Milena Traikovich is a prominent figure in the demand generation space, specializing in helping B2B organizations navigate the complexities of lead nurturing and performance optimization. With a deep background in technical analytics and a sharp eye for market positioning, she has spent years refining strategies that prioritize lead quality over raw volume. Her expertise is particularly relevant in an era where artificial intelligence has fundamentally changed how buyers discover and evaluate software. By focusing on data-driven insights and brand authenticity, Milena helps companies move past generic marketing tactics to build genuine trust with their target audiences. Today, she shares her perspectives on why the traditional B2B playbook of “being everything to everyone” is no longer just ineffective—it is a significant risk to a brand’s digital presence.
Many B2B brands use generic, “one-size-fits-all” language to appeal to every possible buyer. From a demand generation perspective, how does this strategy backfire when AI becomes the primary filter for potential customers?
When a brand uses broad language like “powerful, flexible, and scalable” to describe its product, it essentially abdicates its responsibility to define its own identity. In our current landscape, buyers are increasingly leaning on AI models to do the heavy lifting of vendor research, and if those models don’t find specific, definitive boundaries in your content, they start making inferences. This is where the strategy backfires; without clear exclusionary markers, an AI might hallucinate a purpose for your tool that doesn’t exist or, conversely, fail to recommend you for a niche where you actually excel. We have seen that the safest-looking copy—the kind that tries not to leave any money on the table—actually leaves the company with the least amount of control over its narrative. To maintain authority, a brand must provide the AI with the “truth” of what the product is and, more importantly, what it is not, or risk having the model fill in those blanks with inaccurate information.
You mentioned that buyers are using AI specifically to find vendors they don’t already know. What did the analysis of 2 million B2B conversations reveal about how these committees are actually using these tools?
The analysis performed by the Watchtower audience-intelligence platform across those 2 million conversations highlights a significant shift in buyer behavior among both technical and non-technical committee members. Instead of just looking for the market leaders they already recognize, buyers are using AI as a discovery engine to find specialist, niche vendors that might have otherwise flown under the radar. Interestingly, one of the most common use cases for AI in this context is to specifically hunt for vendor weaknesses because every company claims to be wonderful on their own website. This means the buying committee is actively seeking the “catch”—the trade-offs and the limitations that sales reps often try to hide. For smaller or less familiar vendors, this provides a massive opportunity to stand out, provided they are honest about their specialization and where they fit in the broader ecosystem.
When a company refuses to define its weaknesses or the specific scenarios where it isn’t a fit, how does that influence the way an AI model describes that company to a prospect?
If a vendor’s documentation and public-facing content are silent on limitations, the AI doesn’t just stop looking; it looks elsewhere or makes an educated guess. In various audits, we have observed models claiming a client is a poor fit for certain scenarios simply because there was no data to support the contrary, which is essentially a hallucination born of a vacuum. For instance, if you are an enterprise-level accounting platform and you never explicitly state that you don’t support solo freelancers, the AI might infer that you are too complex for them—or worse, suggest you to a freelancer who will then bounce off your site in frustration. By not providing citations or clear boundaries, you are essentially letting the model—and by extension, your competitors’ data—dictate your market position. The solution is to feed the model better material, such as stating plainly that an enterprise platform is designed for distributed teams and might be overkill for a startup.
Agreeing on who a product is not for sounds like a nightmare for a sales team. How do you facilitate that “uncomfortable conversation” between marketing, sales, and product departments?
These conversations are incredibly difficult because they force a company to choose between the illusion of a broad market and the reality of a precise one. Sales teams are often trained to keep every prospect in the pipeline if there is a budget attached, but we have to show them that a bad fit eventually becomes a “churn” statistic or a wasted resource in the qualification phase. We start by looking at the datif we attract 40,000+ marketing professionals but only a fraction are actually qualified, the cost of that noise is astronomical. We have to protect the “brand truth,” which means being brave enough to say that a competitor might actually be the better fit for a specific use case. It isn’t about creating a “confession page” on the website, but rather integrating these boundaries into pricing pages, FAQs, and buyer guides where they can naturally assist in the disqualification process.
You’ve advocated for “better disqualification” as a way to improve the sales funnel. Can you explain the logic of how turning away the wrong traffic actually helps a demand gen team’s bottom line?
From a demand generation standpoint, more visibility is only a win if it leads to better opportunities; otherwise, it is just an expensive vanity metric. If an AI recommends an enterprise tool to a small business owner because the brand was too vague about its target, that click is effectively useless and creates a burden for the sales team who has to qualify it. By being explicit about exclusions—like a lack of a self-serve plan or specific industry support—you stop the wrong people from entering the funnel before they even click. This improves the overall health of the pipeline because sales acceptance rates and opportunity creation go up, even if total traffic takes a slight dip. We must judge AI referral traffic with the same scrutiny we use for paid search: if the referrals increase but the quality of deals falls, your brand positioning is failing to communicate the necessary boundaries.
Traditional comparison pages are often seen as biased or unhelpful. How are you changing the way these guides are constructed to satisfy both human buyers and AI models?
The era of the “polite but superior” comparison page is over because neither humans nor LLMs find them credible anymore. We are now building transparent buyer guides that are rooted in real user reviews, highlighting exactly where our clients are strong and, crucially, where they are weak compared to two or three competitors. This approach works exceptionally well because it makes the comparison conditional: we tell the buyer to choose us if they prioritize one set of features, but to choose a competitor if a different set of needs is more pressing. People are hungry for honesty, and AI models prioritize content that provides clear, citable distinctions between entities. When you provide an honest assessment, you give your sales team a factual basis to handle objections rather than forcing them to pretend that every weakness is just a misunderstanding.
What is the long-term risk of remaining silent about your product’s limitations in a digital environment where your competitors are likely speaking for you?
Silence is the quickest way to lose control of your brand’s reputation in an AI-driven search environment. If you don’t define your boundaries, the model will pull information from other sources to fill the gap, and those sources often include competitor comparison pages or disgruntled user reviews. Essentially, you are giving your competitors the opportunity to define your weaknesses for you, and the AI will cite them as the authority on your product’s shortcomings. You cannot dictate every word an AI says about your company, but you can ensure that your own perspective and your specific trade-offs exist as a reference point. If no one in your organization can point to a single sentence that explains who shouldn’t buy your product, you can be certain that a competitor or an AI will eventually write that sentence for you.
What is your forecast for the evolution of brand transparency and its impact on B2B SEO?
I believe we are heading toward a period where “transparency” will become a technical SEO requirement rather than just a moral or branding choice. As AI search engines become the primary gatekeepers of information, the brands that provide the most granular, honest, and citable data about their specific niches will dominate the recommendation results. We will see a shift away from keyword stuffing and toward “entity clarity,” where the goal is to be the most accurately defined solution for a specific problem. Companies that continue to hide their flaws will find themselves excluded from AI-generated “shortlists” because the models will perceive them as high-risk or ambiguous. Ultimately, honesty will be the most efficient way to scale organic growth because it minimizes the friction between what a buyer expects and what a product can actually deliver.
