“Who called me about free Netflix?” is easy to file as a sales annoyance. If a one-time code from a text message was read back during the same call, the more precise problem is a credential-handling one: an inbound voice path was used as if it were a bound channel for a bearer token.

Affiliation and AI note: This is promotional editorial from the website whocallme.co.uk, drafted with AI assistance. The worksheet below is a manual record design. It is not a deployed tool, a laboratory test, or a claim that the author contacted EE or Netflix.

Stay inside the public note

Our page for 020 3807 0999 is presented as a landline, with an allocation hint of DIDWW Ireland Limited and a numbering area of London. Those labels do not identify the caller or a physical origin for the call. The page shows 6 searches and one user report. A “first seen” label of “4 days ago” is a page label, not the time the comment was written, and not a reason to describe the note as new this week.

The note is imported, and its original publication date is unknown. It carries a visitor category of Scam/Fraud. That category is a visitor label, not an independent finding about the number holder.

The visitor wrote that someone was “Pretending to be from EE”, “said I was getting free promotional netflix”, and that the visitor “gave them a code texted to me from Netflix”. The visitor “got suspicious when they asked me for my address”, “hung up and contacted both EE and Netflix directly who confirmed it was a scam and secured my accounts for me, saying i wouldnt be charged”.

I did not contact EE or Netflix, and I did not reproduce the call. The statements about confirmation, account securing, and the absence of a charge belong to the visitor’s account. They are not official conclusions I can restate as established facts, and they are not billing results I have inspected.

The public note about a promotional-code call is anonymous and unverified. A displayed caller ID can be spoofed. The named companies remain part of the visitor’s description. That is still enough material for an incident-record discussion, because the note already separates three events: a code was disclosed, later contact was made through channels the visitor chose, and the identity of the inbound caller was treated as untrustworthy.

A one-time code behaves like a bearer token

In access-control terms, a one-time password or SMS code is usually a bearer instrument for a short window. The service that issued the code typically cannot see who is sitting next to the handset. Whoever presents the code may complete the action the code was meant to authorise: a login, a recovery step, a subscription change, or another account operation. This article does not claim to know which action, if any, was attempted in the visitor’s case. The engineering point is the property of the code, not a reconstructed attack path.

Reading the code aloud to an inbound caller is functionally close to pasting it into a form you do not control. The voice call feels like a conversation with a company. The code still travels as a secret that confers whatever power the issuer attached to it. Once disclosed, the useful record is that disclosure happened, not a copy of the digits. Do not store the code in a ticket.

A later question for an address, which is what made the visitor suspicious, is a separate request. It may indicate that the caller wanted more data. It does not tell us whether the code had already been used. The record should not invent a successful or failed redemption.

An inbound caller is not a bound channel

A bound channel, in this narrow sense, is a session or address that the account holder can recognise from a starting point they already control: an official app already signed in, a URL they typed from a known domain, a number printed on a bill they already hold. An inbound voice call does not become that bound channel merely because the speaker uses a company name, or because a text message from a familiar brand arrives during the same conversation.

SMS delivery of a code proves, at most, that a message reached a number the issuer had on file, or that a message appeared to come from a familiar sender. It does not bind the person currently speaking on the other line to that issuer. The inbound caller and the SMS are two channels. Mixing them — reading the SMS into the call — gives the unbound voice path the token.

This is why “they already sounded like EE” is not a channel check. The visitor later chose to contact EE and Netflix directly. In a record, that later choice should be written as official-channel contact reported by the visitor, not as proof automatically attached to the original inbound number.

A minimal incident record, filled only from the note

The table below is a worksheet you can copy into a blank document. No telemetry, packet capture, or live test is required. Do not add fields for a password, a full address, or the code itself.

Field Purpose Example drawn only from the public note
source_kind Distinguish a public visitor note from an internal incident Imported public note; original publication date unknown
claimed_organisation Preserve the claimed affiliation without treating it as verified Claimed EE, as written by the visitor
offered_benefit Record the incentive in the visitor’s words Free promotional Netflix, as written
code_disclosed Record whether a one-time code was read out, not the value Yes, a Netflix-texted code, per the visitor
additional_data_requested Record later requests without inventing answers Address asked for; visitor stopped there
inbound_channel_bound Was the inbound voice path a channel started from the account holder’s own records? No, not established
official_channel_contact Record later contact the visitor says they made Visitor says they contacted EE and Netflix directly
issuer_outcome_claimed Keep any “confirmed / secured / not charged” statement attributed to the visitor Visitor-reported only; not independently inspected here
caller_identity_status Separate identity from the other fields Unverified; not established by the page
page_first_seen_label Preserve a page label without using it as the event time “4 days ago”, page label only

The example column is a manual illustration using the published wording. It is not a newly observed incident at your organisation, and it is not a charge record. If your form requires a date, add a review date that is clearly the date you filled the worksheet. Do not copy the page’s first-seen label into an incident-time field.

Keep three decisions apart

The fact that a code was disclosed is one decision point: secrets may have been exposed, so the account holder may want to use an official channel they already trust. The fact that the visitor later contacted EE and Netflix is a second decision point: that contact, if it occurred as written, is about account state, not about proving who the inbound caller was. The identity of the inbound caller is a third decision, and it can remain unknown without blocking the first two.

Do not fold all three into a single state such as “scam confirmed” or “safe because no charge”. The visitor’s category label and the visitor’s report of what the companies said are still unverified material in this article. A reviewer can read them as claims. They should not be promoted into a subscriber-billing conclusion or a legal finding about the number holder.

Human review is the right point to stop enrichment. A reviewer can check that the code itself was not pasted into the ticket, that no address was added “for completeness”, and that company names stay in the claimed_organisation field rather than in a vendor-ID field. If the reviewer was not a party to the call, they should not invent a test call in order to complete the form.

A collection of published UK telephone reports can supply wording for a source pointer when the source is a public page. It cannot bind an inbound caller to a brand, confirm a subscription charge, or replace the official channel the account holder already uses. Leave the caller’s identity unresolved if the evidence does not resolve it. The useful end state of this worksheet is a short, attributed record: a token was disclosed, later official contact was reported, and the inbound speaker remains unverified.