B2B Engagement Metrics No Longer Predict Revenue, and LinkedIn's Own Research Explains Why

There is something quietly absurd about a company publishing research that undermines its own core value proposition. LinkedIn just did exactly that.

New research commissioned through LinkedIn shows that traditional B2B marketing metrics - reach, engagement, follower counts, click-through rates - no longer "ladder up to being bought." The study attributes this to AI-driven shifts in B2B buyer behavior: buyers are doing more independent research, using AI tools to synthesize vendor comparisons, and making decisions with less direct interaction with sales or content than they did two years ago. The implication is that a brand can be winning on every LinkedIn dashboard metric and still be losing in the room where the actual purchase decision happens.

LinkedIn is, of course, the platform those metrics live on. So this is a little like a ruler manufacturer publishing a study showing that length doesn't predict quality. It deserves to be taken seriously precisely because it cost them something to say it.

Why Engagement Became a Proxy for the Wrong Thing

The story of B2B metrics is a story about instrument drift. In the pre-AI era, engagement was a reasonable proxy for something real. If a potential buyer engaged with your content repeatedly, they were likely building familiarity and reducing perceived risk - two preconditions for purchase. The chain from awareness to consideration to decision was mostly linear, mostly human, and mostly visible to the platforms hosting the content.

That chain is breaking. Intent over impressions has been a theme in brand strategy research for over a year now, but what LinkedIn's data adds is specificity: buyers are now using AI to compress the discovery and evaluation phases. They are arriving at purchase decisions informed by synthesized research that never touched your content pipeline.

What this creates is a gap between activity and outcome that gets worse the more sophisticated your buyers become. Enterprise buyers, technical buyers, financial buyers - these are exactly the segments most likely to be using AI-assisted research tools. They are also exactly the segments that B2B brands have spent the most money trying to reach on LinkedIn.

The AI Buyer Doesn't Need Your Awareness Campaign

The mechanism matters here. Traditional B2B marketing assumes a buyer who starts with low familiarity and needs repeated exposure before they'll engage with a vendor. AI-assisted research inverts this. A buyer using a deep research tool can go from zero familiarity to a shortlist of three vendors with a single query - without ever clicking a sponsored post or downloading a gated PDF.

This is not a future problem. The behavioral shift is already in the data. And what it means practically is that the brands winning B2B deals right now are not necessarily the ones with the highest engagement scores. They are the ones who show up correctly in AI-generated comparisons - which depends on entirely different inputs: structured product data, third-party validation, clear positioning, competitive differentiation that exists in the real world rather than in ad copy.

This is the kind of pattern STI's research tracks systematically - the gap between what brands measure and what buyers actually use to make decisions.

FMCG Insurgent Brands Are Running the Same Experiment at Scale

The same week that LinkedIn published its B2B metrics research, Bain released its annual analysis of insurgent brands in U.S. fast-moving consumer goods. The numbers are striking: small insurgent brands - those with less than 2% aggregate market share - captured 36% of total category growth in 2025.

This is not a new trend. Insurgent brands have been punching above their weight for several years. But the scale is accelerating, and the Bain data cuts against a specific assumption that large FMCG players have been reluctant to abandon: that market share protects you.

It doesn't, for the same reason that B2B engagement metrics don't predict purchase. Market share is a backward-looking measure. It tells you where money went, not why. An insurgent brand with 1.5% share that is growing fast and has a cult following among a specific segment is not well-described by that 1.5% number. The incumbents who look at their 30% share and feel secure are looking at the equivalent of their LinkedIn engagement dashboard - accurate, but pointed at the wrong thing.

Why Insurgents Win When Incumbents Track the Wrong Variables

There's a structural reason why insurgent brands keep outperforming on growth while incumbents retain share. Large brands optimize for defensible metrics - distribution reach, aided brand awareness, unaided recall. These are things that scale efficiently and that large organizations can report upward credibly.

Insurgent brands, often by necessity, optimize for the thing that actually converts: conviction among a specific audience. They can't afford reach, so they go deep instead of wide. They end up with customers who not only buy but advocate. When those customers encounter an AI research tool asking "what's the best [category] product for [use case]?", they are the ones who get mentioned by name. The incumbent with 30% share gets mentioned as the default. The insurgent gets mentioned as the recommendation.

The gap between default and recommendation is where growth lives right now.

Trust Overseas Has the Same Measurement Problem

Harvard Business Review published a piece this week on how American companies can retain trust in overseas markets amid geopolitical volatility. The core argument is that US brands have historically treated international trust as a function of brand awareness - get known enough and the trust follows. That model is breaking down.

The mechanism is similar to the others: buyers in international markets are increasingly making decisions based on proxies that are harder for brands to directly control. Geopolitical associations, supply chain provenance, data governance practices - these are now trust variables in ways they weren't five years ago. A brand can have strong awareness in a European or Asian market and still be losing deals because buyers have updated their trust calculus based on factors the brand has no campaign to address.

The HBR analysis suggests that the path forward involves transparency over messaging - communicating clearly about governance structures, local partnerships, and operational independence from US political dynamics. This is not a marketing play. It's a structural one, and it requires the same honest self-assessment that the LinkedIn metrics research implies: the thing your audience actually cares about and the thing you're currently measuring are probably not the same thing.

Brand trust is increasingly operational, built through what you demonstrably do rather than what you claim. A brand can't address trust erosion caused by supply chain concerns with an awareness campaign. It has to address the supply chain.

Norwegian Cruise Line and the Experience Signal Problem

The fourth data point this week comes from a neuromarketing analysis of Norwegian Cruise Line's dress code backlash. NCL quietly updated its dining policies to ban shorts and flip-flops in six premium restaurant venues. The stated goal was to elevate the experience of those venues. The result was immediate and fierce customer backlash, covered by the New York Post and spreading across social media.

Roger Dooley's breakdown of the incident - drawing on research into servicescapes and brand experience design - identifies the core error: NCL was measuring venue quality by its own internal standards (ambiance, food quality, service levels) but customers were measuring it by a different variable entirely: the freedom to be comfortable while being treated well. The dress code change didn't affect food quality. It signaled something about the nature of the transaction, and customers read that signal differently than NCL intended.

The Signal Interpretation Gap

The NCL story is a clean illustration of what happens when a brand optimizes for a signal it controls rather than the experience its customers are actually having. This is a small-scale version of the same problem showing up in B2B metrics and FMCG market share data.

Brands are making decisions based on what they can measure - engagement rates, share percentages, internal experience ratings. Customers and buyers are making decisions based on variables that are harder for brands to directly observe: whether the product shows up well in an AI comparison, whether the brand is growing in the segments they care about, whether the dress code signals "we trust you" or "we don't."

The gap between what you measure and what your audience uses to decide is not static. It widens every time a new tool or behavior shift makes your current instruments less accurate.

If you're evaluating your own brand's measurement stack against these patterns, our analysis tools can help surface where your current metrics may be pointing at the wrong variables.

What "Buyability" Actually Requires

LinkedIn's word "buyability" is useful even if it's inelegant. It names the thing that all four data points this week are circling: a state of being in which a buyer, at the moment of decision, chooses you. Reach doesn't create it. Market share doesn't protect it. Awareness doesn't substitute for it. A dress code doesn't signal it.

What does create it varies by context, but the structural pattern across all four stories is the same. Buyability is downstream of how accurately a brand understands what its buyers are actually weighing - not what they report they care about in surveys, not what the brand's dashboards show, but the real decision calculus that buyers use when no one from the brand is in the room.

The irony of the LinkedIn research is that it makes this precise point using LinkedIn's own platform as the example. The buyers making enterprise decisions right now are not the ones most likely to engage with a LinkedIn ad. They're doing their research elsewhere, with different tools, using different signals. That gap between where you're measuring and where decisions are actually made is not a campaign problem. It's a strategy problem.

The brands that close this gap will not do it through better creative or higher ad spend. They'll do it by understanding what information their most valuable buyers actually act on, and ensuring that information is accurate, accessible, and favorable - wherever those buyers happen to be looking.

That's a data problem as much as a brand problem. And it requires a different kind of instrument than a dashboard.

For brand leaders who want to understand how decision intelligence applies to their specific category and buyer profile, the research and tools at smarttechinvest.com are built for exactly this analysis.