TF1: ad revenue down 9.3% in Q3 2026. M6: video ad market estimated at -9%. Social: +15.7%. Why falling TV advertising is the best news in your 2027 media plan.
On 29 October, TF1 published its nine-month accounts: 308 million euros of advertising revenue in the third quarter, down 9.3%, and one sentence in the release, reported by AFP, that I have been rereading for a week: "Q3 2026 marks a fourth consecutive quarter of a steeper-than-anticipated decline in the linear advertising market." On 3 November, M6 published its own, up 9.4% thanks to the World Cup, and slipped into its release, according to CB News, that the French video advertising market is down "around 9%" and that the environment remains "without visibility". My view: when the country's two largest sellers of airtime write, five days apart, that their market is falling faster than expected and that they cannot see ahead, that is not the moment to flee television. It is the moment to buy it. And it is exactly the opposite of what the scale-ups I meet are doing.
TV advertising in decline: the autumn 2026 numbers
I start with what the accounts say, not the slogans. At TF1, according to AFP and CB News, group revenue to the end of September reached 1.454 billion euros, down 9% year on year, and advertising revenue fell from 1.12 to 1.02 billion, or -8.9% over nine months. Pierre-Alain Gérard, the chief financial officer, said it plainly: "This decline is essentially driven by advertising revenue." Net profit fell from 123 to 71 million, -42%, and the group is holding its targets thanks to "savings on programme costs". Meanwhile, TF1+ made 51 million euros of advertising in the quarter, +20.5%, and the group claims 42 million monthly streamers: the platform is growing, but as AFP also notes, it "does not make up for the fall in linear advertising".
At M6, the story looks different on the surface: 294.5 million of revenue in the third quarter, +9.4%, and 199.2 million of video advertising revenue, +6.2%, with the group's "best quarter ever on commercial targets", a 23.5% audience share among 25-49s. But read the line underneath: the group claims an "outperformance of 15 points against the rest of the French video advertising market, which it estimates to be down around 9%". Seventeen World Cup matches in July, video programming costs up from 362.5 to 447 million over nine months, and a record audience: it took all of that to post +6% in a market at -9%. Streaming, at 30.9 million and +26.7%, is 13% of the video division. And the group has announced a plan for 80 million of annual savings. A seller that strings together audience records and prepares a savings plan is a seller that knows the price of its airtime is under pressure.
The first-half BUMP barometer, published on 10 September by France Pub, IREP and Kantar Media and reported by The Media Leader, sets the frame: an advertising market of 9.753 billion euros, +5.6%, but the five traditional media at 3.081 billion, -4.5%, and television at -6.2%, with linear advertising airtime down 15% and thematic channels at -29.7%. For the full year, France Pub forecasts the five media at 7.7 billion, -2.5%, and digital at 12.3 billion, +8.6%. Virginie Sablé, its managing director, sums it up: "The erosion of the five media should continue." She is right. And that is precisely my argument.
Where the money leaving TV goes
It goes into the auction. In the same BUMP, social reached 2.219 billion euros in the half, +15.7%, "the strongest growth in the sector" according to Christine Robert, deputy director of IREP; search, 2.740 billion, +12%. I wrote last week about what that demand produces: the price of a Meta ad rose 14% in a year, for an audience growing 3%. The two figures are two sides of the same coin. The money leaving television enters an auction, and an auction turns every additional euro into inflation for everyone else.
Look at who is queuing at each counter. Kantar Media counts 60,597 active advertisers in the first half: 51,474 on digital, 17,708 on the five media, 9,199 in television. Five and a half times more buyers on the digital side, where "3% of advertisers account for 80% of investment", according to Zaïa Ferhaoui of Kantar Media. In other words, on Meta and Google, a scale-up bids against fifty thousand advertisers and against a handful of giants who set the market price. In television, it negotiates with three sales houses that have just written that they lack visibility, at a counter where nine thousand buyers show up, only 9% of whom activate linear, catch-up and addressable TV at the same time. I know where I would rather be a buyer.
Is TV really cheaper? Not the way you think
I am not going to tell you that TV CPMs have fallen 9%. That is not what the numbers say. Advertising airtime down 15% for revenue down 6.2% is a sales house selling fewer breaks to defend its rate card. The channels are not discounting their schedules: they are cutting supply. And the World Cup showed it on a large scale: according to Kantar Media, average revenue per match fell from 5.4 to 4.3 million euros between 2022 and 2026, because the number of matches sold rose from 162 to 270. When supply grows, the unit price falls 20%. When it shrinks, the rate holds.
So for a buyer, the decline of television does not show up in the gross rate card. It shows up in three places. Thematic channels, at -29.7%: an inventory that loses nearly a third of its revenue in six months does not sell at catalogue price. Negotiating terms: volume discounts, bonus airtime, flexibility on positions, everything a sales house grants when, as at TF1, "the evolution of the linear market" determines whether it hits its annual targets. And the calendar: a seller finishing a fourth quarter that was "steeper than anticipated" needs its first quarter of 2027, the time of year when TV is cheapest anyway. That is where I place new entrants. I have detailed what a startup needs before its first TV campaign; the orders of magnitude have not changed, the room for negotiation has.
My rule for buying TV in 2027
I never buy a medium because it is rising; I buy what others are leaving, at the moment they leave it. For 2027, that means: a first quarter negotiated in November and December, when sales houses are trying to secure their year; a share of thematic channels and addressable TV bought at the price of their decline, not their catalogue; the broadcasters' streaming services, TF1+ and M6+, for targeting precision, since that is where their growth is at stake and where the sales houses want to prove it works; and a measurement of my own, brand searches and direct traffic in one area against a control area, never the sales house's wrap-up report. All of it capped: television is not a bet, it is a line in the plan, sized against everything else.
Who should go now
D2C brands and scale-ups saturating Meta. When social weighs more than half the plan, each additional contact costs more than the previous one, that is the physics of the auction, and 14% inflation a year steepens the slope. I have explained why a higher TV CPA can be more profitable than a Meta CPA: when one rises 14% and the other is negotiable, the gap closes without you doing anything. For Sisley, we grew ROAS by 45% across twelve markets without touching the brand image, by working on budget allocation rather than the bid.
Launches. A brand that does not exist yet has nothing to gain in an auction where it pays a price set by advertisers ten times its size. For Exploris, we built the 360° launch of a premium brand: awareness is made on the screen where attention is least contested, and it then converts into cheaper search and social, because people search for your name instead of your category.
Retail and networks. Television remains the medium that fills stores, and its addressable version lets you buy it area by area. It is the kind of network we are currently supporting with Banette, a network of artisan bakeries, on its digital strategy and social activation; for a network, the question is never "TV or not", but "which areas, at what price", and the price has not been this negotiable in years.
What I tell founders
I am not telling you television is going to recover. I think it will keep declining, as France Pub forecasts, and that the sales houses will keep shifting their growth to TF1+ and M6+, +20% and +27% this quarter. I am telling you that a medium losing buyers becomes a buyer's market, and that a media plan is the only place where you decide the price you pay. While 51,000 advertisers fight over the same people on the same six-inch screens, the biggest screen in the house is negotiated with three sales houses that need you. I have written about what you can do with self-serve TV for 1,000 euros; what I describe here is the floor above, the one where you negotiate. The media mix simulator gives you a first order of magnitude of what 20% of your social budget buys on television and CTV. And if you want to know what the sales houses would grant your brand for the first quarter of 2027, that is the question I ask on your behalf in a free audit.
