Skip to content
Imagine

What B2B Actually Posts and What Works

Most advice about LinkedIn is vibes. Somebody's post did well last Tuesday, so now there's a carousel about hooks.

We wanted the boring version of the truth, so we built an index and measured it.

The Imagine B2B Index

The Index is a curated panel of 101 B2B software companies that we scrape, classify, and re-measure on a daily cycle: every original post, its engagement over time, and the people who wrote it - 744 tracked executives and employees so far, 32,277 original posts in the August 2026 snapshot.

It is not a random sample of LinkedIn. It is the part of LinkedIn a B2B team actually competes with:

  • Category leaders people benchmark against: Rippling, Deel, Ramp, Brex, Mercury, MongoDB, ElevenLabs, Supabase.
  • High-growth startups, many YC-backed, running founder-led content: the companies you see on Demo Day and six months later in your pipeline.
  • Vertical clusters mirroring real competitive sets - fintech, dev tools and databases, AI infrastructure and voice, customer support AI, brand protection, benefits.

Inclusion rule, in one line: venture-backed or public B2B software companies with an active LinkedIn presence, tracked completely rather than sampled.11Companies enter the Index when tracked for a customer's competitive set or as category-defining B2B brands. Known companies are indexed up to their last 100 posts on first scrape (500 for new additions), then continuously. See the methodology. The Index grows weekly; figures below are the all-time view as of August 2026, original posts only (32,277 in scope), with 3.5M total engagements - 3.0M likes, 288.8K comments.

Finding 1: The market posts the most of what performs the least

Classify every post into nine categories and count. Then measure what each category earns.22Engagement is likes, comments, and shares per post. The figures use outlier-robust “smart” averages. Categories come from a vision-language classification pipeline reviewed for drift. The two rankings are close to inverted.

B2B companies post the most of what performs the least

Industry Insight, Event, and Product Update make up most posts but have the lowest average engagement. Celebratory Announcements, Personal Stories, Hiring, and Company Updates are less common and perform substantially better.

SHARE OF ALL POSTSAVG ENGAGEMENT PER POST (SMART)22.4%Industry Insight4821.2%Event6118.7%Product Update747.4%Hiring / Recruiting1227.2%Celebratory Announcement1627.2%Personal Story1276.9%Customer Highlight684.9%Company Update1214.2%Meme / Humor81

Source: the Imagine B2B Index · 101 B2B software companies · 32,277 original posts · 744 tracked executives · smart average = outlier-robust mean · pulled Aug 2026

The three most common categories - Industry Insight (22.4% of all posts), Event (21.2%), Product Update (18.7%) - are the three weakest performers, earning 48, 61, and 74 average engagement respectively. The three strongest - Celebratory Announcements (162), Personal Stories (127), Hiring posts (122) - together make up under 22% of what gets posted.

That is a 3.4x gap between the best category and the worst, and the market is spending 62% of its output at the bottom of the range.

Why? The bottom three are the easy ones. Industry insights, event recaps and product updates can be produced without a person attaching their own story to the post. The top three all require someone to be visible - a founder celebrating, a personal story, a team hiring. The market optimizes for what is easy to ship, not for what readers respond to.

The honest caveat: celebratory announcements are partly a consequence of good news, not just a cause of engagement. You cannot announce a Series D every week. But personal stories and hiring posts are repeatable, and they outperform industry insights by 2.5x.33Personal Story averages 127 engagements versus Industry Insight at 48 — 2.6x. Every category shown clears the dashboard's eight-post sample floor by orders of magnitude.

Finding 2: Media roughly doubles engagement

A post with media earns roughly twice the engagement of text alone

Video posts average 108.5 engagements, image posts 93.7, other formats 50, and text-only posts 49.

AVG ENGAGEMENT PER POST (SMART)Video108.5Image(s)93.7Other50Text only49text-only baseline: 49

Source: the Imagine B2B Index · 32,277 original posts across 101 B2B software companies · smart average engagement by media format · pulled Aug 2026

Video posts average 108.5 engagements, image posts 93.7, text-only posts 49. Attaching real media to a post is the single cheapest performance upgrade in this dataset - a 1.9-2.2x multiplier.

Note "real": the top-performing posts in the Index use authentic photos and footage - launch films, team photos, screen recordings - not stock or AI-generated images.

Finding 3: Everyone posts on the same days, for no measurable reason

When the B2B market posts is not when engagement lands

Engagement varies little by day despite most posts being published on weekdays. Sunday at 8 AM is the peak cell, but the weekend sample is small.

12a4a8a12p4p8p11pSun175MonTueWedThuFriSatPeak cell: Sunday 8 AM PT (175 avg) — weekends carry only 5.7% of post volume, so treat small samples with care.

Source: the Imagine B2B Index · average engagement by publish day and hour (Pacific) · weekdays carry 30,446 of 32,277 posts (94.3%) while per-post engagement is nearly flat · pulled Aug 2026

Weekdays carry 94.3% of all posts (30,446 of 32,277). Per-post engagement, however, is nearly flat across the week. The market has collectively decided weekends don't work, and the per-post data does not support the strength of that conviction.44Day-of-week medians run 35–42 across all seven days — roughly a 15% spread — while volume varies sevenfold between weekdays and weekends.

We would not tell you to move your content calendar to Sunday 8 AM because one heatmap cell peaked there - weekend cells have small samples and one viral post can light them up. The defensible claim is the opposite one: the day you post matters far less than what and who is posting. Our companion study of 477 YC founders makes the same point with proper controls: an apparent -13% Saturday penalty shrinks to about -7% and borderline significance once you control for who posts on weekends.

What we would do with this

  1. Rebalance the calendar away from unattributed industry commentary and toward people-led categories - stories, milestones, hiring.
  2. Put real media on everything that deserves reach.
  3. Stop agonizing over posting time; spend that attention on voice and specificity instead.

The Imagine B2B Index lives at benchmark.imagineai.me - every figure here is reproducible on the dashboard, and the methodology page documents engagement definitions, smart averages, and sample floors.

More from Research