Before You Build an Ethics Hotline: Why Zero Reports Is Bad News
In most Turkish hotel groups, the ethics hotline is an email address. It starts with ethics@ or etik@, it sits on an intranet page and on a back-of-house notice board, and it lands in the HR director's mailbox — sometimes the GM's assistant's too. It was agreed in one meeting and has not been reopened since.
This piece is not here to mock that address. It is here to take it seriously enough to test it. A reporting channel has to meet four conditions, and a shared mailbox meets none of them. Not through neglect — by construction.
Then we look at the two Turkish instruments that have written those conditions down. Both are real, both can be read in full, and both almost certainly do not bind you. I will explain why they are still the benchmark. The piece ends with five questions rather than a recommendation.
Every hotel group has an ethics@ address, and almost none of them receive anything
That address rarely comes from a need. It comes from a question. A lender asks for the compliance file, a brand partner sends its own policy set, an audit report leaves the "internal reporting channel" line blank. The following week the address is created, a paragraph goes into the policy document, and the line is filled.
Then years pass. Ask a general manager how many reports came in last season and the answer is almost always the same: none, thankfully. The conversation usually ends there.
Meanwhile the same group's exit interviews are full of content, turnover concentrates in one department, and friction at shift level is common knowledge. The operation is not silent. The mailbox is.
Put those two observations side by side and a question falls out. This whole piece is built on it: what is zero reports actually measuring?
Zero reports is not a result, it is a measurement fault
Zero reports looks identical in two very different worlds. In the first, there genuinely was nothing to report. In the second there was, and it never reached the channel — because nobody knew about it, nobody trusted it, or reaching it was itself the risk. From outside the channel, the two worlds are indistinguishable.
Measurement has a name for this: no data and not measured are not the same thing. An empty cell does not mean the indicator is zero; more often it means the instrument was not running that period. A report that treats the two as interchangeable tells you nothing at best.
The same fault is familiar from employee surveys. When participation drops, the resulting scores are easy to read as improvement — but a mechanism employees have stopped answering is a closed channel, not a good result. In a reporting mailbox the only difference is that you never see the participation rate at all.
The practical consequence: stop reporting zero as a performance figure. That number describes the channel, not the business — and on its own it does not even do that reliably.
So what do you measure instead? Not report volume. The channel's own properties: access, record, timing and impartiality. All four can be audited with no report ever arriving, none of them depends on anyone's good intentions, and together they decide whether what you have is a channel at all.
The four conditions a channel has to meet
Whether a structure counts as an employee reporting channel is not settled by the label on it. It is settled by how it answers four questions.
| Condition | The question | What a shared mailbox gives you |
|---|---|---|
| Access | Who can see the report? | Everyone the mail system grants access to |
| Record | Who did what, and when? | The mailbox itself — editable by the person reading it |
| Timing | Does a clock start? | No; a read receipt is not a commitment |
| Impartiality | Can the recipient be the subject? | Yes, and eventually is |
All four are system properties. None survives the sentence "we trust whoever reads it," because all four exist precisely for the situations where that sentence stops holding.
Before deciding whether the conditions matter to you, price the alternative. People say most of it on the way out — and a good share of what you hear in an exit interview was still fixable six months earlier. Information that arrives at the exit is not merely late information; it is information that can no longer produce a decision. That gap is the financial case for a channel.
Now take the four conditions one at a time.
First: preventing unauthorised access by construction, not by policy
The most detailed text anyone has written on what an internal reporting channel has to do is Directive (EU) 2019/1937. Say the obvious part first: this is EU law, applied through transposition into member states' national law. It does not bind a privately held hotel group in Türkiye. We are reading it as a reference standard, not as an obligation — because nothing else describes internal reporting channel requirements at this level of detail.
Article 9(1)(a) defines the channel itself:
"channels for receiving the reports which are designed, established and operated in a secure manner that ensures that the confidentiality of the identity of the reporting person and any third party mentioned in the report is protected, and prevents access thereto by non-authorised staff members"
Two phrases carry the weight. The first is "prevents access thereto by non-authorised staff members" — not as a policy undertaking, but at the level of how the channel is designed, built and run.
A shared mailbox cannot satisfy that, because whoever opens the mailbox does not control its access list. The list typically includes the mail administrator, the IT team running the infrastructure, the backup and archive layer, anyone holding delegate access, the assistant who covers annual leave, everyone a distribution rule was extended to once and never trimmed back, and any leaver whose access was never revoked. Nobody at the hotel decided that list. The mail system did.
The second phrase is the one almost nobody designs for: "and any third party mentioned in the report". The confidentiality of the person named in the report is in scope too — the person being reported on. In a mailbox read by ten people, the reporter's confidentiality and the accused person's confidentiality are lost in the same instant. The second loss rarely gets discussed at all.
A control you can run this week: pull the access list from the mail administration console, not from the policy document. The answer to "who has access" lives on a screen, not in a procedure. Where the two disagree, the screen is the one that is true.
Second: the record itself — who did what, and when
A record is a separate thing from the report. In a shared mailbox they are the same object, and until they are separated, four things follow.
Deleting a message deletes the record. Read status attaches to a person rather than to the organisation. Where the message was forwarded, who replied, and how long it sat are all invisible. And there is no state — a mailbox does not carry open, under review or closed; what looks like state is a folder name.
All four are uncomfortable. The problem sitting underneath them is the real one: the record can be altered by the person the record is about. The minimum condition for an instrument to be measuring the thing it claims to measure is that the party being measured cannot quietly change the instrument. Where the recipient of a report and the owner of the record are the same person, that instrument is measuring nothing.
There is a short test. If the answer to "how many reports came in last year, how many were closed, and how long did closing take on average" comes out of somebody's memory, there is no record. If it comes off a screen, the next question is who can edit that screen.
Third: a clock has to start
An unacknowledged report is not a report. It is a message that was sent. To the person who sent it, that difference is everything — they are reading the same silence for three signals at once: did it arrive, is anyone reading it, and is anything about to happen to me.
The reference standard sets two separate clocks here. Article 9(1)(b):
"acknowledgment of receipt of the report to the reporting person within seven days of that receipt"
And Article 9(1)(f):
"a reasonable timeframe to provide feedback, not exceeding three months from the acknowledgment of receipt or, if no acknowledgement was sent to the reporting person, three months from the expiry of the seven-day period after the report was made"
The detail here gets lost constantly, so let us keep it explicit: the three months run from the acknowledgement, not from the report. Where no acknowledgement was sent, they run from the end of the seven-day window. The usual shorthand — "seven days to acknowledge, three months to respond" — erases that distinction. The same scope caveat applies: this is EU law, it does not bind you, and we are using it as a design measure.
In a shared mailbox the clock never starts at all. A read receipt is not an acknowledgement: it says a person opened a message, not that the organisation has accepted a report. Nor is an auto-reply — that tells the sender the mailbox is alive, not that anything entered a process.
The control question does not require waiting for a report to arrive: is the time to first response written down anywhere, and who commits to it? If the answer is "it depends," there is no clock.
Fourth: the person receiving the report cannot be the person it concerns
Article 9(1)(c) asks for one thing:
"the designation of an impartial person or department competent for following-up on the reports"
It is easy to read that as a courtesy provision. It is not. It is structural.
The recipient of a shared mailbox is usually the HR director. Where the report concerns the HR director, or a peer of theirs, or the general manager who reviews their performance, the channel does not become awkward for that report — for that report it does not exist. This is not a comment on anyone's integrity, either. The most honest HR director alive cannot be the impartial recipient of a report about themselves.
Which gives you the conclusion: a single-recipient channel has a category of reports it structurally cannot carry, and that category contains exactly the most serious ones. The silence stops being surprising. It is the natural output of a channel whose scope excludes the subjects most worth reporting.
The right way to test this condition is not "who can report." It is "who can be reported on." If a name is missing from the second answer, the condition is not met.
Two Turkish instruments have written these down — and neither one probably binds you
There are two places where Turkish regulators have set down what an internal reporting channel has to do: principle 4.5.9 of the Capital Markets Board's (SPK) Corporate Governance Communiqué, and Article 7(2)(e) of the Banking Regulation and Supervision Agency's (BDDK) Internal Systems Regulation. The first binds listed companies. The second binds banks. Neither binds a privately held hotel group.
That is not a reason to put them aside. These are the only two places in Türkiye where the requirements for an internal reporting channel are written down at all. If a board, a lender, a brand partner or an acquirer is going to benchmark you against something, there is no other domestic reference point available. They are the measure because they are the only written measure — not because they apply to you.
The gap has been recorded from outside as well. In the OECD Working Group on Bribery's due diligence warning on Türkiye, dated 9 July 2026, the listed areas of concern include this line:
"Ineffective detection of foreign bribery allegations, including a lack of whistleblower protection in the private and public sectors"
That warning is about the detection of foreign bribery. It is not a finding about Turkish employment law and it creates no obligation. What it does say is narrower and still worth knowing: an intergovernmental body has put on the public record that whistleblower protection is absent in Türkiye's private sector.
SPK 4.5.9: a method is mandated, not a product
Principle 4.5.9 of the SPK's Corporate Governance Communiqué II-17.1 governs the audit committee. The paragraph that touches our subject reads, in the original:
"Şirketin muhasebe ve iç kontrol sistemi ile bağımsız denetimiyle ilgili olarak şirkete ulaşan şikayetlerin incelenmesi, sonuca bağlanması, şirket çalışanlarının, şirketin muhasebe ve bağımsız denetim konularındaki bildirimlerinin gizlilik ilkesi çerçevesinde değerlendirilmesi konularında uygulanacak yöntem ve kriterler denetimden sorumlu komite tarafından belirlenir."
In substance: the audit committee determines the method and criteria to be applied in examining and concluding complaints reaching the company about its accounting and internal control system and its independent audit, and in evaluating employees' reports on accounting and independent audit matters within the principle of confidentiality.
Three things follow.
Employee reports are named explicitly — "şirket çalışanlarının … bildirimlerinin". That is the text's own wording, not our reading of it.
Confidentiality is part of the principle rather than a preference — "gizlilik ilkesi çerçevesinde".
And the point that matters most here: what is mandated is a method, not a product — "uygulanacak yöntem ve kriterler … belirlenir". The regulation does not tell anyone to buy software. It tells the audit committee to determine a method. Worth writing down plainly, because plenty of sales conversations imply otherwise.
The scope has to be stated in the same breath: muhasebe ve iç kontrol sistemi ile bağımsız denetim — accounting, internal control and independent audit. Reports about harassment, discrimination or occupational health and safety fall outside this principle entirely. In a hotel group, that is where most of the traffic would come from.
BDDK 7/2(e): wider in scope, narrower in who it reaches
The second text is the BDDK's Regulation on Banks' Internal Systems and Internal Capital Adequacy Assessment Process. Article 7 governs the audit committee, and point (e) of its second paragraph requires the committee to ensure:
"Banka içi usulsüzlüklerin doğrudan kendisine veya iç denetim birimine ya da müfettişlere bildirilmesini sağlayacak iletişim kanallarının tesis edilmesini sağlamak"
In substance: that communication channels are established allowing irregularities inside the bank to be reported directly to the committee itself, to the internal audit unit, or to the inspectors.
Two things make this stronger than the SPK principle.
The scope is wider. The subject is banka içi usulsüzlükler — irregularities inside the bank, generally, rather than accounting and audit only. And the direction is specified: the channel has to reach the audit committee, internal audit or the inspectors directly. That is the fourth condition above — separating the recipient from the subject — written by a regulator rather than argued by us. Point (b) of the same paragraph asks for channels letting internal-systems staff reach the committee directly; together they say the line must not run through management.
Its reach, on the other hand, is far narrower: banks only. A hotel group is not within scope. It remains the clearest answer written in Türkiye to the question of who a channel has to reach.
Put the two texts side by side and the position is this. Domestic regulation does not order you to build a channel — but it has twice described what a channel would have to satisfy. Closing that distance is your call, and it is a management decision rather than a compliance one.
Five questions to ask before you build anything
These are the questions to answer before you choose anything. None of them needs a vendor name; each has an answer that either exists in what you have today or does not.
- Who can access that address — in the mail system, not in the policy? If the answer comes from a document, there is no answer. It has to come from the console.
- Is the record of the report held separately from the report itself? If the person the record is about can edit it, there is no record.
- When a report arrives, within how many days is receipt confirmed, and who commits to that? "It depends" is not a timeframe.
- Where does a report go when it concerns the person who reads the mailbox? No answer here means the channel's scope has narrowed to exclude the heaviest reports.
- Who protects the confidentiality of the person named in the report? The question covers the person reported on as much as the person reporting.
If the answers are "no" or "I don't know," what you are missing is not software but a definition: you have an address, not a channel. For most groups, getting to "yes" means collecting reports in one structure first, then auditing that structure against the four conditions above — and the order cannot be reversed, because there is nothing to audit until the structure exists.
Working out which of the four conditions your current etik@ mailbox meets is half a day's work in most groups. If you want a second pair of eyes on your own setup against this framework, get in touch — the conversation starts with what you can measure today, not with what to buy.


