• DevelopingMagnitude 7 earthquake reported in China; GDACS estimates 10,000 in MMI VII+ area — 2 sources
  • New storyCrash on Dixie Highway in Crestview Hills closes I-275 exit ramp
  • DevelopingNWS warns of severe thunderstorm with 60 mph gusts, large hail in parts of Arizona — 3 sources
  • DevelopingNWS warns of severe thunderstorm with 60 mph gusts, large hail in parts of Arizona — 2 sources
  • DevelopingMagnitude 7 earthquake reported in China; GDACS estimates 10,000 in MMI VII+ area — 1 source
  • New storyNWS warns of severe thunderstorm with 60 mph gusts, large hail in parts of Arizona
  • RenderingHurricane Utah flooding forces Gould Wash and WillowWind evacuations
  • New storyOne dead after crash into tree on Carl Baker Road in Starr, coroner says
  • New storyNWS warns of flash flooding in Fresno area through 6:15 p.m. PDT
  • New storyNWS warns of severe thunderstorm near Solomon with 60 mph gusts and quarter-size hail

Tech · Customer data breach

ASOS hackers took customer searches—and the details to make scams convincing

Contact details and shopping searches appeared together in an attacker-supplied sample. ASOS says financial information and passwords were not accessed, but customers still lack a definitive accounting of whose records were taken.

Aired Written 7 min read42 sourcesHeyDay News Desk · About usWatch report

Follow via RSS

Transcript · LOADING

Story trackerDay 3

1 report32 outlets42 sources2 videos

  1. Update 1ASOS hackers took customer searches—and the details to make scams convincingYou are here

A scam message that knows what you have been shopping for is harder to dismiss than one sent at random. That is the practical danger in ASOS’s expanded breach disclosure: hackers obtained customer profiles containing ways to reach people alongside searches they made on the retailer’s website. Payment details and passwords were not accessed, the company says, but those exclusions do not make the other information harmless. 2 3

By 8 October 2026, ASOS had told customers that hackers possessed detailed profiles potentially involving millions of users. The BBC reported that an attacker-supplied sample contained names, addresses, phone numbers, email addresses, customer numbers and website searches. ASOS has not disclosed how many people were affected, how many profiles were taken or how much information was downloaded. “Potentially millions” is possible scale, not a confirmed breach total. 2

Reported searches included “reclaimed vintage,” “glamorous wide fit” and “Asos petite.” Pair an interest with an email address or phone number and fraudulent contact could acquire a convincing shopping connection. That is a risk arising from the exposed information—not evidence that a particular customer has been defrauded or that the records have been publicly released. The search examples come from BBC reporting carried by The Guardian. 2 3

The sample establishes neither the contents of every profile nor the size of the complete collection. ASOS’s own statement identifies names, contact details and “certain non-personal account-related information” as accessed, without fully defining the last category. The complete customer email and dataset were not available for independent examination. Nevertheless, the fuller account materially supersedes the company’s initial warning that basic personal information might have been accessed. 2 3

From rogue message to fuller disclosure

  1. 6 October 2026, approximately 10:00 BST

    Some app users receive an unauthorised notification headed “ASOS HACKED,” threatening a data leak unless they engage.

    1
  2. 6 October, Tuesday afternoon

    ASOS acknowledges the message. It says it restricted access to notification platforms that day; no exact restriction time is established.

    1
  3. 8 October 2026

    The disclosure expands to detailed customer profiles, website searches and access using stolen employee credentials.

    23

Dozens of recipients told the BBC they received the notification, but that is not a total recipient count. SBS reported receipt on phones in Australia, establishing that the message was not confined to the UK. Nor does its arrival time establish when the intrusion or download began. ASOS’s app and website were described as operating normally in the 6–7 October accounts; shopping availability and protection of customer information are separate questions. 1 4

A stolen login, an unresolved route

ASOS says an attacker impersonated a trusted contact to obtain an employee account’s login credentials, then used them to access information on third-party platforms. An unnamed service was used to download customer data. This is an account of credential theft through impersonation—not an established technical vulnerability in the underlying data platform. The services reached, account permissions and access window remain undisclosed. 2 3

The notification claimed, “We have fully compromised the Snowflake instance.” Snowflake said its investigation was ongoing and had so far found no compromise of its platform. Cybercriminals using the name Xuanyewen separately told the BBC they accessed information through Simon AI, described as a platform built on Snowflake. That specific route has not been independently established. 1 2

An ASOS-associated instance, a service built on Snowflake and Snowflake’s underlying platform are not interchangeable. Stolen credentials can open a customer environment without proving that the infrastructure provider itself was breached. The available findings do not settle which environment was accessed. The notification’s extortion-style threat also establishes no verified ransom amount or payment. 1 2

Group-IB research reported by Infosecurity linked the account associated with the Telegram channel xuanyewengateway, which appeared in the notification, to earlier gaming-item trading activity. OSINTsights distinguished the channel’s creation on 6 October from its administrator’s older online history. This is an account-history association, not identification of a real-world attacker—or proof that the channel controller, notification sender and data thief were the same person. 23 39

Cybernews’s 7 October account said researchers had not found leaked files or samples proving possession of ASOS customer records. The 8 October disclosure and BBC examination of a sample advance that earlier evidentiary position. They do not validate the specific claim of full Snowflake compromise. 26 2

What customers can actually know

The latest ASOS statements say payment-card information, bank details and account passwords were not accessed. That is firmer than the initial assessment that cards and passwords were not believed impacted. ASOS told customers they need not take action, while warning them about unsolicited contact. Cybersecurity expert Trevor Dearing separately recommended changing passwords as a precaution and watching for suspicious activity; that recommendation is not evidence that ASOS passwords were stolen. 1 2 3

BBC advice published on 6 October told customers to disregard the unauthorised push notification and not engage with it; the message contained an external link. Skint Dad subsequently reported NCSC advice to assume exposure even without receiving the notification. The exact wording was not visible in the available first-party NCSC passage, and precautionary guidance is not a finding that every customer’s information was accessed. 21 14 9

Regulatory engagement has moved forward, but its formal status remains unclear. The BBC’s 6 October report said ASOS had not informed the Information Commissioner’s Office as of that account. On 7 October, the ICO publicly acknowledged customers’ concerns and encouraged caution with links and attachments. By 8 October, ASOS said it was working with law enforcement and regulators. Neither the available ICO statement nor that company description establishes when a formal breach report was received; there is no supported finding here that ASOS missed a reporting deadline. 1 6 3

The retailer’s reach makes the missing count consequential, but cannot fill it in. The initial report cited more than 10 million Android app downloads, around 17 million annual customers and more than 150 markets. Downloads are cumulative installations, not exposed people; annual customers and markets are business-scale measures, not breach totals. 1

Investors reacted sharply, although the reported percentages remain unsettled. BBC News on 6 October described a fall of around a tenth that Tuesday. FRANCE 24 English that evening reported a 12% intraday maximum and a 7% closing loss. SBS News on 7 October reported a 10% overnight decline. The BBC and SBS descriptions do not specify exact observation times or comparison prices; intraday and closing moves are different measures. 1 5 4

The FT historical-price record gives a 6 October close of 454.00p and daily low of 426.50p. Against Daily Political’s reported previous close of 502.00p, those imply declines of 9.56% and 15.04%, respectively—different from FRANCE 24’s figures. The calculations remain conditional because the available FT excerpt did not expose the prior-day row. Shattered.io’s separate intraday account gave up to 13.2% and 439p, above the recorded full-day minimum. No established evidence attributes the whole day’s movement exclusively to the breach. 33 31 5 29

Known

  • Customer information was obtained, and ASOS attributes access to stolen employee credentials. 23

Unknown

  • The final affected-person count, access window, complete service list and confirmed ICO notification date remain undisclosed or unestablished.

Next

  • A completed accounting of whose records were taken and how the stolen login reached them. No dated next announcement or investigation deadline is established.

The central question is no longer whether this was merely a disturbance in a notification system. Customer profiles were taken. The missing answer is whose—and customers still lack a definitive accounting of their individual exposure.

Sources used for this report

BBC News supplied the initial notification account 1, expanded profile reporting 2, its RSS version 7, the BBC.com notification account 24, live coverage 12 25 and customer advice 21. The Guardian supplied the customer statement in its business live coverage 3. SBS News 4 and FRANCE 24 English 5 supplied broadcast accounts.

Official material comprised the ICO’s ASOS statement 6, ASOS’s announcement carried by Investegate 8, and the NCSC incident page 9 and PDF 10. Other incident coverage came from BleepingComputer 11, Drapers 13, Irish Independent 15, Retail Insight Network 18, SecurityWeek 19 and Daily Mail 20.

Customer guidance and further coverage included Skint Dad 14, ALLATRA Media 17, and The Independent’s protection guide 16, expanded breach account 22, credential-theft report 37 and notification explainer 41. Cybernews supplied its earlier evidence assessment 26.

Account-history reporting came from Infosecurity’s Telegram investigation 23, OSINTsights 39 and Tech Debrief 40. Additional notification coverage came from Infosecurity 38, Forbes 32 and The Paypers 42.

Market sources were the FT historical-price table 33, Daily Political 31, Shattered.io 29, Global Banking & Finance Review 27, Sharecast 28, AskTraders 30, FashionNetwork 34, Investors Chronicle’s historical-price page 35 and The Sunday Guardian 36.

As aired 29 lines
  1. Hackers obtained ASOS customer profiles containing contact details and website searches. The possible scale reaches millions, but no affected-person count is confirmed. Combining shopping interests with ways to contact someone could make fraudulent messages more convincing. How much information was taken, and what can customers actually know about their exposure?
  2. Searches included “reclaimed vintage,” “glamorous wide fit” and “Asos petite.” Those interests could give an unsolicited message a plausible shopping connection. That is a fraud risk, not evidence that a particular customer has been defrauded or that the records have been publicly released.
  3. The detailed records include names, addresses, phone numbers, email addresses, customer numbers and website searches. “Potentially millions” describes possible scale, not a verified total. Neither the number of people affected, the number of profiles taken nor the download volume has been disclosed.
  4. ASOS's disclosure has changed materially. On 6 October, basic personal information, including names and contact details, might have been accessed. By 8 October, detailed profiles were part of the disclosure. The examined sample does not establish what every profile contains or the contents of the complete collection.
  5. The account-related category still lacks a full definition. Names and contact details were accessed on third-party platforms, alongside information described as non-personal and account-related. That description cannot erase the fuller profile exposure. The complete customer email and dataset were not available for independent examination.
  6. An attacker impersonated a trusted contact, obtained an employee account's login credentials, then used that access to reach customer information on third-party platforms. An unnamed service enabled the download. This describes stolen-account access, not an established technical vulnerability in the underlying data platform.
  7. The first access time, download time and duration remain undisclosed. So do the specific services reached and the employee account's permissions. This is an interim explanation, not a completed public forensic reconstruction. Restricting one notification service cannot, by itself, establish the extent of access elsewhere.
  8. The rogue notification reached phones in the United Kingdom and Australia. It arrived on 6 October at approximately 10:00 British Summer Time, or 09:00 UTC. That timestamps the customer-facing message, not the start of data access, and these locations do not define the breach's full geography.
  9. The notification threatened a data leak unless recipients engaged with the attackers, and included an external link. Its claim of a fully compromised Snowflake instance remains unverified. There is no established ransom amount or evidence that a customer or ASOS paid the attackers.
  10. Dozens of recipients reported receiving the message; that is not a total recipient count. ASOS acknowledged it on Tuesday afternoon and restricted access to its notification platforms that day. The exact acknowledgement and restriction times are unknown.
  11. The app and website were operating normally in the 6–7 October accounts. Continued shopping availability does not undo information already obtained. Customers were advised to disregard the unauthorised notification and not engage with it, rather than follow the link or respond to its threat.
  12. Snowflake's ongoing investigation had found no compromise of its platform so far. The name Xuanyewen is attached to a separate, unverified claim of access through Simon AI, described as built on Snowflake. The precise environment accessed remains unresolved.
  13. A customer-associated instance, a service built on Snowflake and Snowflake's underlying platform are different things. Stolen credentials can expose a customer environment without establishing a breach of the infrastructure provider. The employee-credential explanation does not independently validate the attackers' specific platform claim.
  14. The Telegram channel xuanyewengateway appeared in the notification and was created on 6 October. Its administering account had an earlier online history linked to gaming-item trading. That association suggests an identity reorganized around the incident; it does not identify a real-world attacker.
  15. The forum name, matching profile or identifier and a verified connection to the intrusion remain unavailable. It is also unproven that the channel controller, notification sender and data thief were the same person. A 7 October finding of no possession-proving samples preceded the 8 October profile disclosure and sample examination; the evidence advanced, without proving full Snowflake compromise.
  16. Payment-card information, bank details and account passwords were not accessed in the latest company assessment. That is stronger than the initial wording that cards and passwords were not believed impacted. It does not make exposed contact details or search history harmless.
  17. Customers were told they did not need to take action, while also being warned about unsolicited contact. A separate precautionary recommendation was to change passwords and watch for suspicious activity. That recommendation is not evidence that ASOS passwords were stolen.
  18. The precautionary warning's verbatim wording and the expert's organisational affiliation are not established. The practical distinction is between a confirmed password exposure—which is not identified here—and taking an extra precaution while the scope of other exposed information remains uncertain.
  19. The UK data regulator publicly addressed the incident on 7 October and encouraged caution with links and attachments. That establishes awareness, not receipt of a formal company breach notification. On 6 October, ASOS had not informed the regulator as of the account published that day.
  20. By 8 October, ASOS was working with law enforcement and regulators. Whether and when it formally notified the Information Commissioner's Office remains unclear. No confirmed receipt date or case-specific reporting deadline is established, so there is no supported finding that ASOS missed a deadline.
  21. Not receiving the push notification does not settle whether information was accessed. Precautionary advice to assume exposure has been described second-hand, but its exact first-party wording is unavailable. It is not a finding that every ASOS customer's information was taken.
  22. The National Cyber Security Centre's incident page repeated the initial assessment about names, contact details, cards and passwords. Assistance to ASOS was also described, but the available official passage does not establish what was offered or delivered. That gap does not prove assistance or advice was absent.
  23. The Android app had more than 10 million Google Play downloads; ASOS served around 17 million customers annually across more than 150 markets. Downloads are cumulative installations, not active users or notification recipients. None of these figures supplies an exposed-profile count or a verified breach rate.
  24. ASOS shares fell on 6 October, but the reported measures differ: around 10% that Tuesday; a 12% intraday maximum and 7% closing loss; and 10% overnight. The around-10% and overnight descriptions do not establish exact observation times or comparison prices.
  25. The full-day price record gives a 454.00-pence close and a 426.50-pence low. A separately reported previous close is 502.00 pence. The historical-table excerpt did not include the prior-day row, leaving that denominator without a second checked record.
  26. Using 502.00 pence, the close calculates to a 9.56% decline and the daily low to 15.04%. Another intraday account gave up to 13.2% and 439 pence—higher than the recorded full-day minimum. These remain distinct accounts, not one settled percentage.
  27. The conditional closing calculation fits “around 10%,” but differs from the reported 7% close. The conditional low-based calculation exceeds the reported 12% maximum. No established evidence assigns the whole day's share movement exclusively to the breach.
  28. This was more than a notification-system disturbance: customer profiles were obtained. But the total, access window and full service list remain undisclosed. There is no completed public forensic report, law-enforcement outcome, confirmed regulator-notification date or dated next announcement or investigation deadline.
  29. The answer is that identifying details and website searches are part of the exposure, but its full scale and individual reach remain unknown. Customers still lack a definitive accounting of which records were taken, while names, ways to contact them and shopping searches may now sit together outside the company's control.

Sources

42 sources32 outlets2 videos

Source map

Follow the evidence behind this report. The map shows its connections; the links below open the sources cited in this article.

Source map loads as you read.

Explore the full newsroom map ↗

Revision log

  1. r1First published.