Ad Tech's New Pre-Launch Test Gave Converse's Pulled Ad an F

An ad tech model gave Converse's controversial September sneaker ad an F — flagging "hate symbols" and lynching imagery — while ChatGPT rated the same creative a B minus and declined to call it culturally insensitive. That gap is the entire pitch behind Digital Culture Group, which launched its Resonance Studio module in late September to predict how creative will perform before it ever goes live.

The timing wasn't accidental. Converse pulled the ad after two weeks of backlash over imagery many viewers said evoked the KKK and lynching. Some longtime fans threw away or burned their shoes and posted it on social media. CEO and founder Crystal Foote called it "an extremely unfortunate thing for the Converse team to experience," and horrible for "the communities that were impacted by that."

Foote uploaded the ad to both ChatGPT and ARI, the company's flagship Audience Resonance Index. ChatGPT's only critique was that the sneaker depiction "could be interpreted as gender non-conforming styling." ARI instead cited "hate symbols" including robes and hoods, and noted that "the dangling shoes could be read as echo of hanging or lynching imagery" that could offend Black Americans, Foote said, quoting the model outputs directly. Its fix: drop the white robes, cone-shaped hoods and dangling feet, and put the product in a shoebox or on a shelf.

One caveat worth noting: since the Converse ad was already live, ARI was likely influenced by real public reactions to it. The company says the tool would ideally flag such imagery pre-launch, but that specific claim hasn't been tested on a genuinely unseen case.

What does ARI actually draw on?

The model trains on data from the American Community Survey, eMarketer's predictive analytics and research from Nielsen's cultural team, according to Foote. A social listening feature taps conversations on Reddit, X and Meta. Alongside Resonance Studio, the company launched the ARI 100 Ad Pledge, offering 100 pro bono creative assessments to brands and agencies.

Three participants show what practitioners get back — mostly detailed notes, not grades alone:

• Care.org, which signed on in September, submitted a National Women's Day ad asking viewers to sign a pledge to "level the playing field for women and girls everywhere," said Mary Kate Wilson, associate VP for creative and content marketing. ARI wanted specifics: whether signatures went to a petition or a legislator. People ask, "What difference do I make by lending my signature to this?"
• A second Care ad on the history of the WWII-era care package scored high on cultural relevance, but needed to be "more time-bound to today," Wilson said.
• Vella Bioscience submitted an ad for Meltwater, a vaginal moisturizer with hyaluronic acid and aloe vera. Resonance Studio predicted a click-through rate of 106% of benchmark and psychographic alignment around 85%. Its critique: the ingredient-focused ad had only one model shot, and needed more diversity in age and race.

For Vella, the stakes go beyond targeting. President Joy Allen-Altimare said the brand works to challenge "outdated stereotypes and stigma" around topics like menopause, while navigating platform rules that force ambiguous wording around pleasure and intimacy — especially for women's products. Resonance Studio "will help us understand if we're talking to ourselves," she said.

The open question is scale and independence. All three early examples are pledge participants getting free assessments, and none tested the tool on creative before a public controversy. Until a brand credibly reports that ARI caught something pre-launch that nothing else did, the Converse comparison stays a compelling demo — not proof.