Twitter vs LinkedIn for B2B lead gen: what the data shows

The most-cited stat in this debate comes from a 2014 study of one vendor's own platform. It is directionally right and still routinely misread. The current numbers say something more useful: LinkedIn and X are doing two different jobs, and most B2B teams budget as if they are doing the same one.

Which Platform Wins for B2B Lead Gen: LinkedIn or X?

LinkedIn wins direct lead generation, and it is not close. Its visitor-to-lead conversion rate is 2.74%. X/Twitter converts at 0.69%, Facebook at 0.77%. Compilations of B2B lead gen studies put LinkedIn at 277% more leads than Facebook and Twitter combined. The Content Marketing Institute's 2025 B2B report found 84% of B2B marketers rated LinkedIn their most effective organic social platform, against 30% for X. Different measurement approaches, same direction. If you have limited time and one person to run social, the answer to which platform to staff first is settled.

That gap is real and it is still the wrong number to plan an entire program around. We run the same content to both platforms through a single automation layer, with identical UTM schemes on every link, and then we watch what lands in the CRM. LinkedIn attribution shows up 3-5x more reliably than X. Then we read the onboarding survey field where new customers type in how they heard about us, and X accounts for 2-3x more than the UTM data ever showed. Same content, same links, same tracking, two different stories.

That discrepancy is not a tagging bug we failed to fix. It is a behavior difference. X traffic arrives as direct or referral because people open links in an external browser, screenshot the post, or send it to a colleague who searches the brand name a week later. LinkedIn traffic survives the click, because the platform's own analytics and the campaign parameters both make it through. So a meaningful share of the measured conversion gap is a measurement gap, and any honest comparison has to correct both numbers before it means anything.

The decline on X is genuine though, and worth separating from the measurement problem. Its share of B2B social leads ran around 32% in 2020 and sits near 12.73% across 2024-2025. That is not drift inside a confidence interval, it is a role change. X stopped being where B2B deals start. It became where B2B reputations get built, which is a real job, just not the one most pipeline dashboards are configured to notice.

Audience size is the tiebreaker teams reach for, and it is the weakest input available. LinkedIn reports 1.2 billion members across 200+ countries. X claims 586 million monthly active users, though the last SEC-audited figure was 237.8 million mDAU in Q2 2022 and no audited number has been published since. One of those figures has been through an auditor and one has not. Neither tells you how many of your buyers are reachable, how many will see a post without paid support, or how many are in a buying window.

The allocation we would defend: LinkedIn carries direct pipeline, X carries reach into audiences that will never load a LinkedIn feed voluntarily. Fund LinkedIn as a conversion channel with conversion targets. Fund X as an awareness channel with awareness targets and a self-reported source field to catch what the tags miss. Teams that grade both against the same lead-count KPI end up killing X for underperforming at a job it was never doing.

The 80% LinkedIn Stat: Where It Came From and What It Actually Measures

The 80% figure comes from Oktopost, which analyzed more than 100,000 posts flowing through its own B2B social media management platform in 2014. It found LinkedIn responsible for roughly 80% of B2B social media leads, with X/Twitter around 12.73%. That is the source. Not an industry audit, not a panel study, not a cross-vendor benchmark. One tool's customer base, measured over a decade ago, describing the split of leads that arrived through social channels its customers were already using.

Two constraints travel with that number and almost never get quoted alongside it. The first is scope: it measures share of social-sourced leads, not share of all B2B leads. Social is a smaller slice of total B2B pipeline than most teams assume, so 80% of a slice is not 80% of anything a CFO cares about. The second is sample: the customers of a B2B social publishing tool in 2014 were disproportionately LinkedIn-first organizations. The measurement was taken inside a population already tilted toward the answer it produced.

None of this makes the finding wrong. The direction has been independently reconfirmed several times since, on completely different methodologies. Conversion-rate data puts LinkedIn at 2.74% against X at 0.69%. Marketer surveys put LinkedIn at 84% effectiveness against X at 30%. Ad-performance data puts LinkedIn as the only major network with positive B2B return. When independent methods point the same way, the direction is safe even when the specific number is not.

The X decline is on firmer ground than the LinkedIn share is. Multiple sources track X/Twitter's share of B2B social leads falling from approximately 32% in 2020 to about 12.73% in 2024-2025. The absolute percentages vary between studies, which is expected, but the slope is consistent across all of them. The slope is the finding. The decimal places are not.

So the defensible framing is this: LinkedIn dominates social-sourced B2B pipeline, the 80% figure is directionally accurate, and the precise number deserves a wide confidence interval and a citation to its 2014 origin whenever it appears. Citing it bare, as a current benchmark, is the kind of thing a prospect's analyst catches in a board deck. We have watched it happen.

There is a reading habit worth building here. Any time you meet a LinkedIn versus X comparison stat, ask which of three things it measures. Share of social leads, where LinkedIn wins by a large margin. Share of all B2B leads, where social is a smaller contributor than the headline implies. Or awareness contribution, where X is systematically undercounted because attribution tools cannot see most of what it does. Numbers from those three categories get quoted interchangeably, and they are not interchangeable.

What nobody has published, and what would settle this properly, is the same content pushed to matched audiences on both platforms through one distribution layer, with consistent tagging and CRM close data attached. That isolates the platform from the content and the audience. It is the only version of this comparison that is not partly an artifact of who happened to be measured.

Twitter vs LinkedIn for B2B: The Conversion Gap Is Wider Than the Headlines Suggest

On paid, the gap is not a gap so much as a sign change. Dreamdata's 2026 LinkedIn Ads B2B Benchmark Report, built on more than 220,000 customer journeys and over $28M in tracked spend, put LinkedIn's return on ad spend at 113%. Google Search Ads returned 78%. Meta Ads returned 29%. LinkedIn was the only major ad network in the study with positive B2B ROI. Everything else was underwater on a revenue basis, which is a stronger claim than any engagement-rate comparison can make.

Budget behavior tracks it. LinkedIn ad spend grew 31.7% year over year in 2025 while Google Ads grew 6%, and 39% of total B2B ad budgets now flow to LinkedIn, up from 31% in H1 2024. Budget holders are not answering a survey when they move money. They are acting on their own attribution data, which makes this revealed preference rather than stated preference, and revealed preference is the more reliable signal by a distance.

Then there is the part that complicates the whole comparison. One documented cross-posting experiment sent the same content to comparable audiences on both platforms and got 226,500 views on X against 17 reactions on LinkedIn. A second piece from the same experiment got 37,811 impressions on LinkedIn against 59 views on X. Not a small variance. Two pieces of content, two platforms, and outcomes that swap places entirely depending on where each one landed.

The interpretation matters more than the numbers. Audience placement, not content quality, drove those results. Which means that measuring content performance across platforms without holding distribution constant is not a valid comparison at all. The platform is a variable in your test, not a constant. Most content teams treat it as a constant, which is why their performance reviews are measuring the wrong thing.

The failure mode this creates has a shape we see often enough to name it: the rewrite spiral. A post underperforms on LinkedIn, so the team rewrites it. The rewrite underperforms too, so they change the format. Then the hook, then the cadence, then the whole content strategy. Four rounds of edits later, nothing has improved, because the original post was fine and the audience for it was on the other platform. Content quality was never the variable being tested.

The practical correction is cheap. Before rewriting anything that underperformed, publish it once on the other platform without changing the substance. If it lands there, you have a placement problem and your content is fine. If it dies on both, you have a content problem worth fixing. This takes one publishing cycle and it saves entire quarters of misdirected editorial effort.

It also reframes what a good cross-platform test looks like. You are not testing whether the content works. You are testing which audience is present, awake, and willing to engage with that specific idea. Those are different questions, and only the second one is answered by moving a post from one feed to another.

When X (Twitter) Makes More Sense for B2B Than LinkedIn

X beats LinkedIn when your buyers genuinely live there, and in a handful of verticals they do. Security, developer tools, fintech, and media companies see meaningfully higher ICP density on X than most B2B SaaS categories. If your buyer is a security engineer or an infrastructure lead, the conversations that shape their vendor shortlist are happening on X in public, in reply chains, with people they follow. No amount of LinkedIn conversion-rate advantage helps when the audience is not in the room.

Thought leadership is the second case, and the return on it is not small. Top-performing B2B marketers who use X for thought leadership report nearly 4x higher marketing ROI than peers who do not run thought leadership at all. Read that carefully before budgeting against it: the effect is tied to awareness and reputation, not to direct lead conversion. It shows up in shorter sales cycles, warmer first calls, and inbound that arrives already knowing who you are. It does not show up as a form fill with a source tag.

The exodus is real and it is worth understanding why people left. 67% of B2B companies remain active on X as of 2026, down from 82% in 2022. The reasons given for leaving break down as brand safety concerns at 48%, declining ROI at 34%, and platform uncertainty at 18%. Brand safety leads the list, which means most departures were a risk decision rather than a performance decision. Those are different judgments and they deserve to be made separately.

X's speed is a genuine operational advantage that has nothing to do with lead volume. Content on X lives 2-6 hours. On LinkedIn it lives 48-72 hours. That makes X a faster loop for testing angles, hooks, and framings before you commit them to a LinkedIn post that will represent you for the rest of the week. We use it that way constantly: an idea gets floated on X, the replies tell you whether the framing survives contact with practitioners, and the version that survives becomes the LinkedIn post.

The clearest argument for keeping X is what happens when teams drop it. X consistently generates 2-3x more self-reported awareness than UTM attribution ever captures, so on a dashboard it looks nearly free to cut. Then direct traffic softens, inbound gets colder, and nothing in the analytics explains it, because the channel that was feeding those numbers was never credited for them in the first place. The cut looks costless right up until the quarter closes.

If X is not your primary channel, run it small and run it honestly. Personal accounts, a cadence you can sustain without automation carrying the whole load, and a self-reported source field in onboarding so you have some visibility into what it is contributing. Grade it on awareness, mentions, and inbound that names it, not on CRM-attributed leads. Holding X to LinkedIn's conversion KPI guarantees you will eventually kill a channel that was working.

Company Pages Underperform Personal Profiles on Both Platforms

Posts from individual LinkedIn profiles get 8x the engagement of company page posts. Not 8% more. Eight times. The same pattern holds on X, where personal accounts consistently outrank brand accounts in algorithmic reach. If you are choosing where to put the next hour of content effort, that ratio should end the discussion before it starts.

The mechanism is distribution, not sentiment. LinkedIn's algorithm treats company page content as lower-trust distribution and gives it a smaller organic ceiling by default. To reach the audience a personal profile reaches for free, a company page needs paid amplification or employee advocacy pushing the post into individual networks. Both of those cost something. One costs media budget, the other costs internal coordination that most teams underestimate badly.

This is a distribution mechanics decision, not a personal branding preference. That distinction matters because the two get argued differently inside companies. Personal branding sounds like a founder's vanity project and gets deprioritized accordingly. Distribution mechanics sounds like what it is: the algorithm down-ranks brand voices in organic feed placement on both platforms, and working around that costs money you could spend elsewhere.

So the budget conclusion is direct. For B2B lead generation, effort invested in the personal profiles of founders and sales leaders produces more pipeline than the same effort invested in growing a company page. The company page still has jobs to do. It is the destination when someone checks whether you are a real business, it holds the careers content, it anchors paid campaigns. It is a credibility surface, not a distribution engine, and treating it as the latter wastes the budget.

The failure pattern here is metric substitution. A team sets a follower-growth target on the company page, hits it, reports it, and celebrates. Meanwhile meetings booked, pipeline created, and closed revenue have not moved, because page followers were never the input to any of those. The metric was easy to move and easy to report, which is exactly why it got chosen. If your social dashboard leads with page follower count, you are measuring the thing that is least connected to the outcome you want.

The objection we hear most is key-person risk: the profile that drives pipeline belongs to a person who can leave. That risk is real and it is manageable. Spread the load across several individual profiles rather than concentrating everything in one, keep the content substantive rather than personality-driven so it survives a handover, and make sure inbound routes into shared systems rather than one person's inbox. Also plan for the ceiling. LinkedIn caps any single account at 30,000 connections, so a profile-led strategy has a hard limit and needs more profiles before it hits one.

What B2B Teams Get Wrong About LinkedIn vs Twitter Attribution

Both platforms are mismeasured, in opposite directions, and comparing the raw numbers without correcting for that produces a confidently wrong answer. Contact-level attribution underreports LinkedIn's pipeline impact by 60% or more. UTM tracking undercounts X's awareness contribution by a similar order. If you put uncorrected LinkedIn numbers next to uncorrected X numbers, you are not comparing platforms. You are comparing two different measurement errors.

LinkedIn's undercount is structural, not fixable with better tagging. B2B purchases involve multiple stakeholders across long cycles, and the average time from first LinkedIn touch to closed-won runs 211 days per Dreamdata's analysis of more than 220,000 customer journeys. Contact-level attribution credits the individual who filled the form. It cannot see the four other people at that account who read three posts each and shaped the shortlist before anyone touched a form. The buying committee is invisible to a model built around single contacts.

211 days also breaks most attribution windows outright. Standard lookback settings expire long before that first touch can be connected to the deal, so the touch that started the account's interest gets dropped and the last-click channel takes full credit. Account-level measurement is the fix, and LinkedIn's Revenue Attribution Report, updated in July 2025, now incorporates 12 months of CRM activity. It is the starting point rather than the whole answer, but it is the piece most teams are missing.

X's problem is different in kind. X attribution is structurally broken without heavy UTM discipline plus a CRM field capturing self-reported source. X traffic frequently arrives as direct or referral with no campaign tag at all, because users open links in an external browser, screenshot the post and share the image, or send the link to a colleague who searches your brand name days later. The click that finally converts has nothing to do with X in the data, and everything to do with X in reality.

Our own numbers show both distortions at once. Running identical content on both platforms with identical UTM schemes, LinkedIn attribution comes in 3-5x higher in CRM data. Cross-reference that against what customers type into the onboarding survey, and X accounts for 2-3x more than the tags recorded. The tags are not lying. They are recording something narrower than what the question is asking about.

The summary is short. LinkedIn converts directly and shows up in attribution. X builds awareness that converts through other channels and gets credited to those channels instead. Neither number is usable raw. Both need correction before the comparison means anything, and any ROI model that skips that step will systematically over-fund the channel that happens to be more visible to your tracking stack.

The practical setup we would build first, in order: a self-reported source field in onboarding or on the demo form, account-level rather than contact-level attribution in the CRM, and an attribution window long enough to cover a 211-day cycle. Those three cost engineering days, not media budget, and they change which channel your data tells you to cut.

Running Both Platforms Without Letting One Cannibalize the Other

Voice drift is the most common failure mode when teams automate both platforms at once, and it has a recognizable signature. LinkedIn rewards a slower, more considered cadence: posts with setup, evidence, and a stated point of view. X rewards brevity and confidence. Apply the same templated or AI-generated voice to both and you get LinkedIn posts that read too thin and X posts that read too formal. The failure is not the format. It is the voice underneath the format.

The decay curve is consistent enough to predict. The first 2-3 posts on each platform perform at baseline, because the algorithm has not classified the new pattern yet. Then engagement drops 40-60% by week 3 as each platform down-ranks what it is seeing: low dwell time on LinkedIn, no reply engagement on X. Teams usually read that decline as content fatigue and respond by posting more, which accelerates it. Voice calibration per platform, not just format calibration, is the fix, and it is the step almost everyone skips.

Timing works differently enough that the two schedules cannot be merged. LinkedIn content lives 48-72 hours and the first 60-90 minutes after publishing set the algorithm's distribution ceiling, driven by early engagement from 1st-degree connections. Post into ICP active hours, Tuesday through Thursday, 8-10am in the target timezone, and have a small cluster of allied accounts engaging inside that window, and the post reaches 2nd and 3rd degrees. Miss the window and the ceiling is already set before most of your network has opened the app.

X gives you 15-30 minutes. After that a post is effectively dead unless a high-follower account retweets or quote-tweets it, which is not something you can schedule. The operational consequence is the one nobody says out loud: LinkedIn is substantially more automatable, because the timing is manageable and the cadence is slow enough for a queue to handle. X needs real-time human judgment during a window that closes in half an hour. That asymmetry is a large part of why operators running both through automation see higher LinkedIn ROI at lower human time cost.

LinkedIn's connection limits are the other place automation assumptions break. The invitation cap is not a fixed number, it is a trust score that decays and recovers. Published figures put it at 100-200 invitations per week, and high-trust accounts with Premium, a strong SSI, and acceptance rates above 25% do reach 150-200. New or low-trust accounts get capped at 15-30. We watch accounts sitting comfortably at 150 per week fall to 30 after a burst of low-acceptance sends, then climb back over 3-4 weeks of normal behavior. Running at 60-70% of apparent capacity with warm-up ramp schedules is what keeps an account in the high-trust band. Accounts that blast to ceiling immediately after connecting get flagged faster than cadence-throttled accounts on the same IP.

The messaging limits sit alongside that: 8,000 characters per LinkedIn DM, roughly 300 messages per week on a free account and around 700 with Sales Navigator, against a hard ceiling of 30,000 total connections per account. X's constraints are a different species entirely. They are API-layer and cost-based rather than behavioral, priced per call on the current pay-per-use model at $0.015 per post write and $0.005 per post read, capped at 2 million reads per month. LinkedIn punishes behavior patterns. X sends an invoice. Tooling built for one set of rules behaves badly under the other.

Engagement targeting, meaning commenting on ICP posts before your own goes live, works on both platforms and decays at different rates. On LinkedIn, 5-7 genuine comments on target accounts' posts in the 48 hours before you publish measurably lifts early engagement on your post, because the algorithm reads co-engagement as a relevance signal. That stays productive for 4-6 week cycles before it needs variation. On X the same pattern works for 2-3 weeks, then the algorithm starts treating high-comment-frequency accounts as engagement bait and suppresses organic reach. Rotate the X approach every 2-3 weeks. Rotate LinkedIn every 4-6. Running one schedule across both is how the X account quietly stops reaching anyone.

LinkedIn Ads Cost More Per Click, but the Qualified Lead Math Is Closer Than Expected

The headline cost difference mostly evaporates once you filter for qualified leads. LinkedIn's average cost per lead runs $87-125. Optimized funnels on X run $21-40, which looks like a rout until you separate leads from qualified leads. The average qualified B2B lead on X comes in around $98, inside LinkedIn's range. The cheap clicks were real. They were just buying a different quality of lead, and the difference showed up after the form, not before it.

The underlying economics explain why. LinkedIn's per-impression cost is 3-5x higher than X, and its lead yield per impression is 5-10x higher. Multiply those two and the effective cost per lead lands competitive despite the sticker shock at the top of the funnel. Teams that compare CPM or CPC across the two platforms are looking at the numerator alone and drawing conclusions the denominator reverses.

Execution quality accounts for a large share of the remaining spread. Top-quartile LinkedIn Ads users hit 48% MQL influence, 53% SQL influence, and 53% deal influence, roughly 1.5x the median user. That is a wide band inside a single platform. It means part of what gets attributed to LinkedIn being expensive is really an account being run at median, and it means the platform comparison is contaminated by operator skill unless you know where in that distribution your account sits.

The spend data settles the argument better than any benchmark table. LinkedIn ad spend grew 31.7% year over year in 2025 against 6% growth for Google Ads, and 39% of total B2B ad budgets now flow to LinkedIn, up from 31% in H1 2024. Those are budget holders reallocating money based on their own attribution, not marketers answering a survey about what they believe. Money moving is a harder signal than opinions moving.

The return data backs it up. LinkedIn posted 113% ROAS in Dreamdata's 2026 benchmark, drawn from more than 220,000 customer journeys and over $28M in tracked spend, against 78% for Google Search Ads and 29% for Meta Ads. It was the only major ad network in that study showing positive B2B ROI. Worth pairing with the 211-day average from first touch to closed-won: any paid measurement window shorter than that will show LinkedIn losing money on campaigns that eventually paid for themselves several times over.

The comparison worth running is cost per qualified pipeline opportunity, not cost per click. On that metric LinkedIn's higher CPL usually holds up, because the lead quality is structurally different rather than incrementally better. On the other side, the finding does not license unlimited LinkedIn spend. It says the platform earns its premium when the account is run well, the attribution window covers the actual sales cycle, and the qualification filter is applied to both platforms with the same strictness. Change any one of those and the math flips, which is exactly why the same two platforms produce opposite recommendations across different companies' spreadsheets.

Frequently asked questions

Which platform generates more qualified B2B leads in 2026, LinkedIn or X?

LinkedIn generates significantly more qualified B2B leads. Its visitor-to-lead conversion rate is 2.74%, nearly 4x X/Twitter's 0.69%. LinkedIn accounts for roughly 80% of B2B social media leads, while X/Twitter accounts for about 12.73%, down from approximately 32% in 2020. For direct pipeline generation, LinkedIn leads by a wide margin. X plays a real but secondary role, building awareness that often converts through other channels without showing up in attribution data.

Does the 80% LinkedIn lead stat still hold up, and where did it come from?

The 80% figure originated from a 2014 Oktopost study analyzing 100,000+ posts on their own platform. It measures share of social-sourced leads, not all B2B lead sources. The directional finding remains accurate: LinkedIn is far ahead of X for B2B leads. But the precise number is over a decade old and drawn from a single platform's customer base. It should not be cited as a universal benchmark without that disclosure.

When should a B2B SaaS company prioritize X over LinkedIn for growth?

X makes more sense than LinkedIn when your buyers are highly active there: security, developer tools, fintech, and media verticals see meaningfully higher ICP density on X than most B2B SaaS categories. X is also better for rapid content testing, since its 2-6 hour content lifespan cycles faster than LinkedIn's 48-72 hours. X does not convert as reliably as LinkedIn, but it builds top-of-funnel awareness that closes later through other channels, often without showing up in UTM attribution.

Is X still worth using for B2B marketing in 2026?

For brand awareness and thought leadership, yes. Sixty-seven percent of B2B companies remain active on X as of 2026, down from 82% in 2022. Top reasons for leaving are brand safety concerns (48%) and declining ROI (34%). X is not a reliable direct lead conversion channel for most B2B teams, but operators consistently find it generates 2-3x more self-reported awareness than UTM tracking captures. It builds pipeline that gets credited elsewhere.

Should B2B teams use personal profiles or company pages on LinkedIn and X?

Personal profiles outperform company pages on both platforms. Posts from individual LinkedIn profiles get 8x the engagement of company page posts, and the same pattern holds on X. Company pages on both platforms are algorithmically down-ranked relative to personal accounts. For B2B lead generation, building the personal profiles of founders and sales leaders drives more reach, more inbound, and more pipeline than growing a company page. This is a distribution mechanics decision, not a branding preference.

How often should a B2B brand post on LinkedIn vs X to maximize engagement?

On LinkedIn, 3-5 posts per week from personal profiles is the productive range; the first 60-90 minutes after posting set the algorithm's distribution ceiling, so timing to ICP active hours (Tuesday through Thursday, 8-10am in the target timezone) matters more than raw frequency. On X, daily posting is standard and high-frequency accounts are not penalized the same way, but each post's engagement window is only 15-30 minutes. The cadence requirements differ enough that they should be planned separately.

Can the same content be cross-posted from LinkedIn to X, or does it need to be native to each platform?

Cross-posting the same content without adaptation consistently underperforms on both platforms. LinkedIn rewards considered, longer-form posts with setup, evidence, and a clear point of view. X rewards brevity and a fast-moving, confident voice. When the same content runs on both without calibration, engagement decays 40-60% by week 3 as each platform's algorithm down-ranks content with low dwell time (LinkedIn) or no reply engagement (X). Format adaptation is part of the fix; voice calibration per platform is what most teams skip.

Is paid advertising on LinkedIn worth the higher cost per click compared to X?

The headline cost difference narrows once you filter for qualified leads. LinkedIn's average CPL runs $87-125; X's optimized funnels run $21-40, but the average qualified B2B lead on X comes in around $98 after unqualified clicks are removed. LinkedIn's 2026 return on ad spend was 113%, the only major ad network with positive ROI per Dreamdata's analysis of 220,000+ customer journeys. For B2B pipeline, the right comparison is cost per qualified opportunity, not cost per click.

What is the biggest attribution mistake B2B teams make when measuring LinkedIn vs X performance?

Using contact-level attribution instead of account-level measurement. Contact-level attribution underreports LinkedIn's pipeline impact by 60% or more because B2B deals involve multiple stakeholders and long sales cycles (the average LinkedIn touch-to-close runs 211 days). For X, the opposite problem applies: UTM data significantly undercounts X's awareness contribution because X traffic frequently arrives as direct or referral with no campaign tag. Both platform numbers require correction before they can be meaningfully compared.

What percentage of B2B social media leads come from LinkedIn vs Twitter?

By current estimates, LinkedIn accounts for roughly 80% of B2B social media leads, while X/Twitter accounts for about 12.73%, down from approximately 32% in 2020. These figures measure share of social-sourced leads, not all B2B lead sources, and they originate from platform-specific studies rather than independent audits. The directional split is consistent across multiple data sources: LinkedIn dominates direct social lead generation, while X's contribution is primarily at the awareness stage.

Sources and further reading

• LinkedIn 2025 B2B Marketing Benchmark Report (Ipsos, n=1,500 senior B2B marketers)
• LinkedIn Decision Makers Insights Report on C-suite presence and buyer behavior
• X Business advertising capabilities and audience reach data