LinkedIn document and carousel posts drive the most B2B leads in 2026, outperforming both video and single-image posts on engagement and lead capture. UK B2B accounts see document posts average around 6.10% engagement, with Socialinsider reporting median carousel engagement near 21.77%, roughly 196% higher than video and 585% higher than plain text. Multi-slide carousels earn around 3.4x the reach and 2.1x the engagement of a single image. Video reach fell roughly 35% year on year as LinkedIn's algorithm favoured saveable educational formats. LinkedIn drives 75 to 85% of all B2B social leads, and 40% of marketers rate it their best channel for quality leads. The honest verdict: lead with carousels and documents to capture and educate, use video to build founder trust, and stop confusing engagement rate with pipeline. Format choice should map to funnel stage, not vanity metrics.
Last updated: June 2026
Carousels and native document posts drive the most B2B leads in 2026, and the gap over video is wider than most agencies admit. Across UK B2B company pages and founder profiles, the saveable multi-slide formats consistently produce more profile clicks, more inbound connection requests, and more comment-to-DM conversations than any other content type. Video remains powerful for one specific job, which is building trust in a face, but it has stopped being the reach engine it was two years ago.
Here is the part competitors skip. The question is not "which format gets the most likes" but "which format moves a stranger toward a sales conversation". Those are different questions with different answers. A carousel that teaches a buyer how to fix a real problem earns a save, and a save is a deferred intent signal. The buyer comes back, reads your profile, and books a call. A video that makes someone laugh earns a like and is forgotten by lunchtime. Both have a place. Only one of them reliably fills a pipeline.
Our view, formed from running content for UK software, automation and professional-services clients, is blunt: if you can only commit to one format, build carousels. They are cheaper to produce than video, they survive longer in feeds, and they convert browsers into leads because they deliver genuine value before asking for anything. Video is the second priority, used sparingly and deliberately. Single images and pure text posts are distribution glue, not lead engines.
The summary table below sets the stage for everything that follows.
| Format | Primary job | Lead strength | Production cost | 2026 algorithm trend |
|---|---|---|---|---|
| Carousel / document | Educate, capture, nurture | Highest | Low to medium | Favoured, rising |
| Native video | Build founder trust | Medium | High | Reach down ~35% YoY |
| Single image | Announce, reinforce | Low | Low | Flat |
| Text only | Distribute, start debate | Low to medium | Very low | Stable, high comments |
If you take one thing from this article, take this: lead with carousels, support with video, and measure leads rather than applause. The rest of this guide proves why, with the numbers, the funnel logic, and a practical cadence you can run from next Monday. If you would rather have a team build and automate the whole engine for you, our AI automation agency in London does exactly that.
The UK data tells a clear story: document and carousel posts top the engagement table, video sits in the middle, and text and single images trail. Document posts on LinkedIn average around 6.10% engagement on UK B2B accounts, with broader datasets such as Socialinsider reporting median carousel engagement as high as 21.77%. That figure is roughly 196% higher than video and 585% higher than plain text. Native document engagement has grown around 14% year on year, while video reach has fallen by approximately 35% over the same period as the algorithm rebalanced toward saveable, dwell-heavy content.
Reach numbers reinforce the point. A multi-slide carousel earns around 3.4 times the reach and 2.1 times the engagement of a single image post. A well-built 10-slide deck can pull two to three times the impressions of an equivalent text update because each swipe is a fresh dwell signal that tells the algorithm the content is worth surfacing to more people. Dwell time, not likes, is the currency LinkedIn now optimises for, and carousels manufacture dwell time by design.
The lead-generation context matters as much as the engagement context. LinkedIn drives 75 to 85% of all B2B leads generated through social media. Around 93% of B2B marketers use the platform, 84% say it delivers the best value of any social channel, and 40% rate it the single best channel for quality leads. So the platform itself is settled as the B2B channel. The only open question is which format on that platform earns the most pipeline, and the data points firmly at the multi-slide saveable formats.
| Format | Approx. UK B2B engagement | Reach vs single image | Year on year trend |
|---|---|---|---|
| Document / carousel (median) | 6.10% to 21.77% | ~3.4x | +14% |
| Native video | ~3.0% to 6.6% | ~1.5x to 2x | -35% reach |
| Single image | ~3.0% | 1x baseline | Flat |
| Text only | ~1.5% to 3.0% | ~1.2x | Stable |
A word of caution on these numbers. Engagement rates vary by industry, follower count and the quality of the writing, and any single benchmark is an average that hides enormous spread. A poor carousel will underperform a brilliant video every time. The format gives you a structural advantage, not a guarantee. Treat the percentages as a starting prior, then measure your own account, because your audience is the only benchmark that pays your invoices.
Carousels and documents win because they combine three things the LinkedIn algorithm rewards heavily in 2026: high dwell time, high save rates, and a low barrier to consumption. Every swipe is an interaction the platform reads as interest, so a single carousel generates seven to ten micro-engagements from one reader where a text post generates none. That accumulated dwell time is the strongest organic reach signal LinkedIn has, and saveable formats produce it almost automatically.
The second reason is the save. When someone saves a carousel, they are telling the algorithm the content has lasting value and telling themselves they intend to return. Saves are the closest thing LinkedIn has to a deferred intent signal, and intent is what separates a lead from a like. A buyer who saves your guide to choosing a CRM is a buyer who is, on some level, in market. That is why our work on custom CRM development in London consistently generates inbound from educational carousels rather than from promotional video.
The third reason is psychological and it is the one most teams underrate. Documents feel like free value, not advertising. A reader swiping through a tightly designed ten-slide breakdown of a problem they have is receiving a gift, and reciprocity does the rest. By the time slide ten asks them to book a call or comment for a template, you have already earned the right to ask. Video can do this too, but it demands more from the viewer: sound, attention, and a willingness to watch a face. Carousels work in silence, on a commute, with one thumb.
The honest rule we apply: if the content teaches something a buyer can act on today, make it a carousel. The format and the value compound.
None of this means video is dead. It means video has a narrower, more specific job in 2026, which the next section covers honestly.
Video still wins when the goal is trust in a person rather than transfer of information. No carousel can replicate the effect of a buyer watching a founder explain how they think, hearing their tone, and deciding they are someone worth doing business with. For high-value B2B services where the buyer is essentially buying judgement, that human signal is decisive, and video delivers it in a way no other format can.
There are four situations where we actively recommend video over a carousel. First, founder-led thought leadership, where the personal brand is the asset. Second, complex demonstrations, where seeing a product or process in motion beats describing it. Third, top-of-funnel awareness for a cold audience that does not yet know you exist, where a scroll-stopping face outperforms a text-heavy slide. Fourth, social proof, where a short clip of a client talking carries more weight than a written quote ever will.
The catch is cost and consistency. Video that builds trust has to look and sound credible, and that means decent lighting, clear audio, and tight editing. A shaky, mumbled clip damages trust rather than building it. Because production cost is high, most teams cannot sustain a high-frequency video cadence, and inconsistent video underperforms consistent carousels. This is why our recommendation is video as a deliberate supplement, not a primary engine.
| Use video when | Use a carousel when |
|---|---|
| Selling judgement or a founder's expertise | Teaching a repeatable, actionable method |
| Demonstrating a product in motion | Breaking down a framework or checklist |
| Reaching a cold awareness audience | Nurturing an audience that already knows you |
| Showing a client testimonial naturally | Capturing inbound via a comment-for-resource CTA |
Be sceptical if anyone tells you video is finished on LinkedIn. It is not. Its reach advantage has shrunk, but its trust advantage is intact, and trust is what closes B2B deals. The mistake is treating video as a volume play. Treat it as a precision instrument and it earns its keep. If you want production and distribution handled end to end, our business process automation team can systematise the workflow so your video output stays consistent without eating your week.
Engagement rate is the wrong primary metric because likes and comments do not pay invoices, and the two often move in opposite directions to pipeline. A controversial text post can earn 400 comments and zero sales conversations, while a quiet, technical carousel earns 30 saves and three booked calls. If you optimise for engagement, you will be tempted to chase reactions, and chasing reactions pulls your content away from the buyers who actually purchase. This is the central error in almost every "which format wins" article, and it is the gap we are deliberately filling.
The metrics that correlate with leads are different from the metrics that correlate with applause. Saves indicate deferred intent. Profile clicks indicate research. Document downloads or comment-for-resource actions indicate hand-raising. Direct messages and connection requests from your content indicate a conversation has started. None of these appear in a headline engagement-rate figure, yet all of them predict revenue better than likes ever will.
Here is the reframe in practice. We track a lead funnel for content, not an engagement leaderboard. The shape looks like this.
Once you measure this way, the format debate resolves itself. Carousels win because they manufacture saves and profile visits, which sit at the top of the lead funnel. Video wins at the trust-to-call conversion step, which sits lower down. The honest conclusion: stop reporting engagement rate to your board and start reporting cost per booked call by format. The numbers will reorganise your strategy faster than any benchmark table. A simple GoHighLevel automation pipeline can attribute each booked call back to the post that started it, which is where format strategy stops being guesswork.
Our stance is firm here. Any agency that reports your LinkedIn performance purely in engagement percentages is either avoiding accountability or does not understand B2B buying. Demand lead metrics.
Map carousels to the top and middle of the funnel, video to the trust-building middle, and text to distribution across the whole funnel. The mistake most teams make is treating LinkedIn as one undifferentiated feed where you just post "content". A B2B buyer moves through stages, and each stage responds to a different format. Get the mapping right and your content stops being noise and starts being a sequence that walks a stranger toward a contract.
At the awareness stage, the buyer does not know they have a problem you can solve. Here a strong cover-slide carousel or a scroll-stopping founder video does the heavy lifting, because the job is interruption and education. At the consideration stage, the buyer knows the problem and is comparing approaches. This is carousel territory: frameworks, comparisons, mistakes to avoid, the exact value that positions you as the credible choice. At the decision stage, the buyer is choosing a provider, and trust signals win: client video testimonials, case-study carousels, and founder content that proves judgement.
The decision matrix below is the practical core of this article. It maps format to funnel stage to the type of lead you can expect, with rough UK effort and cost guidance per piece.
| Funnel stage | Best format | Lead type produced | Effort per piece | Typical UK cost per piece |
|---|---|---|---|---|
| Awareness | Carousel + occasional video | New followers, profile visits | Medium | £120 to £350 |
| Consideration | Educational carousel | Saves, resource requests, DMs | Medium | £150 to £400 |
| Decision | Video testimonial + case-study carousel | Booked calls, demo requests | High | £300 to £900 |
| Distribution | Text post linking to the above | Comments, reach, debate | Low | £40 to £120 |
Notice that no single format owns the whole funnel. The teams that win on LinkedIn in 2026 run a portfolio, weighted toward carousels because that is where the volume of intent sits, with video reserved for the moments where trust must be built before a deal can close. If your content calendar is all of one format, you are leaving leads at whichever stage that format does not serve. The fix is not more posting; it is the right format at the right stage. For service businesses, pairing this with an AI chatbot on the landing page captures the buyers who click through but are not yet ready to book a call.
The single most important technical decision is dimensions: post carousels and documents at 1080 by 1350 pixels in 4:5 portrait, because portrait occupies more vertical feed space on mobile and stops the scroll. Square 1:1 works but is weaker on mobile, and 16:9 landscape is the worst choice for feed real estate. Roughly 70% of LinkedIn consumption is on mobile, so design for the thumb first and the desktop second.
For carousels and documents, the sweet spot is seven to ten slides. Fewer than five feels thin and fails to build dwell time; more than twelve loses people before the CTA. Each slide should carry one point and one point only, with generous white space, large readable type, and a clear visual swipe cue on the first slide so people know to keep going. The cover slide is everything: it is the headline that decides whether anyone swipes at all, so it should name the reader's problem or promise a specific outcome in six words or fewer. The final slide is your CTA: comment for the template, book a call, or follow for more.
For video, keep it short, captioned, and front-loaded. Captions are non-negotiable because most B2B video is watched on mute in an open-plan office. Hook in the first three seconds, keep founder pieces under 90 seconds and demos under three minutes, and export in a 4:5 or 1:1 ratio for the same feed-space reason as carousels. Audio quality matters more than camera quality: a clear voice on a phone beats a mumble on a cinema camera.
| Spec | Carousel / document | Native video |
|---|---|---|
| Best dimensions | 1080 x 1350 (4:5) | 1080 x 1350 (4:5) or 1:1 |
| Optimal length | 7 to 10 slides | 30 to 90 seconds |
| Rule per unit | One point per slide | Hook in first 3 seconds |
| Captions | Native (it is all text) | Mandatory burned-in captions |
| CTA placement | Final slide | Verbal plus on-screen text |
A practical design checklist we run before any carousel ships:
Specs are necessary but not sufficient. A perfectly formatted carousel with weak ideas still fails. Get the dimensions and length right so the format never works against you, then put your real effort into the thinking on each slide.
A winning 2026 cadence posts four to five times per week with a roughly 50/20/30 split: half carousels and documents, a fifth video, and the rest text posts that distribute and amplify the heavier formats. This is the hybrid model that beats single-format strategies, because it serves every funnel stage in a single week rather than betting everything on one content type. Consistency over months matters more than any individual post, so the cadence has to be sustainable, which is why carousels anchor it.
The logic of the split is straightforward. Carousels do the lead-capture and nurturing work, so they get the largest share. Video does the trust work, so one credible piece per week is enough; more is a bonus, not a requirement. Text posts are cheap and fast, and they keep your name in the feed between the heavier pieces while driving comments that lift reach for everything else. Run this for 90 days and you build the compounding presence that produces inbound, rather than the start-stop posting that produces nothing.
Here is a concrete weekly template you can adapt.
The piece almost everyone gets wrong is repurposing. One good carousel is not one post; it is a content engine. The same ten slides become a blog article, a newsletter section, five standalone text posts, and the script for a short video. Producing genuinely new content five times a week is unsustainable for most teams, and the unsustainability is why most LinkedIn strategies die at week six. Repurposing is how you keep the cadence alive without burning out, and automating the repurposing is how you keep it alive at scale. This is precisely the kind of repetitive, rules-based workflow our AI automation agency in London builds for clients, so the engine runs whether or not anyone remembers to post on a Friday afternoon.
You capture LinkedIn leads lawfully by collecting data with a clear lawful basis, telling people exactly what you will do with it, and giving them a genuine choice. Under UK GDPR and the Data Protection Act 2018, a LinkedIn comment is not consent to be emailed, and connecting with someone is not permission to add them to a newsletter. This is the compliance gap that competitor articles ignore entirely, and getting it wrong exposes UK businesses to complaints and Information Commissioner's Office action.
The practical principles are not complicated. When someone comments "send me the template", you may message them the template, but you should not silently add them to a marketing list off the back of that comment. If you want to email them ongoing marketing, you generally need their consent, and that consent has to be specific, informed and freely given, which means a real opt-in rather than a pre-ticked box. Keep a record of when and how each person consented, because under accountability you must be able to demonstrate it. And honour unsubscribe and deletion requests promptly, because the right to object and the right to erasure are not optional.
A compliant LinkedIn-to-CRM flow looks like this in practice:
Our honest position: compliance is not a reason to avoid lead capture, it is a reason to do it properly, and properly built systems convert better anyway because trust is part of the offer. When we build a GoHighLevel lead pipeline or a custom CRM for a UK client, consent logging and clean unsubscribe handling are baked in from day one, not bolted on after a complaint. If you are unsure whether your current flow is lawful, the safest assumption is that connecting and commenting are not consent, and you build from there. The ICO publishes clear, free guidance on direct marketing that every UK B2B team should read before they capture a single lead.
Softomate builds and automates the entire LinkedIn lead engine for UK B2B businesses, from carousel production to compliant CRM capture, on a fixed-quote basis with no surprise invoices. We are a London-based automation and software agency in Stanmore (HA7), and we run a five-stage process designed to take you from inconsistent posting to a measurable, repeatable inbound system within weeks rather than months. The goal is not more content for its own sake; it is more booked calls per pound of effort, tracked end to end.
The five stages are deliberately sequential, because skipping the strategy and tracking steps is how most LinkedIn efforts waste money on content that no one can attribute to revenue.
Pricing is transparent and fixed before we start. The table below shows indicative 2026 starting points; your exact quote depends on scope, and we agree it in writing up front.
| Engagement | What it includes | Indicative timeline | Starting price |
|---|---|---|---|
| Strategy and template sprint | Funnel map, format mix, carousel and video templates | 2 to 3 weeks | From £1,450 |
| Automation and CRM build | LinkedIn-to-CRM flow, consent logging, attribution | 3 to 5 weeks | From £2,900 |
| Managed content engine (monthly) | 4 to 5 posts a week, repurposing, reporting | Ongoing | From £1,200 / month |
| Full engine (build plus 3 months managed) | Everything above, end to end | ~12 weeks | From £6,500 |
Every engagement is fixed-quote. You will know the price before we begin, and you will not receive an invoice for anything we did not agree. We would rather scope carefully and quote honestly than win a project on a low number and surprise you later. If you want the engagement metrics to translate into actual booked calls, that translation is the entire point of how we work. Explore the full business process automation service or get in touch via our contact page to talk through your funnel.
Yes. Carousels and native document posts remain the highest-engaging LinkedIn format in 2026, with UK B2B document engagement around 6.10% and broader median carousel engagement reported near 21.77%. They earn roughly 3.4 times the reach of a single image because each swipe adds dwell time, the strongest organic reach signal the algorithm uses.
No, but its reach advantage has shrunk by around 35% year on year. Video is no longer the volume engine it was. It now does one job well: building trust in a founder or showing client proof. Use it deliberately for awareness and decision-stage trust, not as your primary lead-capture format.
Seven to ten slides is the sweet spot. Fewer than five feels thin and fails to build dwell time; more than twelve loses readers before the call to action. Put one point on each slide, lead with a strong cover slide that names the reader's problem, and finish with a single clear CTA.
Use 1080 by 1350 pixels in 4:5 portrait for both carousels and video. Portrait occupies more vertical feed space on mobile, where roughly 70% of LinkedIn consumption happens, so it stops the scroll more effectively than square or landscape. Always design for the mobile thumb before the desktop view.
Four to five times per week is the proven cadence, split roughly 50% carousels and documents, 20% video, and 30% text. Consistency over 90 days matters more than any single post. Repurpose one anchor carousel into the rest of the week's content so the cadence stays sustainable without burning out your team.
Yes, by a wide margin. LinkedIn drives an estimated 75 to 85% of all B2B leads generated through social media. Around 93% of B2B marketers use it, 84% say it delivers the best value of any channel, and 40% rate it the single best channel for quality leads. It is the settled B2B platform.
A save signals deferred intent: the reader found your content valuable enough to keep and intends to return. Likes are momentary applause that rarely correlate with pipeline. Saves, profile visits, resource requests and DMs predict revenue far better than engagement rate, which is why you should report cost per booked call rather than likes.
Not automatically. A comment or connection is not consent to be marketed to under UK GDPR and the Data Protection Act 2018. You may send what was requested, but ongoing marketing generally needs a specific, informed opt-in. Log consent, explain what people will receive, and make unsubscribing easy.
Indicative 2026 costs run from around £120 to £400 per carousel and £300 to £900 per video, depending on scope. A managed content engine typically starts from about £1,200 per month. Softomate works on a fixed-quote basis, so you agree the price in writing before any work begins.
Use both, mapped to the funnel. Lead with carousels to educate and capture intent at the consideration stage, and support with founder or testimonial video to build the trust that closes the deal at the decision stage. A single format leaves leads stranded at whichever stage it does not serve.
The verdict for 2026 is clear: lead with carousels and document posts, because they deliver the highest engagement, around 3.4 times the reach of a single image, and the saves and profile visits that sit at the top of the lead funnel. Use video deliberately for the trust-building moments that close B2B deals, where its 35% reach decline matters far less than its human signal. Run a four-to-five-posts-a-week hybrid cadence split roughly 50/20/30 across carousels, video and text, and repurpose one anchor piece into the rest. Above all, stop reporting engagement rate and start reporting cost per booked call by format, because likes do not pay invoices and LinkedIn still drives 75 to 85% of B2B social leads. Capture those leads lawfully, with logged consent and clean opt-outs. Get the format-to-funnel mapping right and a stranger's swipe becomes a booked call within weeks.
If you want this engine built, automated and measured for you, explore our GoHighLevel automation services in London or tell us about your funnel for a fixed quote.
Written by Deen Dayal Yadav, Founder of Softomate Solutions, a London-based AI automation and software development agency in Stanmore (HA7). With over 12 years building software, CRM and automation systems for UK businesses, he helps founders turn content effort into measurable pipeline rather than vanity metrics. Softomate Solutions is a registered company at Companies House and builds compliant, attributable lead engines for B2B clients across London and the UK. Learn more about Softomate and how we work.
We protect the real names of all clients featured in examples and case studies. Every testimonial is from a real client.
More from the blog
Work with us
Book a free 30-minute discovery call with DD and get a personalised automation roadmap.
Softomate Solutions Limited
Deen Dayal Yadav
Online
We use essential cookies to keep the site running. With your permission, we also use analytics cookies to understand how visitors use our site so we can improve it. No data is sold. Privacy Policy
