+269 net new followers in September, a 10.9% gain and Colleen's strongest month on record. The curve climbed steadily all month and stepped up hardest around the two viral reaction posts (Sep 17 and Sep 29), closing at 2,733. Growth was broad-based rather than tied to a single day, a sign the reach is converting to audience.
"I am speechless" alone drove 216K, 60% of the month's reach, and the "Traits to Avoid" reaction added another 93K. Below those two, the educational hot takes (gap in resume, candidates using AI, job boards) formed a solid mid-tier. Everything else sat under 3.2K.
Each list ranks by a different signal: impressions for raw reach, saves for "I want to come back to this" intent, and engagement rate for resonance on smaller-reach posts. The big story is that the two reaction posts carried reach and topped saves, a break from August when reach and intent came from different posts. The engagement-rate list stays dominated by small-reach podcast clips.
Who got seen the most.
B2B's strongest signal. Readers bookmarking to revisit.
Smaller-reach posts that overperformed for their audience.
Image led volume (11 of 19) and, carried by the two screenshot reactions, dominated reach. Six podcast Videos, one long-form Text essay, and one reposted Quote rounded out the calendar. The screenshot-plus-reaction image is now the signature format.
Image's average (31.7K) is inflated by the 216K and 93K reactions. Strip those two out and the other nine images average near 4.9K, still the strongest format. Text ran on one 3.1K essay, and the six podcast videos averaged just 740.
We read each of the 19 posts and classified them by the job the post was doing. Five recurring patterns emerged. The breakdown shows volume and performance per pattern so we can see which deserve more calendar share next month and which to rethink.
Hot takes and reaction posts tower over everything else at 46.5K average, carried by the two viral screenshots. Educational thought leadership is a distant but healthy second at 6.4K. Event, product, and podcast all stay under 1.5K.
Hot takes and reactions also led saves this month with 174, a reversal from August when educational carried intent. The leaked-list reaction alone drove 127. Educational held second at 15 saves on a fraction of the reach. Product and event drove zero.
Five content patterns observed across 19 published posts.
| Pattern | Posts | Total imp. | Avg imp./post | Total saves | Top performer |
|---|---|---|---|---|---|
| Hot takes / reactions | 7 | 325,268 | 46,467 | 174 | "I am speechless" (216.0K imp) |
| Educational / thought leadership | 4 | 25,679 | 6,420 | 15 | Gapless-resume myth (16.2K imp) |
| Podcast (The Talent Bar) | 5 | 3,269 | 654 | 2 | Dana Neiger episode (1.6K imp) |
| Product & company updates | 2 | 1,566 | 783 | 0 | Lavalier feature launch (1.2K imp) |
| Event & appearance promo | 1 | 1,380 | 1,380 | 0 | RecFest Nashville tease (1.4K imp) |
Two posts reacting to real hiring artifacts drove 309K impressions between them, 87% of the month's reach. "I am speechless" (one line over a screenshot) hit 216K on 530 likes and 272 comments, and the leaked "Traits to Avoid" list hit 93K with 127 saves, by far the month's highest. Unlike August's meme (35.6K and only 3 saves), these reactions reached and got bookmarked, because each one broke down a real document the reader wanted to keep.
The two reactions reached a similar order of magnitude, but the leaked-list post earned 127 saves (1.4 per 1K) versus 35 for "I am speechless" (0.16 per 1K). The difference is that the leaked-list post itemized four specific hiring practices the reader could reference, while "I am speechless" was a single line. Outrage buys the reach; the itemized breakdown underneath is what gets it saved.
Below the viral spikes, the reliable performers were educational hot takes that named real companies and cited numbers: job boards won't exist (6.0K, 13 saves, naming Ashby, Juicebox, Paraform with funding figures), recruiters-can-use-LLMs-but-candidates-can't (12.5K, 8 saves), and the gapless-resume myth (16.2K). This is the save engine that shows up every month, not just on a viral day.
Five Talent Bar video clips averaged 654 impressions and 2 saves combined, and the two product posts averaged 783 and drove zero saves. They hold a respectable 2.8 to 4.6% engagement rate from a warm audience, but they do not spread and they almost never get bookmarked. They are a baseline, not a growth lever.
The first sentence does almost all of the work on LinkedIn. The algorithm decides reach within the first 90 minutes based on early dwell and engagement, so the hook is where the post is won or lost. We analyzed the opening lines of Colleen's top 6 posts (top 5 by impressions and top 5 by saves, overlap counted once). Five ingredients recur.
Observed across the top performers this month.
The two biggest posts both opened on a screenshot of a real document: a reaction to something outrageous and a leaked internal "Traits to Avoid" list. "Just look at this" invites the reader to inspect evidence rather than read an opinion. A visible artifact is the single strongest trait of this month's breakouts.
"100 million applications across 200,000 jobs," "$80M at an $850M valuation," "300+ applications," "3x since 2021," "2 to 4 month resume gaps." The educational posts all dropped a hard number in the first few lines, which creates instant credibility and stops the scroll.
Ashby, Juicebox, Paraform, Workday, Textio. The job-boards post anchored its whole argument to named startups and funding rounds, which is exactly why it traveled and got saved. Real names read as reporting, not opinion.
"Recruiters using LLM yes, job seeker using LLM no. Make it make sense." "A gapless resume is one of the most overrated signals in hiring." "I honestly don't think job boards will exist." Naming a common belief and breaking it is the most consistent trait of Colleen's high-engagement posts.
"I had to read this three times." "I am speechless." Fast, personal opening lines with no corporate warmup, followed by a numbered or bulleted breakdown the reader can act on. The posts that opened with a list of takeaways are the ones that got saved.
Setting aside the viral spikes, the floor of the calendar was a context-free meme, a generic rant that also hit a publishing snag, and a hiring-manager lecture whose scheduled version failed to deliver. Hypotheses below.
This was a bare two-line joke with no screenshot, no number, and no breakdown underneath. It makes the same point as the 12.5K "recruiters use LLMs, candidates can't" post, but stripped of the evidence and the argument that made that one travel. A punchline with no setup and no takeaway gives the reader no reason to stop, comment, or bookmark.
The take is fair but generic, with no named company, no number, and no visible artifact to anchor it. It also hit a publishing problem: the scheduled Sep 20 version delivered 1 impression (it never properly went out) and was reposted natively on Sep 21, which cost it the early-momentum window the algorithm rewards. Even published cleanly, a plain opinion rant without evidence is now the weakest format on this feed.
The Sep 14 scheduled version returned zero impressions (a publishing failure) and the post was reposted natively on Sep 15. On top of the lost momentum, it is a long three-point lecture aimed only at hiring managers, a narrower audience than the candidate-facing posts that reached this month, and it opened with a soft "Hot take:" label rather than an artifact or a number.
1. Publishing reliability: two scheduled posts (Sep 14 and Sep 20) failed to deliver, landing 0 and 1 impressions before being reposted by hand and losing the critical first 90 minutes. Verify every scheduled post actually went out.
2. Context-free memes and plain opinion rants (82 and 116 impressions) now badly trail reactions built on a real screenshot with a breakdown.
3. Don't over-read the 216K "I am speechless" post. It ran on 530 likes and 272 comments but only 35 saves (0.16 per 1K). The leaked-list reaction (93K, 127 saves) is the model to copy, because it handed readers an itemized breakdown.
4. Podcast clips (avg 654) and product posts (avg 783) stay small. Keep them as a baseline, but do not let them crowd out the reaction and educational posts that reach and save.
Five concrete moves, each grounded in something we observed this month.
The two reaction posts drove 309K of the month's 357K reach and 162 of its 191 saves. Ship one reaction a week built on a real screenshot of a bad hiring practice (a job post, a rejection email, a leaked policy), and always itemize what is wrong underneath so it gets saved, not just liked. The breakdown is the difference between 127 saves and 35.
Two scheduled posts failed to deliver this month (Sep 14 at 0 impressions, Sep 20 at 1) and had to be reposted natively a day late, forfeiting the early-engagement window. Confirm each scheduled post actually publishes, and if one fails, treat the repost as a fresh slot rather than letting it trail a dead original.
Under the viral spikes, the durable save drivers were educational hot takes citing real companies and figures: job boards (6.0K, 13 saves), recruiters-vs-candidates on AI (12.5K, 8 saves), the gapless-resume myth (16.2K). These reach a wide candidate audience and bookmark reliably, so they should anchor the calendar even in a week with no viral reaction.
"I am speechless" reached 216K but converted only 35 saves because nothing saveable followed it on the same theme. When a reaction post goes big, follow it inside three days with a framework the same reader would bookmark (for example, "the 5 job-post red flags that mean don't apply"). Convert the borrowed attention into intent while it is warm.
Podcast clips averaged 654 impressions and product posts 783, with zero saves between the product posts. Keep one podcast clip and one product post a week as a baseline, but do not let them displace reaction and educational slots. When a launch does run (the Lavalier features video stalled at 1.2K), open it with a hiring problem Colleen has lived rather than a feature list.