Sydney Sweeney's Novig Ad Reveals a New Risk for Brands

Sydney Sweeney's ad for the sports prediction app Novig drove a ninefold jump in brand mentions. That figure comes from Peak Metrics data, reported via Business Insider on September 15, 2026. Only 4 percent of that conversation named the app itself. Novig gained the audience. Sweeney absorbed the reputational cost, even though she is a strategic partner and equity holder in the company, not a hired spokesperson.

Updated November 5, 2026.

What actually happened with the Novig ad?

Novig launched its "Just Sports" campaign on September 9, 2026, with Sweeney appearing nude and covered only by sports equipment. Novig CEO Jacob Fortinsky told Front Office Sports on September 15, 2026, that Sweeney approached the company about the campaign. She served as its creative lead. She holds equity in Novig. Four-time Olympic gold medalist Ariarne Titmus called the ad "infuriated" on Instagram, and the backlash spread quickly among professional female athletes.

The criticism followed a familiar pattern for Sweeney. Her 2025 American Eagle "great jeans" campaign drew a similar wave of backlash over its wordplay on "jeans" and "genes." That earlier controversy is part of why the Novig ad got read instantly as a repeat. Reporters and commentators reached for the comparison within hours of the Novig launch, and the comparison stuck.

What are the key numbers at a glance?

The figures below come from the reports cited in this article. They are listed together so the scale of the campaign, and the imbalance inside it, are easy to compare.

Measure	Figure	Source
Campaign launch date	September 9, 2026	Front Office Sports interview with Novig's CEO
Increase in Novig brand mentions	More than ninefold	Peak Metrics, via Business Insider
Average weekly brand mentions	392 rising to 4,633 in four days	Clayton Durant, CAD Management, via Inc.
Estimated potential reach	786.6 million rising to 3.98 billion	Clayton Durant, CAD Management, via Inc.
Share of conversation naming Novig	4 percent	Peak Metrics
Share of conversation naming Sweeney	96 percent	Peak Metrics
Novig's 2026 funding	$75 million Series B at a $500 million valuation	Reported in coverage of the campaign

Did the campaign work for Novig?

By Novig's own numbers, yes. Peak Metrics found the company's mentions across X, Instagram, TikTok, Reddit, Threads, and Bluesky surged more than ninefold after launch. A separate analysis by Clayton Durant of CAD Management, reported by Inc. on September 14, 2026, found average weekly brand mentions climb from 392 to 4,633 in four days. Estimated potential reach rose from 786.6 million to 3.98 billion over the same window. Novig, which raised $75 million in a Series B round this year at a $500 million valuation, bought the attention it wanted.

Two independent firms measured the same spike using two different methods. That kind of agreement is rare in social measurement. It removes most of the room to argue the numbers are an artifact of one firm's methodology. Whatever else is true about the campaign, the awareness result is not in dispute.

Why did Sweeney absorb the cost instead of Novig?

Peak Metrics found only 4 percent of the social posts about the campaign named Novig. The other 96 percent named Sweeney directly. This is the same asymmetry I described in celebrity marketing beyond star power. Attention concentrates on the more familiar, more searched name, regardless of who owns what. Novig grew its name recognition. Sweeney personally carried the reputational weight of a campaign she helped design.

That gap matters because it is measurable and repeatable, not a one-off quirk of this particular ad. Any campaign that pairs a well-known face with a lesser-known brand should expect the same lopsided split. The brand's own visibility grows in the background while the public conversation fixes on the person. Sweeney's case is unusual only in how cleanly Peak Metrics quantified the split.

What does AI have to do with this?

Four AI engines, ChatGPT, Claude, Gemini, and Perplexity, independently named Sweeney's equity stake in Novig. None were prompted to look for it. The finding comes from 5W's AI visibility study on the Novig campaign. That detail is now part of the retrievable record. I've written before about how AI reputation managers now shape a brand's story whether a company plans for it or not. I've also covered how social listening vs AI brand monitoring measure two different things entirely.

Novig's own social listening data shows the awareness win. What it doesn't show is that four separate AI systems have already filed away who actually owns a piece of the outcome. Social listening tracks what people posted this week. AI systems retain what gets cited and keep surfacing it in answers for as long as the underlying source material stays live. Those are different kinds of permanence, and brands running provocative campaigns need to plan for both.

How should brands structure an equity-for-endorsement deal?

A brand should structure an equity-for-endorsement deal so that the public can tell the owner's role from the spokesperson's role, and so that both parties have agreed in advance how backlash will be handled. When talent is also an owner, audiences stop separating the two roles, and the person usually receives more of the reaction than the company. These five terms are worth settling in writing before any creative is approved. They reflect my own advice, not a legal standard, and counsel should review the final agreement.

• Who speaks for what. Decide who answers press questions about the product and who answers about the talent personally. One voice for each, named in advance.
• Creative approval. When the talent is creative lead, agree on who has final sign-off and what happens if the brand and the talent disagree late in production.
• A shared backlash threshold. Define the point at which the brand and the talent issue a joint statement, such as a named group publicly objecting or a sustained run of critical coverage.
• Plain disclosure of the stake. The equity stake surfaced in AI answers without anyone asking for it. A brand that states the arrangement clearly on its own site controls how the arrangement is phrased.
• Separate measurement. Track mentions of the brand and mentions of the talent separately from the first hour, so the split is visible within a day and not discovered weeks later.

What does a 30-day plan after a provocative launch look like?

A 30-day plan starts before launch day and ends only when the record has been checked and corrected. The sequence below is a working template for any campaign that is likely to divide opinion. The timing is a recommendation, and each step should be adjusted to the size of the audience and the risk.

1. Before launch: Draft holding statements for the three most likely criticisms. Name the spokesperson, brief them and agree who approves each statement.
2. Launch day: Publish a page on your own site that states the purpose of the campaign and the plain facts, including who created it and who owns what. A citable source should exist from the first hour.
3. Days 1 to 3: Watch the split between brand and talent mentions. Correct factual errors quickly and politely. Do not argue about taste, because it draws more coverage and settles nothing.
4. Days 4 to 14: Publish a second piece that adds information, such as results, the thinking behind the creative or a response to a specific criticism. Keep it factual and short.
5. Days 15 to 30: Check what the AI engines say, update your own pages where the record is wrong and note what you would change in the next campaign.

What should talent and their teams do differently?

Talent with an ownership stake should assume that the public conversation will attach to their name first and plan their own communications accordingly. In this case the data shows 96 percent of the discussion named Sweeney, even though Novig owned the product and the marketing budget. A talent team can reduce the exposure with a few habits.

• Hold a separate statement. Prepare a personal statement that covers the talent's reasons and role, distinct from the company's statement.
• Read the first 100 responses. The first responses show which criticism is spreading and whether it is about the product, the creative or the person.
• Keep ownership facts consistent. The talent and the company should describe the stake in the same terms everywhere, because AI engines repeat whichever description is easiest to find.
• Plan for repetition. If an earlier campaign drew similar criticism, as happened here with the 2025 American Eagle campaign, assume reporters will connect the two and prepare a clear answer to the comparison.

How do you check what AI says about your campaign?

You check by running a fixed set of questions across the main AI engines every week and recording what each one says and which sources it names. The method takes less than an hour a week and gives you a record you can compare over time. It follows the same logic as measuring Citation Share for a brand.

1. Write ten questions a curious reader would ask, such as what the controversy was about, who is behind the campaign and who owns the company.
2. Run each question in ChatGPT, Claude, Gemini and Perplexity, and save the answers with the date.
3. Note which sources each engine cites and whether the facts are correct, especially ownership and timing.
4. Publish or update your own pages where the answers are wrong or incomplete, so there is a correct source to cite.
5. Repeat weekly for a month, then monthly. Compare against the first run.

For how the engines choose their sources, see how AI engines decide what to cite.

What should brands take from this?

Model the attention split before signing an equity-for-endorsement deal, not after the campaign runs. Novig's own numbers show the audience grew around Sweeney personally, not around the product she is invested in. I made a related point in prediction markets and AI search. The underlying facts of a fast-moving story get locked into the record quickly. Whoever controls that record early controls how the story gets told later.

Price the talent's equity stake against the attention split a campaign is likely to produce, not the split a pitch deck assumes. Build a plan for what gets published in the days after launch. That is the material that ends up defining the story for AI systems and search results alike. A brand that waits to respond until the backlash cycle has run its course has already ceded the record to whoever wrote about it first. Everything-PR's comparison of celebrity marketing agencies shows how the main firms structure and price talent deals, and the corporate apology guide covers what a response should contain when a campaign draws backlash.

Sources: Peak Metrics data reported by Business Insider (September 15, 2026); Clayton Durant of CAD Management, reported by Inc. (September 14, 2026); Front Office Sports interview with Novig CEO Jacob Fortinsky (September 15, 2026); 5W's AI visibility study on the Novig campaign. The 30-day plan and the deal terms above are my recommendations, not findings from those sources.

Ronn Torossian is the founder and chairman of 5W, a PR, creative and AI Visibility agency.

Frequently asked questions about the Sydney Sweeney Novig campaign

What was Sydney Sweeney's role in the Novig campaign?

Sweeney approached Novig about the campaign and served as its creative lead. She also holds equity in the company.

How much did Novig's brand mentions increase after the launch?

Peak Metrics reported that Novig's mentions surged more than ninefold across social platforms. A separate analysis found that average weekly brand mentions grew from 392 to 4,633 in four days.

How was the conversation split between Sydney Sweeney and Novig?

Only 4 percent of social posts about the campaign mentioned Novig. The other 96 percent named Sweeney directly.

Why did the Novig advertisement draw public criticism?

The campaign featured Sweeney appearing nude and covered only by sports equipment. This prompted backlash among professional female athletes, including Olympic gold medalist Ariarne Titmus.

How should a brand protect itself when talent owns equity?

Agree in writing who speaks for the brand and who speaks for the talent, set a threshold for a joint statement, disclose the stake plainly on the brand's own site and track brand and talent mentions separately from launch day.

How can I see what AI engines say about a campaign?

Run a fixed set of ten questions in ChatGPT, Claude, Gemini and Perplexity each week, record the answers and the sources they cite, and correct your own pages wherever the facts are wrong.