This site's own company graph records real venture funding rounds, sourced from newsletters, RSS feeds, and Google Alerts. Auditing it this week turned up Anthropic's $65 billion round logged as four separate rows -- same company, same amount, four different dates a few weeks apart, each one a different outlet's mention of the same event.[1] The instinct is to call that a duplicate and delete three of the four rows. That's half right: for "how much money did Anthropic actually raise," there's exactly one real answer, and the other three rows were noise. But how many independent times an event gets reported is itself real information -- not about the event, but about how much attention it pulled. Collapsing the four rows into one and throwing away the count would have deleted that second measurement along with the duplicate data.
What this site's own numbers show
Of 4,382 real funding rounds on record, 4,237 were mentioned exactly once -- the overwhelming norm. 145 were mentioned more than once, up to a high of 8. What's notable is how loosely that repetition tracks deal size: CuspAI's $450 million round was mentioned 8 separate times, the same count as Atoms' $1.7 billion round -- nearly 4x the money, the same amount of attention. Blue Origin's $10 billion round -- one of the largest on record -- was mentioned only 5 times. Bigger checks get slightly more repeat coverage on average, but the relationship is weak, not proportional: attention is being driven by something beyond the size of the number.
[1]
That gap -- between what happened and how much got said about it -- isn't unique to funding rounds, and it doesn't work the same way twice. Two other well-studied domains show the identical surface pattern with two genuinely different underlying mechanisms.
When attention leads the event, not just reports it
Harvard Kennedy School's Shorenstein Center analyzed thousands of 2016 presidential campaign news statements across eight major outlets during "the invisible primary" -- the year before any actual voting. Donald Trump received roughly a third of all coverage in 2015, nearly twice the volume of the next-most-covered Republican candidate, despite low initial polling numbers. The report's finding: the volume and tone of that coverage
preceded his rise in the polls, not the reverse.
[2] A separate peer-reviewed study (Reuning, 2019,
Perspectives on Politics) found that most 2016 candidates did NOT see their polling improve after a surge in coverage -- Trump was the exception.
[3] In this domain, attention isn't just a delayed echo of what already happened -- for at least one candidate, it appears to have helped cause the outcome it was reporting on.
When attention is scheduled before anyone knows the outcome
A company preparing to go public enters a legally defined "quiet period" between filing its S-1 and the actual offering -- a real, dated process milestone, not a spontaneous news event. Bushee et al. (
Journal of Accounting and Economics, 2020, cited 157 times) found that media coverage clusters heavily in exactly this window, and that it measurably shapes retail investor trading behavior before the company has even priced its shares.
[4] This is a third, distinct mechanism from the other two: not a mega-deal drawing repeat mentions after the fact (funding rounds), and not coverage volume itself moving an uncertain outcome (primaries) -- a predictable, structural buildup tied to a known calendar date, whether or not the eventual IPO turns out to matter.
4,237 / 4,382
funding rounds on this site's graph mentioned exactly once (97%)
~33%
of all 2015 GOP primary coverage that went to one candidate (Shorenstein Center)
157
citations on the peer-reviewed IPO quiet-period coverage study (Bushee et al., 2020)
Three domains, three different reasons attention and event size come apart -- a mega-deal draws repeat mentions after it's already real; a candidate's coverage volume can shape the very outcome it's supposedly just reporting; a company's coverage clusters around a scheduled date regardless of what that date turns out to mean. Treating all three as "the media hypes things" erases what's actually different about each one -- and that's the same collapsing-distinct-things-into-one-number mistake this site has already named elsewhere: two real measurements, not one blurry one.