France Killed Its Cold-Call Opt-Out List

The Day France Scrapped Its Do-Not-Call List

On August 11, France banned unsolicited sales calls. You’ve probably already seen that in the news. What caught my eye, though, was the system that got scrapped alongside the ban: a do-not-call registry called Bloctel¹. Launched in 2016, it lasted ten years before being shut down on the very same day.

Reader, this move actually bundled two things together. It banned unsolicited sales calls, and it eliminated the registry where consumers had to sign up their own numbers to opt out. The structure flipped: instead of consumers having to register first to avoid getting calls, businesses now have to get consent first.

So what actually changes here isn’t the size of the fines — it’s the cost structure of outbound sales and how companies manage consent data. Consent is no longer something you collect once and use indefinitely; it now expires after a year, meaning it has to be renewed annually. And over this same period, voice spam in Korea has been moving in the opposite direction — it’s going up.

Cold-Calling Ban and Bloctel’s Abolition Took Effect the Same Day

The basis is Article 13 of Law No. 2025-594, enacted on 30 June 2025. It takes effect on 11 August 2026. What’s interesting is where this clause sits. It wasn’t tacked onto a consumer-protection reform — it rode along with a public subsidy fraud prevention law. That means the French government treated the problem not as a consumer nuisance but as fraud infrastructure.

Behind this lies years of accumulated consumer frustration. French authorities estimated that roughly three-quarters of citizens receive at least one unsolicited sales call per week. In 2024, 11 consumer groups jointly demanded an outright ban. The fact that this was still the situation after a do-not-call list had existed for 10 years is the starting point for this law.

The substance is simple. Businesses can no longer call consumers for sales purposes without prior consent. There are only two exceptions: contact related to an ongoing contract, or cases where clear consent was obtained in advance. Newspaper and magazine subscription solicitations get a separate carve-out. Conversely, three sectors — energy retrofitting, aging-housing renovation, and the CPF (individual training account) — face a complete ban on sales calls even with consent, since these are where fraud has been most concentrated.

Penalties are capped at €75,000 for individuals and €375,000 for corporations, per call. Converted at the exchange rate on 11 August 2026 (roughly ₩1,637 (~$1.19) per euro), that’s about ₩120 million (~$87,000) and ₩610 million (~$444,000) respectively. These are caps, not automatic fines — France’s Directorate General for Competition, Consumer Affairs and Fraud Control (DGCCRF)² investigates and imposes penalties, and can publicize the sanctions. When vulnerable individuals are targeted, penalties can rise to 5 years’ imprisonment and €500,000 or 10% of revenue.

But from a practitioner’s standpoint, the real weight isn’t the fines. Any contract signed via a call made without consent is void. In other words, a sale doesn’t count as revenue. Fines are a matter of probability, but a voided contract is a matter of the transaction itself — and that’s what will change sales organizations’ behavior far more quickly.

And consumers don’t need to do anything. No need to register with Bloctel, no need to cancel it either. The default is now that a call can’t be made at all without consent.

France Moves to Opt-In, Korea’s Voice Spam Doubles

Look at Korea’s numbers over the same period, and the trend runs the opposite way.

According to the “2025 Second-Half Spam Distribution Status” report released on May 14, 2026 by the Broadcasting and Media Communications Commission and the Korea Internet & Security Agency (KISA), the average monthly spam volume per person rose to 10.35 messages, up roughly 30.8% from the prior half-year. Within that total, voice spam alone hit 4.26 calls — double the previous half-year figure and the highest in five years. The top advertising category was telecom subscription solicitation. The Commission attributed this to expanded voice-based sales by distributors following the July 2025 repeal of the Mobile Device Distribution Improvement Act.

What matters here is that text and voice spam moved in exactly opposite directions. Over the same period, reported/detected text spam fell from 150.2 million cases to 12.88 million — a drop of about 91%. Voice spam reports/detections, meanwhile, rose from 5.04 million to 8.73 million, up roughly 73%.

Why the divergence? The answer lies in the rules. Article 50 of Korea’s Information and Communications Network Act requires explicit prior consent from recipients before sending commercial advertisements through electronic transmission media. On text messages, email, and automated calls, Korea is already an opt-in³ country. But there’s one exception: under the Door-to-Door Sales Act, calls placed by a telemarketing seller⁴ using a live human voice — provided they disclose the source of the personal information used — are exempt from this consent requirement.

The exception this piece focuses on is specifically live-voice calls placed by telemarketing sellers under the Door-to-Door Sales Act, where the source of the personal data is disclosed. This doesn’t mean every live-voice call is exempt from the consent rule. Against the backdrop of this exception, text spam fell while voice spam rose. Still, these two indicators alone aren’t enough to conclude that the entire shift is due to the regulatory gap.

There is an opt-out mechanism. But it’s the same registry-based opt-out model France scrapped on August 11. There’s Do Not Call, built in 2014 by the Fair Trade Commission under Article 42 of the Door-to-Door Sales Act and operated by the Korea Consumer Agency under delegation, and there’s a separate Financial Sector Do Not Call registry run independently by 12 financial industries including banks, insurers, and card companies. Financial product solicitation is entirely excluded from the FTC’s Do Not Call registry. So consumers have to register with both systems separately.

The speed is even more disappointing. Businesses are only required to cross-check the opt-out list once a month or more. That means someone can still receive calls for up to 30 days after registering. The Financial Sector Do Not Call registry states it can take up to two weeks for a registration to take effect. Consumers have to go through the registration process themselves, wait for it to be reflected, and deal with a system split across two separate registries. That’s the burden placed on anyone simply trying to say no to phone calls.

A One-Year Consent Expiry Date Rewrites the Cost of Doing Business

The real weight of France’s regulation sits in the implementing decree issued on 23 July 2026, because that’s where the specifications for consent were laid out.

Consent must be free, specific, and clear, and must be revocable at any time. Pre-checked boxes are invalid, and burying consent inside terms-of-service can’t substitute for it. So far, none of this is surprising. The problem is in the next three lines.

• Consent is valid for a maximum of 1 year and does not auto-renew
• Consumers can revoke consent verbally, mid-call
• Businesses must keep proof of consent for at least 3 years, provide it to consumers on request, and bear the burden of proof themselves

Under these terms, simply having a contact isn’t enough to keep calling it forever. Once consent expires after 1 year, the business has to re-obtain it, and that process costs money. This doesn’t mean a contact list depreciates 100% every year on the books — it means the business now has to actively manage which consents remain usable for outreach.

Here’s what shifts on the cost sheet: where a per-unit list-purchase price used to sit, a per-unit consent-acquisition cost now takes its place. It’s effectively a new line item inserted ahead of customer acquisition cost (CAC). And the annual re-consent campaign stops being a one-off expense and becomes a fixed cost that recurs every year. The clause allowing verbal revocation mid-call makes the burden even heavier — if a call goes badly, the consumer can revoke consent on the spot, and that contact becomes unusable for sales going forward.

The trouble is that most organizations aren’t even ready to start doing this math. Contact data typically sits in a single table, blended from records that flowed in through multiple campaigns, often with no record of when consent was originally obtained. If you don’t know when it was given, you can’t know when it expires.

The shockwaves cross borders. Morocco’s Minister of Employment reported to parliament that up to 50,000 jobs in the country’s call centers are at risk. The industry has drawn roughly $100 million in investment and generates over $1 billion in annual revenue. An industry source in Morocco’s outsourcing sector said the French market has historically accounted for more than 80% of the industry’s revenue. The same source did stress diversification, noting that pure telemarketing now makes up only 15-20% of overall activity. Still, I don’t read that number as a reassuring signal. The staff and operational infrastructure built to handle French-language calls with French consumers also get deployed for other kinds of work, so a decline in the 15-20% telemarketing share can drag down those adjacent operations too.

When the country writing the regulation and the country actually staffing the phone calls are different places, the burden — job losses included — shows up first outside the regulator’s own borders. A single clause passed by the French parliament ends up rewriting employment plans in Casablanca.

Oswarld’s Lens

There’s a scene I run into constantly while building GTM strategies. It’s the moment a client casually mentions “our 300,000-record database” as if it were an asset. I always ask the same question in response: where did these numbers come from, when was consent obtained, and where’s the proof? A clear answer rarely comes back.

That’s why I recommend treating consent not as a checkbox field in a CRM but as a ledger. Source, time of acquisition, expiration date, withdrawal history — all of it needs to be recorded per contact. Once the burden of proof shifts to the business, any contact whose consent can’t be verified becomes hard to use for sales and carries the risk of legal disputes. It’s far cheaper to do this work before the regulation arrives, because retroactive reconstruction is essentially impossible.

Still, I’m not ready to assume this regulation will succeed. Germany has enforced a similar ban since 2009, yet unauthorized sales calls haven’t disappeared. The head of the French consumer group UFC-Que Choisir has said she worries fraudsters will simply move their operations to door-to-door sales instead. Regulation tends to raise costs for the compliant side first. So the likely sequence is that legal outbound calling drops first, while illegal calls decline only much later.

I think Korea also needs to examine how its prior-consent principle and the exemption for telemarketing sales actually function in practice. If AI voice agents drive down the cost per call, the volume of sales calls could actually increase. We also need to evaluate whether a system where consumers register individually on opt-out lists is sufficient on its own to address that.

Closing

If I compress today’s rundown into three lines, it’s this.

In France, businesses now have to confirm consent before making sales calls, and consumers are protected even without registering on a separate opt-out list. Businesses have to manage the 1-year validity period for consent and any withdrawal history, and keep documentation on hand as proof. In Korea, prior consent is the principle as well, but telemarketing has a qualifying exception for “nurture calls” (solicitation calls to existing customers). With voice spam on the rise, it’s worth taking a closer look at how this exception and the do-not-call registry actually operate.

If you’re running outbound campaigns, I’d suggest checking three things this week. Where did the numbers on our list come from, when was consent obtained, and does the proof still exist in a form you can actually produce right now? If you can’t answer even one of these three, that list isn’t an asset yet.

💬 Over the past month, how many sales calls did you get where you thought, “I have absolutely no memory of consenting to this”? Drop the industry in the comments, and I’ll compile them by sector in the next issue.

---

📨 If you have a colleague working on outbound or CRM, please forward this to them.

---

References & Further Reading

Primary sources

• Légifrance, “Consumer Code Article L.223-1 et seq.: Consent to Telephone Solicitation” (version effective August 11, 2026). Link ··· This is the statutory text itself. If you want to check exactly how far the exceptions extend, this is the most precise source.
• French Ministry of the Economy and Finance (DGCCRF, France’s competition and consumer-fraud authority), “Telemarketing Rules That Apply to Businesses.” Link ··· It lays out practical specs like the 1-year validity period for consent and the 3-year record-retention requirement. Good reading from a business operator’s perspective.
• Service-Public.fr, “Ban on Telemarketing Calls: What Are the New Rules?” Link ··· A government explainer, written for consumers, on the abolition of Bloctel and the two exceptions.
• Broadcasting Media Communications Commission and Korea Internet & Security Agency, “Spam Distribution Status, Second Half of 2025,” published May 14, 2026. Link ··· Every Korean figure in today’s piece comes from this report. Even just looking at the voice-spam section on its own tells the story.
• France 24, “France bans unsolicited telemarketing calls to protect consumers,” August 2026. Link ··· This article covers the impact on Moroccan call-center employment in the most concrete detail.

Background

• CNIL, “Rules on Commercial Prospection by Telephone.” Link ··· Consent requirements as seen from the perspective of France’s data protection authority. If you’re curious about the overlap with GDPR, this is the one.
• Korea Fair Trade Commission, “Do-Not-Call Registry for Telemarketing Sales.” Link ··· Try registering yourself, and you’ll immediately understand why financial products are excluded — and why one registration doesn’t settle things for good.

Related past issues

• The Cloudflare Bot That Got a Name and a Wallet ··· A case that redesigned how access itself gets granted. It connects to today’s consent structure.
• Manatoki Dies, But the Invoice Keeps Coming ··· Covers the pattern of regulation and liability drifting toward whichever party is easiest to grab hold of.

---

The author is Oswarld (Kwangseob Ahn). Current roles: Adjunct Professor at Sejong University, Strategy Consultant at INLEVEL9. Career, research, books, and recent work are kept current on the About page. Latest · July 2026: HEMA-2: A Consolidation-Aware Tri-Memory Architecture with Multi-Channel Scheduling for Lifelong Conversational AI.

📝 Glossary

Footnotes

1. Bloctel: A do-not-call registry France introduced in 2016. Once a consumer registered a number, businesses were required to exclude it from telemarketing. It was abolished as of August 11, 2026. ↩

2. DGCCRF: France’s Directorate General for Competition Policy, Consumer Affairs and Fraud Control. It’s the agency that investigates violations of consumer law and imposes administrative sanctions — roughly analogous to the consumer-protection function of Korea’s Fair Trade Commission. ↩

3. Opt-in / opt-out: Opt-in means a business must obtain consent in advance before making contact; opt-out means the business can contact people by default, and only excludes those who explicitly object. The dividing line is who bears the burden of doing the refusing. ↩

4. Jeonhwa-gwonyu-panmae (telemarketing sales): A concept under Korea’s Door-to-Door Sales Act, referring to sales of goods or services conducted by soliciting over the phone or inducing a callback. Businesses engaging in this must register with their local city, county, or district office. ↩