• by Laurie Sullivan , Staff Writer, 6 hours ago

Verve Group can activate conversational intent signals for targeting from Large Language Models (LLMs) by unifying zero-party data, search intent, and pseudonymized AI chat activity into one layer of intelligence.

The company claims that it supports a privacy layer while processing more than 1 billion daily signals. This is done by leveraging several integration layers from recent acquisitions to transition targeting capabilities from traditional search to AI discovery from queries. 

Verve says this development makes its technology the first open advertising platform to offer "high-fidelity intent data from AI chat interfaces for programmatic" campaigns.

The platform aggregates and activates signals from conversational AI engines, allowing advertisers to reach audiences based on complex, real-time behaviors that surface in these chatbots.

Although AI is able to predict behavior, advertisers may need that extra data, despite early tests suggesting that chatbot ads like OpenAI's require simplistic ads.

ChatGPT ads seem to favor clarity over creativity, which may mean many tech platform developers have to rethink strategies -- unless this is temporary learning curve.

If true, or if the trend persists, one company that comes to mind is Smartly.io, which just signed a deal with OpenAI, ChatGPT's parent company, as the first creative tech platform for its ad content. 

Adthena, a tech platform that analyzed about 40,000 daily placements, found that sentence structure, among other things, plays a key role.

"In some ways it’s like stepping back in time to Google AdWords early days 15+ years ago and working with ad copy, but there is masses of opportunity here to fine tune copy if you're currently in the trial or about to launch your first OpenAI ad campaign,” Adthena CMO Ashley Fletcher wrote in a post on LinkedIn.

Similar to a news article, poem or work of fiction, he wrote that every word must carry weight and contribute directly to the clarity of the message, as well as the conversion. In fact, the average headline clocks in at just 30 characters and around five words.

Nearly every headline in ChatGPT ads leads with the brand name, and averages 30 characters, with the longest at about 36 characters. The constraint forces concise messages in ads.

Every headline reads like a product stamp rather than a tagline, according to the data. Body copy generally averages 116 characters and roughly 19 words.

The dollar symbol drives conversions and anchors value in the ad. Adthena explains that a concrete financial claim outperforms vague promises when users are researching products and services.

Offers like “free” are the most common conversion triggers because they reduce commitments, but calls to action like “shop” or “now” are explicit.

While post ChatGPT ads have seen success, there are some that have failed. Adthena has noted in the past that one client’s clickthrough rate (CTR) was just 0.91%, roughly 7 times lower than the 6.4% benchmark for Google Search in that same sector.

“Low tech, no performance data, and limited targeting,” Glenn Gabe, SEO and AI Search Consultant at G-Squared Interactive,” wrote on X, citing reporting from The Information. “Advertisers that bought ChatGPT's first ad campaigns say the process was low tech and that they haven't received much data showing if their ads worked.”

He wrote that two ad agency executives worked with ChatGPT advertisers who said they had not yet proved the ads have driven any measurable business outcomes for their clients.