Don't Call Me at Work: Inconvenient Times and Places, and What the Call Record Has to Show

At 4:52 on a Tuesday afternoon a consumer picks up, listens to the opening, and says: "I'm at work, I can't do this here. Don't call me on this number during the day."

The collector says "no problem, I'll try you another time", and ends the call. It was handled politely and it lasted nineteen seconds. On most scorecards it would not be reviewed at all.

Three days later the dialler places a call to the same number at 2:15 in the afternoon.

Nobody on the floor did anything they would recognise as wrong. The collector was courteous. The dialler stayed inside its permitted hours. And the consumer has now been called at a time and place she said, out loud, on a recorded line, was inconvenient. This piece is about that gap: what the rules put in play, which half of the problem your telephony already handles, and which half exists only in the conversation.

It is orientation for people who run QA, not legal advice. Where a question turns into a legal conclusion we say so and stop.

What the text puts in play

The Fair Debt Collection Practices Act limits when and where a debt collector may communicate with a consumer, and Regulation F restates and elaborates those limits. The statute is on the FTC's site; the regulation is at 12 CFR Part 1006, and the Consumer Financial Protection Bureau publishes it with its official interpretations in an interactive version. Read section 1006.6 and section 1006.14 there rather than relying on anybody's summary, including this one.

In outline, the provisions a call reviewer meets are these:

Provision (as we read the outline)	What it concerns	Where the triggering fact usually lives
Unusual or inconvenient time	Communicating, or attempting to, at a time the collector knows or should know is inconvenient	The clock, and what the consumer has said
The 8 a.m. to 9 p.m. presumption	Absent knowledge to the contrary, times outside that window, local time at the consumer's location, are treated as inconvenient	The dialler and its time-zone data
Unusual or inconvenient place	Communicating at a place the collector knows or should know is inconvenient	What the consumer has said
Place of employment	Communicating at work where the collector knows or has reason to know the employer prohibits it	What the consumer or employer has said
Medium of communication	Using a medium the consumer has asked the collector not to use	What the consumer has said or written

Two features of the wording carry most of the operational weight.

First, the limits reach attempts as well as communications. A call that rings out, or reaches voicemail, at a time the consumer designated as inconvenient is still within the provision's reach as we read it. "She didn't answer" does not make the question go away.

Second, the standard is knows or should know. The rule does not wait for the consumer to use a particular word. That is the whole difficulty, and we come back to it below.

The clock half is a dialler problem, and mostly a solved one

The 8-to-9 presumption is the part of this every floor already knows. It is enforced by configuration: the dialler holds a time zone for each number and will not launch outside the window.

It is worth being precise about what that configuration assumes. The presumption runs on local time at the consumer's location, and a dialler does not know where a consumer is. It knows an area code and perhaps a postal address. Mobile numbers travel with people who move, so an area code can sit in one time zone while the address on file sits in another. The official interpretations address this conflicting-information situation directly; the approach they describe, in substance, is to call at a time that would be convenient in every location the collector's information points to. How your platform implements that — or whether it does — is a question for whoever configures it.

From a QA standpoint the clock half has one important property: it is not in the call recording. The time a call was placed is a telephony fact. A QA record that timestamps each call can corroborate it, but the rule's question is about the consumer's local time, and that depends on location data the QA record does not hold. This is the same division of labour we described for call counts in the seven-in-seven frequency presumption: the system that places the calls owns the proof.

The spoken half is where floors lose it

Everything else in the table is triggered by something a person said. And people do not say "I hereby designate weekday afternoons as an inconvenient time." They say:

• "I'm at work."
• "I work nights, I sleep till two."
• "Not now, I'm driving."
• "Don't call this number, it's my mum's house."
• "I'm not allowed personal calls here."
• "Text me, don't ring."

The official interpretations make the point that a collector can know, or be in a position where it should know, that a time or place is inconvenient even though the consumer never used that word. They also indicate that a collector may ask follow-up questions to pin down what the consumer means — which times, which place, which number. Whether any particular phrase on any particular call crossed the line is a judgement for your counsel. What a QA programme can do is make sure the phrase was noticed, clarified and carried forward.

Those are three separate steps, and they fail separately.

Noticed. Statements like these arrive mid-sentence, usually in the first fifteen seconds, often while the collector is still delivering the opening. A collector focused on right-party confirmation hears "I'm at work" as an obstacle to the call rather than as information about the account.

Clarified. "I'm at work" could mean not right now, or never at this number, or never between nine and five. Those produce three different restrictions. A collector who does not ask has left the floor to guess, and both guesses are bad: too narrow and the consumer is called again at the wrong time; too broad and the account is needlessly unworkable.

Carried forward. The restriction has to be recorded somewhere that the dialler obeys. A free-text note reading "cx at work, cb later" changes nothing about when the next call launches.

The seam, drawn precisely

The statement is made in the conversation. The restriction has to live in the platform. The only bridge between them is a collector's fingers in the seconds after the call.

That is the same seam we described for cease requests and disputes, with one difference that makes this version harder to audit. A cease request tends to be emphatic and memorable. A time-or-place statement is casual, the call that contains it is short and friendly, and nothing about it looks like a compliance event. On a sampled programme, a nineteen-second call that ended politely is the last thing anyone pulls for review.

So the failure is quiet. It surfaces weeks later as a complaint that reads "I told them not to call me at work", and the first question anyone asks is whether she did. If the call was recorded and can be found, you can answer. If it was recorded and cannot be found without knowing the date, you effectively cannot.

It is also the kind of gap an outside review is built to find. The Consumer Financial Protection Bureau's supervision and examination materials frame compliance management around whether what staff are trained to do is what the records show they did, and the Bureau's rulemaking history for Regulation F, published in the Federal Register, explains the reasoning behind the time and place provisions at length.

Six ways it goes wrong on real calls

1. Acknowledged, not recorded. "No problem, I'll try you another time" — and no restriction is entered. By far the most common.

2. Pressed past the statement. The consumer says "I can't talk right now" and the collector replies "this will only take a minute" and continues into the balance. Asking when would be better is one thing; carrying on with the substance of the call is another. Reviewers should mark what happened and how long it continued, and leave the characterisation to counsel.

3. Never clarified. "Don't call me in the mornings" is logged as-is. Mornings until when? On which number? Every day?

4. Recorded against the wrong thing. The consumer said not to call her work number; the collector flagged the whole account as do-not-call, or flagged her mobile. The first costs the client a workable account. The second leaves the work number live.

5. The employer statement missed. "I'm not allowed personal calls here" is information about what the employer permits, which the place-of-employment provision is specifically concerned with. It is often heard as small talk.

6. The inbound call at the "inconvenient" time. A consumer who said not to call before noon rings in at 9:30. The official interpretations discuss a consumer who initiates contact at a time or place they earlier designated as inconvenient, and the latitude described is narrow. A collector who treats the inbound call as the restriction being lifted for good has assumed more than the consumer said.

The complaint that taught us this was not about an aggressive call. It was a perfectly nice fifteen-second call where she said she was a nurse on nights and we said "sure". Then we rang her at eleven in the morning four more times. Every one of those calls was inside permitted hours and every one was a problem.

— Compliance manager, healthcare receivables agency, 9 years, name withheld by request

What a reviewer should listen for

A workable scorecard treatment has one detection row and three follow-through rows, all binary:

• Did the consumer say anything indicating a time, place, number or medium was inconvenient or unwelcome? Yes or no, with the quoted words and the timestamp.
• If yes: did the collector acknowledge it?
• If yes and the statement was ambiguous: did the collector ask what it covered?
• If yes: after the statement, did the collector continue with collection content beyond asking for a better time? Note the duration.

A fifth question cannot be answered from the call: was a matching restriction entered on the account? That needs the platform. A sensible routine is for the reviewer to list every "yes" from the first row each morning and for someone with platform access to check each account the same day. It is ten minutes of work for a floor of twenty and it closes the seam.

Wording the detection row is harder than it looks, and it is a good candidate for a calibration session. "I'm busy" — yes or no? Reviewers will split, and the scorecard should say which way your floor leans.

What automated screening can and cannot do

Transcription makes the spoken half searchable, and that changes the economics. A phrase like "at work", "don't call", "I'm driving" or "text me" can be found across every call rather than across the 2 percent somebody happened to pull, and a short friendly call is no less likely to be screened than a long hostile one.

It does not settle anything. "I'm at work" appears on a great many calls where the consumer goes on to talk happily for ten minutes. A screen that surfaces the phrase is giving a reviewer a place to listen. The reviewer decides whether the consumer was designating anything, and counsel decides what follows from it. A flag is not a finding covers that triage.

And the screen sees only the conversation. It cannot tell you what time it was where the consumer was standing, and it cannot tell you what the collector typed afterwards.

Where our product fits, and where it does not

As of October 2026, our Voice tier — $99 per monitored collector per month, supervisors and admins free — records each call on two channels, transcribes it with timestamps, and screens every transcript against a collections rubric, reporting each dimension separately. A consumer's statement is attributed to the consumer's channel, so "who said it" is a fact about which device the audio came from. Six months of calls can be searched for a phrase. Features has the detail and pricing the tiers.

The limits that matter for this topic:

• No account linkage. Calls tie to collectors and times, not to accounts in your platform. We can show that a consumer said "don't call me at work" on a given collector's call at 4:52; we cannot show whether a restriction was then entered, or whether a later call went to the same consumer. That reconciliation happens in your system of record.
• No dialler control. Nothing we do stops a call launching. Permitted hours and time-zone logic belong to your telephony.
• Next-morning review. A statement made today is surfaced tomorrow. If the dialler's next attempt is scheduled before then, the collector's own note is the only protection.
• Softphone calls only. A call made from a desk handset that does not route through the PC produces no audio for us to capture.

Our compliance page sets out the same boundaries more generally. Nothing in this article confers compliance on a floor that adopts it; it describes where to look.

Frequently asked questions

Does a consumer have to say the word "inconvenient"?

The official interpretations of Regulation F indicate that a collector can know, or should know, that a time or place is inconvenient without the consumer using that word. Which statements meet the standard on which facts is a legal question for your counsel. Operationally, treat any statement about when, where or how the consumer does not want to be reached as something to notice, clarify and record.

What are the default calling hours under the FDCPA and Regulation F?

Absent knowledge to the contrary, times before 8 a.m. and after 9 p.m. local time at the consumer's location are presumed inconvenient. That is a presumption about the clock; a consumer's own statements can make other times inconvenient as well. Read the text of section 1006.6 for the exact wording.

Does an unanswered call count?

The provisions refer to attempting to communicate as well as communicating, so as we read them an attempt at a designated-inconvenient time is within their reach. Confirm how that applies to your dialling practices with your own counsel.

Can a QA tool tell us whether we called at an inconvenient time?

Only partly. A call record can show what the consumer said and when the call took place. It generally cannot show the consumer's location, and without a link to the account it cannot show that a later call went to the same person. The dialler and the collections platform hold those facts.

What should a collector do when a consumer says "I'm at work"?

That is a scripting and training decision for your floor and your counsel. From a QA perspective, the pattern that produces a clean record is: acknowledge, ask what the statement covers if it is unclear, stop the substantive conversation, and enter a restriction the dialler will obey rather than a free-text note.

How do we find these statements if we only sample calls?

With difficulty. The calls that contain them are short and polite, which makes them the least likely to be selected for review. Searching transcripts of every call for the common phrases, then having a reviewer listen to each hit, is the practical route.

Is this the same as a cease-communication request?

No. A cease request concerns stopping communication; a time, place or medium statement concerns when, where or how. They are addressed in different provisions and they produce different restrictions on the account, which is one more reason to clarify what the consumer meant.