Bitcoin ETFs eke out a third weekly inflow — after $465 million walked out the door on Thursday and Friday

The U.S.-listed spot bitcoin ETFs closed the week ended July 24 in the black, but only just, and only because the damage came late. The funds netted $33.79 million across the five trading days, per data from analytics firm SoSoValue cited by CoinDesk — their third consecutive week of inflows and the first such streak since early May. Strip out the last 48 hours and the number would have been far larger. Add them back in and the streak nearly didn't survive.

What happened, in sequence

The week started strong enough that the funds were riding a seven-day winning streak into Thursday. Bitcoin itself rallied to a July high above $66,500 on Tuesday, according to CoinDesk. Then the flows reversed. The ETFs shed roughly $225.2 million on Wednesday, July 23, and another $240.1 million on Thursday, July 24 — about $465 million over the two sessions combined, per SoSoValue's tally as reported by CoinDesk. Decrypt put the same two-day figure at "$465M," corroborating the number to the dollar range, and both outlets agree on where it came from: BlackRock's IBIT, the largest of the spot funds, accounted for nearly $415 million of the exits — the vast majority of the total.

Bitcoin's price tracked the mood, retreating below $64,000 by the end of the week amid what CoinDesk described as profit-taking and a soft equity tape, with the Nasdaq 100 dragged down by chipmaker stocks. At the time the two pieces published on July 27, bitcoin was quoted at $65,495.22 by CoinDesk and $65,538.00 by Decrypt — a small gap consistent with snapshots taken minutes apart.

The context that matters

The $33.79 million net figure is best read against what came before it. The ETFs had bled for eight consecutive weeks starting the week ended May 15, per CoinDesk. July has been the repair phase: three straight inflow weeks of $197 million, $75.67 million, and now $33.79 million. That is a positive trajectory, but a decelerating one — each week smaller than the last, and this latest one the smallest of the three. Roughly $34 million spread across five days of trading, for a product class that measures its biggest days in the hundreds of millions, is a trickle, not a flood.

Key facts

• Weekly net flow: +$33.79 million for the week ended July 24, the third straight weekly inflow and the first three-week streak since early May (SoSoValue data, via CoinDesk).
• Late-week outflows: ~$225.2 million on July 23 and ~$240.1 million on July 24; ~$465 million combined (SoSoValue via CoinDesk; "$465M" per Decrypt).
• Lead exit: BlackRock's IBIT, ~$415 million of the two-day outflow (CoinDesk and Decrypt).
• Prior two inflow weeks: +$197 million and +$75.67 million (CoinDesk).
• Preceding stretch: eight consecutive weeks of outflows from the week ended May 15 (CoinDesk).
• Price: July high above $66,500 on July 21, then below $64,000 by week's end; quoted at $65,495.22 (CoinDesk) and $65,538.00 (Decrypt) on July 27.

The real-world read

The honest framing here is one CoinDesk itself reached, and it's worth keeping: institutional demand has come back, but it's "anemic and not as powerful as typically observed during bull runs." A headline that reads "third straight week of inflows" is technically true and slightly flattering. A more complete headline is that the week ended with a two-day rush for the exits roughly fourteen times the size of the net weekly gain, and the streak held only because the earlier days banked enough to absorb it.

Watch the concentration. Nearly $415 million of a $465 million outflow ran through a single product, IBIT. When one fund dominates both the inflows and the outflows this way, "ETF flows" as a category number can mask what is really a small number of large allocators moving in and out of one vehicle. A $33.79 million weekly "inflow" that hinges on the timing of BlackRock's book is a fragile thing to build a narrative on — one more day like Thursday and the third-week streak would have snapped.

Note also what's doing the explaining. The tidy "repair phase" framing comes from BRN, a crypto analytics firm, in an email to CoinDesk — a characterization, not a fact, and one from a firm whose business is commentary on these markets. Their own words hedge it: "institutional demand is still cautious." The attribution of the price pullback to profit-taking and chipmaker weakness is CoinDesk's read, plausible but not the kind of thing that can be proven from a flow table.

One structural caveat: the underlying flow numbers here trace to SoSoValue, a third-party aggregator, not to the issuers' own daily disclosures. The two secondary outlets agree on the two-day total and on IBIT's lead, which is reassuring, but the primary record is each fund's official creation/redemption data, and neither piece cites it directly. Treat the precise per-day splits as SoSoValue's estimates until the issuers' numbers confirm them.

Opinion, and whose

• BRN (crypto analytics firm), via CoinDesk: "After May and June's heavy outflows, July's repair phase has brought relief, but institutional demand is still cautious." A characterization of the trend, not a forecast — and self-hedged.
• CoinDesk: framed returning institutional demand as "anemic and not as powerful as typically observed during bull runs," and attributed the price pullback to profit-taking and a weak Nasdaq 100. Reasonable interpretation, presented as such.

No source in this set forecasts a price target, and neither should be read as doing so.

Sources

• CoinDesk (Tier 2, reputable secondary), July 27 — the anchor for this story: the weekly net figure ($33.79M), the per-day outflows ($225.2M and $240.1M), the IBIT concentration (~$415M), the three-week streak, the eight prior outflow weeks, the price action, and the BRN quote. Flow data attributed to SoSoValue; the "repair phase" characterization attributed to analytics firm BRN.
• Decrypt (Tier 2, reputable secondary), July 27 — corroborated the $465M two-day total and that IBIT led the outflows; also supplied a July 27 price quote ($65,538.00). Most of the page was a live price ticker rather than reporting; its substantive contribution was confirmation of the headline figures.
• No Tier 1 primary source (issuer disclosures or SoSoValue's raw dataset) was cited directly by either outlet; the flow figures originate with SoSoValue.

This is news coverage, not investment advice.