6.9% of ChatGPT Interactions Indicated Buying Interest in U.S. Study

Try Our Free Tools!
Master the web with Free Tools that work as hard as you do. From Text Analysis to Website Management, we empower your digital journey with expert guidance and free, powerful tools.

Insights into Consumer Purchase Intent Through ChatGPT

Recent findings reveal that consumer purchase intent manifests in approximately 6.9% of interactions within ChatGPT, as observed by an opt-in panel analyzed by SparkToro.

This investigation provides brands with valuable insights into consumer behavior in the realm of conversational AI.

The comprehensive data arises from Verso, which meticulously monitors users’ ChatGPT conversation histories upon account linkage.

The analysis encompasses a substantial sample size of 41,961 dialogues, generated by 12,321 individuals spanning from January 1 to October 29, 2026.

Notably, the majority of this data has been statistically weighted to mirror the demographic contours of the U.S. population.

Within the sub-sample of interactions with ChatGPT, an overwhelming 87.9% of responses refrained from referencing any specific brand, product, or business.

Conclusively, Rand Fishkin, co-founder and CEO of SparkToro, posited that the significance of AI visibility primarily stems from Google’s rollout of AI Overviews and AI Mode, rather than from ChatGPT itself.

Frequency of Purchase Intent in Conversations

The analysis delineated three distinct categories of conversation goals. Of the total dialogues analyzed, 4.9% indicated that the interlocutor was contemplating a purchase and evaluating options; an additional 2.1% consisted of queries explicitly related to the acquisition of a specific brand, product, or service; while a staggering 93.1% exhibited no purchasing intention whatsoever.

Fishkin remarked, “This could signify troubling news for ChatGPT’s advertising prospects.”

He also referenced SparkToro’s December 2024 analysis, which scrutinized approximately 332 million Google searches. He noted that 15% or more of these searches could be classified under his definition of commercial intent.

The analysis was conducted on search data from a U.S. panel employing browsers from January 2023 to September 2024.

Within this dataset, the category defined as ‘commercial intent’ accounted for 14.5% of searches, even as the searchers may not have been poised to finalize purchases at that moment.

Transactional searches constituted a meager 0.69%. However, it is pertinent to mention that these categories do not align with those identified in the ChatGPT analysis, thus rendering direct numerical comparisons infeasible.

Mentions of Brands in Conversations

The report further delineated inquiries regarding brands, revealing that within the weighted group, a substantial 86.7% of prompts made no request for brand, product, or company information.

Of the prompts studied, 6.6% pertained to multiple brands, while 6.7% focused on a singular brand. On the response side, 7.0% of all replies referenced multiple brands, products, or companies, whereas 5.1% referred to just one. Hence, the cumulative rate of responses mentioning at least one brand was reported at 12.1%.

The post’s conclusions characterized 12% of ChatGPT interactions as integral to the buyer journey, although its purchase intent analysis distinctly indicated a figure of 6.9%.

Utilization of ChatGPT

Prominent categories for ChatGPT utilization were elucidated, with requests for images and visual design leading the chart at a considerable 17.6%.

Following closely were inquiries related to information and entertainment (14.4%), writing and communication (12.7%), while shopping, brand, and local service inquiries comprised 5.3% of the total.

For categorization purposes, each conversation was assigned a singular category reflecting its primary objective.

This method prioritizes the output type over the topic of inquiry; for instance, a request for assistance in drafting an email seeking a refund would fall under writing, not shopping.

Notably, the percentages presented in the report’s visual chart diverge from those in the textual analysis by fewer than two percentage points across categories.

Comparison with OpenAI’s Data

The panel data was juxtaposed with OpenAI’s Signals data from September. OpenAI’s dataset appears to skew more heavily towards writing and technical support queries, while the weighted panel data inclined towards image requests and instructional questions.

The author asserts that comparisons between the two datasets are justified, acknowledging that OpenAI’s figures likely encompass API requests absent from Verso’s data.

According to OpenAI’s documentation, the dataset exclusively reflects messages within ChatGPT Free, Go, Plus, and Pro accounts, omitting Codex and enterprise usage, thereby excluding API calls as well.

Furthermore, OpenAI’s data represents a global message distribution, while the panel’s figures are confined to U.S.-based conversations.

Data Methodology

Verso provides a complimentary tool enabling users to analyze their ChatGPT interaction history. According to its sign-up terms, Verso retains imported conversations, utilizing them in de-identified formats for datasets licensed to research and commercial entities. Notably, Aaron Aknin, co-founder of Verso, provided the data underlying the analysis.

The report also indicates that the panel is somewhat biased towards younger demographics, including women and individuals from lower-income brackets.

Specifically, 9,675 panelists possessed demographic attributes suitable for statistical weighting to align with the 2024 United States Census.

No usage category diverged by more than 1.4 percentage points after applying these demographic adjustments.

Prior to commencement of data analysis, the 131 most active users (responsible for over 33 prompts per active month) were excluded from the sample.

Despite representing merely 1% of the total panelists, these heavy users generated 21% of all prompts.

Conversely, the average member of the adjusted group (excluding the aforementioned top users) contributed 9.9 prompts from January to October.

As most panelists participated via mobile devices, there exists a possibility that usage through desktop computers and work accounts was underreported. Notably, API requests were not incorporated into the overall dataset.

Each conversation was meticulously coded utilizing a structured codebook supplied to Anthropic’s Claude model, based on user-generated prompts and the initial responses from ChatGPT.

A secondary coder independently recoded a random sample of dialogues, achieving an agreement rate of 90% for categories and 97% to 99% for brand tags. However, no distinct agreement figure has been disclosed concerning the identification of purchase intent.

SparkToro offers a suite of audience research tools.

The Importance of These Findings

For brands navigating the complexities of AI visibility and ad budgeting, the U.S. panel elucidates that less than 7% of ChatGPT interactions are entwined with purchasing intent.

Fishkin concludes that AI visibility’s importance is predominantly driven by Google’s AI Overviews and AI Mode, raising pertinent questions about the merit of tracking ChatGPT visibility in isolation, unless it is considered alongside Bing rankings and AI visibility metrics.

Large digital screen displaying Google AI with a globe graphic in a modern server room; several people are visible in the background.

The report underscores that the panel provides a more accurate reflection of the inquiries directed to ChatGPT than the frequency of overall usage.

For pay-per-click (PPC) teams assessing ChatGPT Ads, the 6.9% purchase intent share delineates a snapshot of conversation dynamics within a consumer panel with limited ChatGPT engagement, rather than serving as a comprehensive account of purchase-driven conversations across the platform.

Future Directions

Until the panel data consolidates desktop, workplace, and API usage, it remains prudent to align purchase intent metrics such as these with one’s own ChatGPT referral and advertising analytics prior to incorporating them into strategic budget considerations. Fishkin has expressed anticipation for further collaborative research endeavors with Verso.

Source link: Searchenginejournal.com.

Disclosure: This article is for general information only and is based on publicly available sources. We aim for accuracy but can't guarantee it. The views expressed are the author's and may not reflect those of the publication. Some content was created with help from AI and reviewed by a human for clarity and accuracy. We value transparency and encourage readers to verify important details. This article may include affiliate links. If you buy something through them, we may earn a small commission — at no extra cost to you. All information is carefully selected and reviewed to ensure it's helpful and trustworthy.

Reported By

Ranjana Banerjee

I’m Ranjana Banerjee, Creative Content Manager at RSWEBSOLS in Kolkata, India, with 10+ years of experience in blogging, SEO, digital marketing, and e-commerce. I create high-quality content and SEO strategies that boost traffic, improve rankings, and help businesses grow in competitive markets.
Share the Love
Related News Worth Reading