LinkedIn Still Works. Most People Are Using It Wrong.
- robbie6761
- 14 hours ago
- 6 min read

By Henry McIntosh, Founder of 2112 Marketing and Noteworthy · Updated 2026-08-25
People love to complain about LinkedIn. Too much AI content, too much noise, too many people shouting into a feed that's stopped listening. In our latest Twenty One Twelve Growth Forum we flipped the script and asked:
“Can LinkedIn still drive commercial outcomes?
We put that question to three personal branding experts: Taryn Talley, Josh Huggins and Judi Hayes.
Their answer was more or less unanimous. LinkedIn, when used correctly, can be game changing for your business.
Is LinkedIn still worth the time?
Yes, but not for the reason most people think. The platform has surpassed 1.2 billion registered members worldwide, and it's still the default place professionals go, whatever else has launched around it. That scale isn't the argument on its own. Vanity metrics never were.
Split LinkedIn into two buckets. Are you booking real conversations through it? And are the sales conversations you're already having noticeably warmer because the other person already has context on you? If both are true, the platform is doing its job. If neither is, more posting won't fix it.

Where do you actually start?
Not with a content calendar. With your story.
Josh's advice was to lead with real experience and credibility, the specific background nobody else on a client shortlist can claim. Judi Hayes went further back: define your audience with real precision before you write a word. "You can't post your way to authority," as she put it. The profile has to demonstrate who you help and how, or the posting is wasted effort.
Taryn Talley's version of the same idea: your headline and About section aren't about you, they're about what you represent to the person reading them. She estimates hers shows up in ten or more places across the platform. Most people never optimise it once. As with all good marketing, the most important person is the customer - you need to have an intricate understanding of who you are trying to reach.
What and how frequently should you actually post to LinkedIn?
Lived experience, not commentary on the news. Taryn's rule: if you're producing content like a news aggregator, expect the engagement to match. The better source material is sitting in your client calls. Josh's suggestion was blunt and practical: look at the themes coming up in the conversations you're already having, and build content around what's real rather than what you think you're supposed to say.
The consistency question came up too - how much should you post on LinkedIn? The panel agreed on the answer even if their cadences differ: find a rhythm you can sustain, not a rule you read somewhere. One, two, three posts a week or more, it doesn’t overly matter. What matters is having something authentic to add and doing so consistently, not filling a slot for the sake of it.
One thing all three were unambiguous about: don't let AI write the post for you. LinkedIn is already deprioritising content that reads as generated, and audiences can tell. Use AI to think faster. Don't use it to skip having a take.
If you’re struggling to find your voice, try Noteworthy - the AI video interview platform. One 15 minute interview produces a week of multimedia content ready to edit and publish. It’s the new cheat code for ambitious marketers, agencies and founders.
Do comments and DMs still work?

More than posts do, in some respects. Comments and direct messages are two of the few things not directly affected by algorithmic suppression, which all three panellists said is real, whatever LinkedIn calls it officially. Judi's practical fix: track members reached against impressions, not impressions on their own. A high impression count with a low reach percentage is the tell.
Anecdotally, I’ve often seen my comments achieve greater impressions than my posts, but the panel agreed this is often the case - and shows the power of diving in and writing relevant comments. But once again, they stressed not to use AI to generate comments.
Direct messaging came up repeatedly as the underused channel, on the condition that it's genuinely personalised. Taryn described her own approach: read the person's profile and recent posts properly before reaching out, and write something that could only have been sent to that person. The alternative, the generic scraped-looking pitch, does more damage to credibility than it saves in time.
Why is this suddenly an AI search story?
Because it is one. Independent research from Semrush, analysing 325,000 prompts across ChatGPT Search, Google AI Mode and Perplexity, found LinkedIn is the second most-cited domain overall behind Reddit, appearing in roughly 11% of AI responses on average. A separate analysis by Profound, tracking six AI models from November 2025 to February 2026, found LinkedIn had become the single most-cited domain for professional queries specifically, with citation frequency more than doubling in that window.
The detail that matters most for anyone posting under their own name: on ChatGPT and Google AI Mode, 59% of LinkedIn citations come from individual creators, not company pages. Perplexity is the reverse, favouring company pages 59% of the time. Build both, but don't assume the company page is doing the heavy lifting on its own (Semrush, 2026).
Judi's advice on the mechanics translates directly into what the AEO research is finding: publishing articles and newsletters gives content a life beyond the feed, because it gets indexed. Several of her own clients have told her they found her through ChatGPT rather than a search engine. Taryn's practical add: think of your website as your home base, and LinkedIn as one part of a wider digital footprint that includes press mentions, podcasts and guest appearances. The bigger and more consistent that footprint, the more often the LLMs surface you.
On the content side, the same structural changes that help human readers help AI systems extract you: clear FAQs, direct-answer H1s, and headings phrased the way people actually search rather than the way you'd write a chapter title.
Personal brand or company brand?

Personal, when reach is the goal. Josh's read: LinkedIn's algorithm rewards a personal story with real history over a company page post, because trust in institutional content has fallen while trust in identifiable individuals hasn't. Judi's own experience backs this up: a financial adviser she used for years changed firms, and she followed the person, not the brand he'd been representing.
That doesn't make company pages redundant. Taryn's point on employee advocacy was one of the sharper ones on the call: get your team commenting and resharing under their own names, not just posting from the company account. She's taken agencies into industry top-100 rankings by getting advocacy participation above roughly 9%, off briefs as simple as a Slack message flagging which posts need support and why.
Where this leaves you?

None of this requires doing everything at once. The panel's shared advice on bandwidth was to repurpose deliberately: one good conversation becomes a LinkedIn post, that post becomes a newsletter, that newsletter becomes a blog post. Depth in one place, redistributed well, beats fresh content everywhere. The platform's usefulness hasn't changed much in the past year. What's changed is who's watching. Every post is now also being read by something that isn't a person, and that widens the audience for the same work rather than adding more of it.
Useful Links

Watch the full session, including audience Q&A on suppression, virality and what to do when consistent posting isn't converting: https://www.youtube.com/watch?v=aeqGO8xAn9I&t=140s
Get invitations to future Forums and recordings of past ones via the Twenty One Twelve newsletter.
We also mentioned Noteworthy in the session, our tool for turning a single recorded conversation into LinkedIn posts, clips, blogs and newsletters. The beta is open if you'd rather talk than type: benoteworthy.io
FAQ
Is LinkedIn still commercially viable for B2B in 2026?
Yes, and the case for it got stronger this year. LinkedIn is the second most-cited domain in AI search overall and the most-cited for professional queries specifically, showing up in around 11% of AI responses on average (Semrush, 89,000 URLs analysed, March 2026). That's on top of over 100 million verified human members. Commercial viability now runs on two tracks: real conversations booked through the platform, and whether AI tools cite you when your buyers ask the questions you'd want to answer.
Does personal branding outperform company pages on LinkedIn?
For reach and AI citation, generally yes. Individual creators account for 59% of LinkedIn citations on ChatGPT and Google AI Mode, though company pages still lead on Perplexity. The strongest approach combines both.
Is LinkedIn algorithm suppression real?
The panel's consensus was yes, though LinkedIn doesn't confirm it directly. The practical fix is tracking members reached against total impressions, and leaning on comments and DMs, which aren't affected the same way.
Is LinkedIn content really being cited by ChatGPT and other AI tools?
Yes, and it varies significantly by platform. LinkedIn shows up in 14.3% of ChatGPT Search responses and 13.5% of Google AI Mode responses, but only 5.3% of Perplexity responses (Semrush, March 2026). Separately, Profound's tracking found LinkedIn's citation frequency on ChatGPT more than doubled between November 2025 and February 2026, the fastest-moving domain they measured that year.
How often should you post on LinkedIn? There's no universal number. The panel's advice was to post when you have something worth adding, and prioritise quality and engagement (particularly commenting) over hitting a fixed frequency.
Henry McIntosh is founder of Twenty One Twelve, a B2B marketing agency working with clients in financial services, pharma and other regulated sectors, and Noteworthy - the AI video interview platform which helps you extract insight and compelling content from the people that matter most to your business.




Comments