Sample audit · public posts as of 1 Oct 2026
We took nineteen of the most repeated LinkedIn rules and checked each one against what actually happened on his feed. Every post is compared only with his other posts from the same 30 days, so his audience growing or shrinking cancels out.
How to read the numbers. "+48%" means posts that follow the rule typically get 48% more reactions than his other posts from the same month. "Not enough posts to tell" means he has not done it often enough for a real difference to show; we say how big the difference would have needed to be.
He is one of the founders most often held up as proof that LinkedIn alone can build a company. RB2B went from zero to $5M in annual revenue in 12 months with his LinkedIn as its only marketing channel, and he teaches his approach openly, so his habits get copied. He has also posted almost every day for three years, which gives enough history to test nineteen rules properly; most accounts do not have that. If the standard rules hold anywhere, they should hold here.
We have no connection to him. We chose his account because it is public, prolific and influential. Sources: SaaStock, Founderpath.
| Result | Posts | What it means | |
|---|---|---|---|
| Put a specific number in the post | +48% | 471 | His strongest lever. A dollar figure, a count or a percentage. Several other habits (longer posts, lots of line breaks) look good mainly because those posts also carry a number. |
| Don't write short | −30% | 237 | His shortest posts get 30% fewer reactions. Even short posts with a number still lose 24%. The usual "keep it short" advice is wrong for him. |
| Attach an image | +21% | 212 | Smaller than the first two, and the easiest to act on. |
Also true for him: posts that open with "I", "my" or "we" get 23% more (160 posts). A link in the post costs about 15% (338 posts). That matches the usual advice, though it is a cost, not the reach-killer people describe.
| Verdict | In plain words | |
|---|---|---|
| Asking questions to get comments | Makes no real difference | Posts with a question get 9% fewer reactions across 456 posts. The difference is real but too small to matter. |
| Tagging people | Makes no real difference | 6% lower across 466 posts. Tagging is not bringing him extra reach. |
| Lists and bullet points | A few hits, not a pattern | A handful of list posts did very well and pull the average up. Most of his 421 list posts did no better than his other posts that month. |
| Lots of white space | Luck and numbers | It looks like +24%, but that disappears once you compare only posts that both have a number. With 19 rules tested, one or two will look good by chance, and this is one of them. |
| Hashtags | Not enough posts to tell | Only 72 posts used them, and they show no difference. With that few, only a change of 30% or more would have shown up. Anything smaller is invisible, so hashtags are not worth his attention either way. |
| Ending with a question | Not enough posts to tell | Only 42 posts. A difference would have needed to be 50% or more to show up. |
| Carousels and all-caps hooks | Not enough posts to tell | 8 and 7 posts. He barely uses them. |
| Posting in the morning | Not enough posts to tell | 2 morning posts in three years. No data. |
Never tried, so nothing to test: video.
Both point in a promising direction, but his history alone cannot settle them. A test can.
| So far | Posts | What we would expect from a test | |
|---|---|---|---|
| Opening with a story | +17% | 135 | Our prediction: story openings lift his reactions by around 15 to 20%. Two of his three best posts ever were stories about a person. A test on his next 50 or so posts would tell whether that lift is real and, with his own impressions, whether stories also reach more people. |
| Posting at the weekend | +16% | 165 | Our prediction: weekend posts do no worse, and possibly better. That would contradict "never post at the weekend" and give him two more posting days. |
Median reactions per post, by half year. *H2 2026 runs to 30 September. His typical post is down 59% from its peak in early 2024, while his following kept growing to 159k. The comparisons above already remove this trend. Whether the drop comes from him, his topics or LinkedIn itself cannot be told from one account.
Around November 2024, people started resharing his posts less: reposts per reaction fell by 37%. That is a change in behaviour, not in audience size. Whether it was him or LinkedIn would show up in the same check on other accounts over the same dates.
It cannot prove cause. If he adds images to the posts he already thinks are strongest, the image gets the credit. Looking back at past posts can never separate those two. The only way is a test on his own account: a coin flip decides which upcoming posts carry the change, then we measure.
It measures reactions and comments, not impressions. Impressions are visible only to the account owner, so a public audit shows what people responded to, not how many people saw it.
Appendix
They are not Adam's rules. They are the pieces of advice that come up again and again in LinkedIn growth content: creator playbooks, ghostwriter threads, "how I grew to 100k" posts. Things like "keep it short", "never put a link in the post", "use three to five hashtags" and "post in the morning". We chose them because they are the rules people actually follow, so they are the ones worth checking. In a paid audit you can add your own: the rules your ghostwriter or your playbook swears by.
Each post compared with the twenty posts nearest it in time, so an early post is not buried by a later, bigger audience. Reposts of other people are excluded.
"44×" means 44 times the reactions of his typical post at that time. Eight of the ten carry a specific number. Two of the top three are stories about a person, which is why story openings are worth testing.
Result = how posts following the rule did against his other posts from the same month. "Smallest visible change" = how big a difference had to be for this many posts to show it reliably.
| Rule | Result | Posts | Smallest visible change | Verdict |
|---|---|---|---|---|
| Specific numbers beat vague claims | +48% | 471 | 15% | Works |
| Keep it short | −30% | 237 | 23% | The opposite is true |
| Long posts build authority | +37% | 236 | 21% | Works, mostly via numbers |
| Never put a link in the post | −15% | 338 | 15% | True, a modest cost |
| Always add an image | +21% | 212 | 20% | Works |
| Write in the first person | +23% | 160 | 26% | Works |
| Ask questions to drive comments | −9% | 456 | 13% | No real difference |
| Tag people to borrow their audience | −6% | 466 | 15% | No real difference |
| Lots of white space | +24% | 238 | 20% | Luck and numbers |
| Listicles beat prose | +21% | 421 | 18% | A few hits, not a pattern |
| Never post at the weekend | +16% | 165 | 27% | Worth testing |
| Open with a story | +17% | 135 | 27% | Worth testing |
| Hashtags extend reach | −3% | 72 | 31% | Not enough posts |
| Always end with a question | −9% | 42 | 52% | Not enough posts |
| Use three to five hashtags | +6% | 19 | 76% | Not enough posts |
| Carousels get the most reach | −17% | 8 | 5× | Not enough posts |
| Shout the hook in capitals | −24% | 7 | 3× | Not enough posts |
| Post in the morning | n/a | 2 | n/a | Not enough posts |
| Video wins everywhere | n/a | 0 | n/a | Never tried |
Numbers, long posts and lots of line breaks tend to show up in the same posts, so their effects overlap rather than add up. We re-checked each one using only posts that already had a number:
| Among posts with a number | Result | Posts | Holds on its own? |
|---|---|---|---|
| Short | −24% | 137 | Yes |
| Long | +23% | 218 | Not clearly |
| Lots of line breaks | +13% | 227 | No |
Short posts lose even with a number, so "don't write short" stays. Long posts and line breaks were mostly riding on the number.
Independent analysis of public posts. Not affiliated with or endorsed by Adam Robinson, RB2B or MoltSets.
Your last 300 to 800 posts, the same nineteen rules plus up to five of your own, the full report, and a test plan with our predictions. PDF within 48 hours. Send your LinkedIn profile URL at checkout.
The audit tells you what goes with better results. To know what actually causes them, the next step is a test on your own account: one change at a time, a coin flip per post, our prediction written down before we start, and your own impressions data. From $1,500 a month, three months minimum. Ask after your audit.