We invited Arjun Moorthy, Co-founder & CEO of Rocksalt AI, to share some of the interesting data his team has been digging into. The following is a guest post.
The best GTM asset a B2B startup has is hiding in plain sight: the founder's genuine point of view. Not the product pitch. Not the polished company page. The founder — their hard-won experience, their specific frustration with the problem, and their authentic conviction about how to fix it. That passion is something AI cannot replicate. And it's the fastest way to build the kind of trust with buyers that actually converts.
Most founders never tap it. Here's why they should — and what the data says about how to do it right.
The B2B buyer journey now has two stages: before AI, when buyers aren't actively in-market, and in AI, when they are actively researching options. Research shows that by the time a buyer opens ChatGPT or Perplexity to shortlist vendors, they already have 4 out of 5 names in mind.
The real question is: how did they form that list?
They formed it on LinkedIn. That's where decision-makers spend their professional attention when they're not yet buying: browsing, learning, and slowly deciding who they trust. Over 65 million of them are on the platform every day.
A founder with a consistent LinkedIn presence gets onto that mental shortlist before the buying process starts. Something a company ad or blog post rarely does. And that trust over time helps the vendor make the short list when the buyer goes to AI.
Most founders are leaving this channel completely untapped.
Three things come up almost every time I talk to a B2B founder about LinkedIn:
No time. When you're managing product, fundraising, hiring, and customers simultaneously, LinkedIn feels like a nice-to-have. It gets deprioritized and eventually abandoned.
No clarity on what actually works. There is a lot of conflicting advice on what makes LinkedIn work and most of them are largely wrong. Without a clear signal that the effort is working, it's rational to give up.
No way to connect it to the pipeline. Impressions and follower counts don't show up in a board deck. If you can't draw a line from LinkedIn activity to qualified conversations, it's nearly impossible to justify the time.
All three are solvable. But solving them starts with replacing guesswork with data.
At Rocksalt AI, we analyzed 6,753 LinkedIn posts from 39 B2B marketing influencers — people like Dave Gerhardt, Elena Verna, Rand Fishkin, and Kieran Flanagan — to find out what actually drives engagement. Crucially, we controlled for follower count and segmented posts by funnel stage, which eliminates the two biggest flaws in most LinkedIn advice: comparing people with wildly different audiences, and ignoring what different post types are actually designed to do.
The findings challenge a lot of what gets passed around as LinkedIn best practice. A few that stood out:
1. Having a healthy mix of content type is important, don’t over-index on just promoting your company
Top influencers split their posts roughly 50% marketing-specific (MOFU), 30% personal/general (TOFU), and 20% company-promotion (BOFU). Most executives over-index on self-promotion — this data suggests that's a mistake.
2. Posting more than 1–2 times per week hurts, not helps. There is a negative correlation between posting frequency and average engagement above roughly two posts per week. LinkedIn itself has acknowledged the platform sees higher audience fatigue than other social networks. More is not better.
3. Personal posts (think selfies, or career advice type of posts) are not automatically the highest-engagement option. The popular belief that personal stories and photos always win is not supported by the data. Median engagement is nearly identical across personal, industry, and company-focused posts. Quality drives performance, not category.
4. Below 50K followers, audience size doesn't predict engagement at all. For most founders, chasing follower growth won't move the needle. Content quality matters far more. That's actually good news — it means you don't need a big audience to start.
The full analysis, including the raw dataset for anyone who wants to run their own numbers, is here: rocksalt.ai/linkedin-2026-analysis
When I was at HubSpot running Business Development, the company bet everything on inbound. The insight was simple: buyers were searching on Google, and content was how you earned their trust before they were ready to buy. If you became the authority in your space, inbound got easier and outbound got more effective. The whole machine ran on that principle.
AI has changed the medium. It has not changed the principle.
In 2025 and 2026, AI has made generic content free and abundant. Blog posts, LinkedIn updates, email sequences - all of it can be produced at near-zero marginal cost by anyone with a ChatGPT account. The result is a collapse in signal quality. Buyers are swimming in polished, competent, forgettable content.
What cuts through is exactly what AI cannot replicate: a founder who has lived inside a problem for five years. The specific frustration that made them start the company. The insight they formed from a hundred customer conversations. The conviction that comes from having real skin in the game.
That authenticity is not a soft advantage. It is a structural moat and it becomes more defensible every month that AI-generated content gets cheaper and more widespread. The Rocksalt data makes this concrete: the LinkedIn influencers who held their engagement through LinkedIn's 360Brew algorithm change in late 2025 were the ones posting consistent, genuine expertise. The ones who leaned on volume and formatting tricks saw declines.
The HubSpot inbound thesis still holds. It has just moved from companies to people.
The research translates into a clear operating framework for any founder who wants to build this channel seriously.
1. Get clear on what you want to be known for. It doesn't have to be one single thing, but it cannot be five. The founders with the strongest LinkedIn presence are legible — you understand what they're about within a few posts. Find the intersection of your genuine expertise and your ICP's most pressing unsolved problems, and own that space.
2. Follow the 50/30/20 content mix. Roughly 50% of posts should be industry insight — your informed take on what's happening in your market. Around 30% should be personal and career perspective — lessons learned, hard-won experiences, honest reflections. The remaining 20% can be company and product focused. Most founders invert this, spending the majority of their posts on self-promotion. The data shows that makes them easy to scroll past.
3. Post 1–2 times per week — and mean it. The data is clear that frequency above this threshold hurts rather than helps. One genuinely useful post outperforms five average ones. Post when you have something worth saying, not because you feel you should be posting.
4. Structure and format matter more than you think. Top-performing posts contain 4–7 times more list-formatted items and significantly more line breaks than lower-performing posts. Write the way a busy person needs to read. Short paragraphs, clear structure, no walls of text.
The founder's authentic expertise is the most underleveraged GTM channel in B2B — and AI has made it more valuable, not less.
The data from Rocksalt's analysis of 6,753 LinkedIn posts shows: post 1–2 times per week, ditch hashtags, use lists and structure, and follow the 50/30/20 content mix. Below 50K followers, content quality matters far more than audience size. And every great LinkedIn presence needs a clear, ownable point of view at its centre.
What worked for HubSpot at the company level — becoming the trusted authority so inbound does the heavy lifting — now works for founders at the individual level. The buyers are on LinkedIn. The channel is ready. The question is whether the founder shows up.
Arjun Moorthy is the Co-founder & CEO of Rocksalt AI, a platform that helps B2B founders and executives build consistent LinkedIn and Reddit presence. Before Rocksalt, he was Head of Business Development at HubSpot.