How it works

How Cut Through Crypto scores the news

Our method, in plain terms — so you can trust the filter. Crypto produces thousands of headlines a day. Most are noise, many are paid, and the few that matter get buried by the ones that don't. Cut Through Crypto surfaces the small number of stories that actually move a long-term investment thesis — and says why. Here's exactly how, with nothing hidden.

Scored against a thesis, not a mood

Most feeds rank by engagement — what's loud, trending, or being shilled. We don't. Every story is scored 1–10 for how directly it touches a structural investment thesis: the specific conditions that would confirm, threaten, or erode the long-term case for crypto. A quiet technical development can outrank a viral price headline — because it moves the thesis and the price headline doesn't.

And we don't score for agreement. A story that contradicts the thesis — one we'd rather weren't true — scores just as high as one that confirms it. Contrarian signals are the most valuable kind. The filter is for relevance, not comfort.

What the score means

9–10 · Critical — a direct hit on a make-or-break condition: a Bitcoin security break, a CBDC displacing stablecoins, privacy-coin delistings going universal, confirmed sovereign Bitcoin buying. Things that genuinely change the picture.
7–8 · High — concrete, trackable data on the trends we watch: ETF flows, key legislation, on-chain metrics, tokenisation milestones. Movement you can measure.
4–6 · Moderate — relevant to a theme we cover, but no metric has moved yet. Context, not catalyst.
1–3 · Noise — price-pump fluff, SEO filler, off-topic macro. Anything scoring 3 or below is hidden. You never see it.

Forty stories become one

The same event gets reported dozens of times across dozens of outlets. That duplication collapses into one clean, citable story — with a link to every source. You read the development once, not forty times, and can always check the originals.

Threaded over time

News isn't standalone; it's a storyline. Each story is tracked across days, so you see at a glance what's new, what's progressed, and what's gone quiet. A regulatory fight or an accumulation trend reads as one through-line, not a scatter of headlines.

The engine proposes; I dispose

The AI does the reading, scoring, de-duplicating and threading — at a scale no person could. But it doesn't get the last word. Every major story gets a sharp human take: the so-what, not the summary. The machine surfaces the signal; a person decides what it means.

Verified, or not — who stands behind a story

The steps above are the machine's work. A separate human accuracy pass is what puts a person's name behind a specific story: each claim fact-checked against the primary sources, headline figures traced to what the source actually says, and assertions attributed — never a source's hedge or projection restated as settled fact.

So every story carries one of two labels. “✓ Verified”, with the date it was checked, means a person has done that pass and stands behind it — and because verification is a snapshot, editing the story afterwards clears the tick. “Unverified — auto-generated summary, not yet reviewed” means exactly that: the machine's draft, not yet checked, its specifics not to be relied on until they are. Most of the archive predates this pass and reads unverified until each entry is reviewed one by one — the honest state, not fine-print to skim past.

Six sectors

Every story is filed into one sector, so you can follow what you care about:

—

What this is, and isn't

A daily, thesis-driven filter for a long-horizon investor: one clear point of view, applied consistently, with the work shown. It is not financial advice, and it isn't neutral — it's a specific thesis, transparently applied, so you can weigh the signal for yourself.