When Someone Tells You to Stop Contacting Them

A consumer says stop calling, or don't call me at work, or I've got a lawyer. The restriction takes effect when they say it, not when your systems find out — and the gap between those two moments is where nearly every violation in this area happens. Not because anyone decided to keep calling, but because a dialer campaign was already built, a letter cycle was already queued, and an agency working the account never received the flag. That makes this a propagation problem with a technical answer, which is good news, because propagation is something you can engineer and intent isn't.

The restrictions are not the same

Treating every restriction as one thing is the first error, because they have different scopes and different requirements.

Request	Scope	Operational effect
Stop all communication	Broad	Suppress everything, subject to limited exceptions
Don't call at work	One number or context	Restrict that number; others may continue
Don't call this number	One channel instance	Restrict it specifically
Don't email / text me	A channel	Restrict the channel
Not before 9 / after 6	Timing	Window restriction
I have an attorney	Who may be contacted	Escalate immediately
I dispute this	Different mechanism entirely	Per our dispute analysis

The specifics of what each requires depend on who is collecting, the jurisdiction, and the applicable rules — and those distinctions matter enough that the system should record which restriction was requested rather than a single flag meaning "restricted."

Two practical points. Over-restricting is far safer than under-restricting, and an operation that treats an ambiguous request as broad has made the right error. And the two bottom rows aren't contact restrictions at all — they're different triggers with different processes, and conflating them into a contact flag loses the actual instruction.

Where the failures happen

The core of the problem, and it's the same structure our bankruptcy guide identifies with filings.

A consumer asks a collector to stop on Tuesday. What can still happen:

• A dialer campaign built Monday calls Wednesday.
• A letter queued last week arrives Thursday.
• An email sequence running on its own schedule continues.
• An agency working the account never received the flag.
• A different internal team — retention, servicing — contacts them about the same account.
• The note is in free text and no automated system reads it.

Six ways to violate a restriction that everyone involved intended to honour.

The recurring pattern this desk has now found in three separate contexts — bankruptcy filings, vulnerability disclosures per our disclosure analysis, and contact restrictions — is identical: a signal reaches one system and not the others, and automated processes continue because nobody told them. It's the same finding our complaint analysis makes about frequency complaints.

Which means the answer is architectural rather than procedural. If suppression is a property of the account that every system reads before acting, all six failures disappear at once. If it's a flag each system has to be separately told about, you'll fix them one at a time forever.

Capturing it properly

What has to happen at the moment the request is made:

1. Record it in a structured field, not free text. A note nothing reads stops nothing.
2. Record which restriction — the specific scope, not a generic flag.
3. Record how it arrived — verbally on a call, in writing, through a portal.
4. Record when, with a timestamp, since the effective moment is the request.
5. Apply it immediately, in the same session.
6. Confirm to the consumer what you've applied, which prevents a second request and creates a record.
7. Retain the record permanently, per our records guide — you may need to demonstrate when it was applied.

The collector needs to be able to do all of this without escalation. Per our monitoring analysis, a response requiring approval frequently doesn't happen — the call ends, the queue is long, and the restriction is applied a day later than it should have been.

And collectors need training to recognize an informal request as a request. "Please don't call me at work anymore" is a restriction whether or not it uses any particular words, and a consumer isn't required to phrase it correctly for it to count.

Propagation

The section that determines whether any of the above matters.

Every system that can initiate contact must read the restriction before acting. The inventory:

• The dialer — and campaigns already built need to check at dial time, not at build time. This is the single most common gap.
• Letter generation, including items already queued at the printer.
• Email and text platforms, which frequently run on separate infrastructure.
• Agencies and third parties — propagation to them is a data feed, not an email to a contact, and it needs to run on the same cadence as your own systems.
• Other internal teams touching the same customer.
• Any vendor performing outreach on your behalf.

The agency point deserves emphasis because it's the largest residual exposure. Per our agency analysis, their conduct is attributed to you — and an agency working from a file extracted before the restriction will contact the consumer with no way of knowing.

What that requires contractually and operationally:

• Restrictions transmitted on every file exchange, not just at placement.
• A defined maximum lag between your recording it and their receiving it.
• Contractual obligation to apply within that window.
• Immediate recall of accounts where a broad restriction is applied.
• Audit rights, exercised.

Notice of representation

The trigger requiring the fastest and least discretionary response.

Notice that a consumer is represented changes who may be contacted about the matter, and continuing to contact the consumer directly afterwards is a serious problem.

What makes it operationally difficult:

• It frequently arrives mid-conversation and informally — "my lawyer is handling this" in the middle of a call.
• Collectors may not recognize it as the trigger it is.
• It may arrive by letter to a department that doesn't route it quickly.
• It requires stopping immediately, not at the end of the sequence.

What the process should be:

1. End the call politely and take the attorney's details if offered.
2. Apply full suppression immediately, before verification.
3. Escalate to whoever handles represented accounts.
4. Record it as a distinct flag, not as a general contact restriction.
5. Propagate, including immediate agency recall.
6. Take advice on what contact remains permissible.

Step two is the important one: suppress first, verify afterwards. The asymmetry is the same as with bankruptcy filings — a wrongly suppressed account costs a few days of collection activity, and a wrongly contacted represented consumer costs considerably more. That asymmetry should be built into the system rather than left to a collector's judgment under time pressure.

What a restriction doesn't do

Worth being clear about internally, because misunderstanding runs in both directions.

A communication restriction limits contact. It does not:

• Extinguish the debt. The obligation continues.
• Stop credit reporting, which per our furnisher guide continues under its own obligations.
• Prevent other lawful action on the account.
• Resolve anything. The balance remains and the situation is unchanged.

Two consequences for the operation.

Don't treat a restricted account as resolved. It still needs a decision — an arrangement if one can be reached through permitted channels, a write-off decision per our small balance analysis, or whatever else is appropriate. A restricted account left indefinitely in a suppressed state is a deferred decision, not a handled one.

Don't use the limits of the restriction as a workaround. A consumer who asked you to stop calling and then receives daily emails has been complied with technically and not in substance — and per our complaint analysis, that's exactly the conduct a complaint rate is supposed to catch.

Working the account afterwards

What's left, and it's more than operations assume.

• Permitted channels, where the restriction was channel-specific. A consumer who restricted calls may be entirely willing to resolve it by email.
• Inbound contact. Make it easy for them to reach you — a self-service path per our small balance analysis costs nothing and requires no outreach at all.
• Self-service resolution, which is the highest-value option on a restricted account since it needs no contact whatsoever.
• A decision about the account, made deliberately.

The second and third are worth building specifically for this population. A consumer who restricted contact hasn't necessarily refused to pay — frequently they've objected to how they were being contacted, which is a different thing entirely. An operation that makes paying easy without requiring a conversation recovers from accounts it had written off as uncooperative.

That reframe is worth stating plainly: a contact restriction is information about your outreach, not only about the customer. A rising rate of restrictions is the same kind of signal as a rising complaint rate, and it should be read that way.

The audit that proves it works

The control that tells you whether everything above is functioning, and it's simple.

Run a standing report of any contact occurring after a recorded restriction. Every instance is a finding.

What to examine on each:

• Which system initiated it.
• The lag between recording and the contact.
• Whether the restriction was recorded correctly at capture.
• Whether propagation reached that system, and when.
• Whether it was an agency, and what their lag was.

What to track over time:

• Contacts after restriction, per thousand restricted accounts.
• Propagation lag, internal and to each third party.
• Restriction rate per thousand contacts, as an outreach quality signal.
• Whether the failure count is falling, which is the only test that matters.

An operation that has never run this report doesn't know whether its restrictions work — and given the six failure modes above, the prior should be that some don't. The report costs a query and it finds problems that no amount of training would have surfaced.

Frequently asked questions

---

This guide is educational and does not constitute legal or compliance advice. The effect of communication restrictions, which parties they bind, available exceptions, requirements following notice of legal representation, and rules governing electronic contact vary by jurisdiction, by whether the collector is a first party or a third party, and by product, and they continue to develop. Consult qualified counsel about the requirements applicable to your operation.