outside the company, it becomes much harder for a buyer to discover and validate.
The same applies to AI-assisted discovery.
This does not mean there is a simple formula for “optimizing for an LLM.” AI systems differ in how they retrieve, rank, summarize, and cite information. But the direction is clear: the information available about your company across owned and third-party sources increasingly affects whether your brand can be surfaced and understood during AI-assisted research.
Forrester has documented the same shift. Its research found that 95% of B2B buyers planned to use generative AI in at least one part of a future purchase, while more than half said AI had already led them to consider more or different vendors. Forrester argues that B2B visibility increasingly depends on content being relevant, credible, authoritative, and discoverable by both people and AI systems.
This changes what we mean by lead generation.
Call the earlier stages demand generation, brand building, content marketing, category education, or something else internally. The terminology matters less than the commercial reality: the competition begins before the form fill.
Your website matters. So do product and service pages, case studies, articles, interviews, expert commentary, customer evidence, industry media, professional communities, review platforms, and conference appearances.
I used to view this work primarily as expert positioning.
In complex B2B, clients have always evaluated more than the product. They evaluate the company and the people behind it.
Now that digital footprint has an additional function.
It helps buyers, search systems, and AI-powered discovery tools understand what the company actually does, where its expertise lies, and what evidence exists to support its claims.
Simply writing “we are experts” on a corporate website is not much of an argument.
THE BUYER IS OFTEN A GROUP, NOT A PERSON
There is another shift that matters particularly in complex B2B sales.
Working only with the person who first makes contact has never been enough in a complex deal. Today, that reality is becoming harder to ignore.
In my own work, I have long used stakeholder mapping: who is involved in the decision, who evaluates the technical side, who builds the business case, who handles procurement, who will use the solution, who can support the project, and who can stop it.
This same logic increasingly shapes marketing.
Hence the growing focus on account-based marketing, buying groups, and intent signals.
6sense’s 2025 global research found that typical B2B purchases involve more than ten people, while buyers still initiate roughly four out of five first engagements with vendors themselves.
So marketing should not be looking only for another individual lead.
The more useful question is whether a need is beginning to form inside a specific account.
Perhaps several people from the same company are researching a topic. Perhaps different stakeholders are engaging with related material. Perhaps the organization is examining a particular category of solution or revisiting an issue it explored in the past.
The goal is to understand not merely who filled in a form, but what may be happening inside the account.
That changes the handoff between marketing and sales.
A name, an email address, and a phone number are useful.
But a much more valuable piece of information is:
“Why does it make sense to speak with this company now?”
YOU CAN AUTOMATE LEAD GENERATION. YOU CAN AUTOMATE CHAOS TOO
There are already many parts of the commercial process that AI and automation can accelerate.
Account research. Data enrichment. Website analysis. Qualification. Prioritization. Message preparation. Analytics.
And I keep seeing the same mistake.
A company tries to automate a process that does not work properly even when people do it manually.
The target segment is poorly defined. Qualification criteria are unclear. The CRM is incomplete. Data is fragmented. Nobody properly analyzes why deals are lost.
Then AI is introduced with the expectation that it will fix the system.
It will not.
It will allow the existing system to operate faster.
If the process is good, that creates scale.
If the process is bad, it scales the disorder.
So my sequence remains the same:
Process and data first. Automation second.
AI should amplify a functioning commercial system, not compensate for the absence of one.
THE MORE AI WE USE, THE MORE VALUABLE REAL EXPERTISE BECOMES
At first glance, there is an obvious contradiction here.
If AI is getting better at research, account analysis, content creation, and helping buyers compare suppliers, does that make the salesperson less important?
In some types of transactional sales, certain functions will almost certainly become less valuable.
But complex B2B is different.
The more automatically generated emails, presentations, messages, and content a buyer receives, the easier it becomes to notice someone who actually understands the industry and the situation.
AI can collect information.
Someone still has to determine whether that information applies.
Someone has to recognize risks that do not fit a standard scenario.
Someone has to understand why a solution that looks excellent on paper may fail inside a specific organization.
And someone has to connect technical capability with economics, implementation risk, internal stakeholders, and the realities of the business.
The Gartner findings are useful here because they show both sides of the change. Buyers increasingly want digital self-service, yet 69% still prefer to validate AI-generated insights with a sales representative.
That is not a contradiction.
It suggests that the value of the seller is moving away from information delivery and toward interpretation, validation, and decision support.
The buyer does not need a salesperson for information that can be found in thirty seconds.
The buyer needs a salesperson when information has to become a decision.
AI is not necessarily removing the strong seller from a complex deal.
It is raising the standard for what a strong seller needs to be.
WHERE I WOULD START
If I were rebuilding a B2B lead generation system today, the first thing I would do is check how the company looks from the outside.
This does not require a large budget.
Take 10 to 15 real questions a potential customer might ask when looking for your product, service, or type of supplier. Ask those questions across several AI systems and traditional search.
Do not include your company name.
Phrase the queries the way a buyer actually would.
Does your company appear? Which competitors do? How are those competitors described? What criteria are used to compare them? Which sources are cited or referenced? Is your specialization understood correctly? Are your strengths visible? Are there obvious gaps or inaccurate descriptions?
This exercise is particularly relevant in software, where the shift toward AI-assisted discovery is already pronounced. G2’s 2026 survey of 1,076 B2B software buyers and decision-makers found that 51% started software research with an AI chatbot more often than with Google, while 71% used AI chatbots somewhere in the research process. These figures should not be generalized to every B2B category, but they are a useful indication of where buyer behavior is heading.
The next step is to inspect the digital evidence behind your positioning.
Can a buyer quickly understand what problems you solve? Are the product or service pages substantive? Are there case studies? Specific results? Named experts? External publications? Customer evidence? Independent mentions?
Then I would return to the target account.
Not:
“We need manufacturing companies.”
Which manufacturing companies?
In what situation?
With what processes?
What has to be happening inside the organization for your solution to become relevant?
Then map the people involved.
Who initiates the project? Who evaluates the technical side? Who builds the financial case? Who manages procurement? Who will use the solution? Who can support it internally? Who can block it?
Only after that does it make sense to connect marketing, CRM, sales, data, and automation into one system.
The goal is not simply to count inquiries.
You need to understand which accounts are showing genuine interest, what happens after the first contact, where conversion is being lost, and which channels and activities eventually contribute to revenue.
PART OF YOUR FUTURE PIPELINE MAY ALREADY BE SITTING IN THE CRM
As customer acquisition becomes more expensive, it makes little sense to focus exclusively on finding new contacts.
In many companies, a significant commercial resource is already sitting inside the CRM.
Old leads. Lost opportunities. Former clients. Existing customers with cross-sell or upsell potential.
Yet this database often functions primarily as an archive.
There is no systematic work with it because most attention remains focused on generating new inbound demand.
That is also a lead generation problem.
The question is not only where the next new contact will come from.
It is also where you already have accounts that know you, have interacted with you, or considered buying from you in the past.
Ignoring that resource while continuing to pay for new acquisition is difficult to justify.
CONTENT SHOULD HELP THE BUYER MAKE A DECISION
Content is another area worth reconsidering.
I would not set the objective as simply publishing more.
Volume alone solves nothing.
Good B2B content should answer the questions a buyer is actually asking while making a decision.
What options exist? How are they different? What are the limitations? Where does a particular approach work? What risks appear during implementation? When should a solution not be used at all?
This becomes even more important when AI sits between the company and the buyer.
Forrester notes that AI-powered and zero-click search can summarize or interpret content without necessarily sending the buyer to the original website. In that environment, visibility increasingly depends on whether information is useful, trustworthy, structured clearly, and credible enough to be surfaced.
So good content performs several functions at once.
It helps a potential buyer understand the problem.
It shows how the company thinks.
It provides sales with something useful to continue the conversation.
It creates evidence of expertise.
And it improves the chances that the company can be discovered and correctly understood during independent research.
Content therefore becomes part of the commercial process before the buyer has decided to speak with sales.
THE MAIN QUESTION NOW SOUNDS DIFFERENT
For years, conversations about lead generation often started with some version of the same question:
“Where do we get more leads?”
That question has not disappeared.
But in 2026, it is no longer enough.
By the time a lead enters your pipeline, a meaningful part of the buying process may already have taken place without you.
The buyer may have researched the market.
Built a shortlist.
Defined requirements.
Compared different approaches.
Asked AI for recommendations.
Read third-party content.
Discussed the options internally.
And formed an opinion about which suppliers appear most credible.
Only then do they contact you.
6sense’s latest global data captures the point clearly: buyers are speaking with sellers somewhat earlier than they did a year ago, but 94% still rank their shortlist before seller engagement, and the pre-contact favorite still wins around four out of five deals.
So I would formulate the challenge differently.
How do we get into the buyer’s field of view before they decide to contact us?
And there is a second question that matters just as much:
What happens when the person who finally reaches sales already believes they understand the market?
That space between those two points is where more and more of the real competition in B2B is taking place.
