Is ChatGPT Driving Your Sales? How to Approach the Unanswerable Question

A board member asks the CMO a simple question: "Is ChatGPT sending us customers?" The CMO pauses. The question seems straightforward, but the honest answer is anything but simple.

According to TechCrunch, ChatGPT now has over 800 million weekly active users. Many of them ask for product recommendations, research purchase decisions, and seek advice that influences where they spend money. Your brand might be benefiting from these conversations—or being left out of them entirely. But you can't actually prove either.

Welcome to the most frustrating question in modern commerce: Is ChatGPT driving your sales, and how would you even know?

The honest answer is that you can't know with certainty. But that doesn't mean you're completely in the dark. With the right frameworks, you can develop informed perspectives about ChatGPT's likely impact—even without definitive attribution.

The ChatGPT Revenue Mystery

ChatGPT has become a mainstream research tool. When consumers want unbiased advice, when they're overwhelmed by options, when they want to quickly understand a category they know nothing about—they increasingly turn to AI assistants for help.

This behavior matters enormously for commerce. The recommendations AI provides shape consideration sets. The brands mentioned become candidates for purchase. The products highlighted get priority attention. In categories ranging from consumer electronics to home goods to business software, ChatGPT responses influence billions of dollars in purchasing decisions.

Yet from a measurement perspective, this influence is almost entirely invisible.

When a consumer asks ChatGPT "What's the best mattress for back pain?" and receives a response that includes your brand, real value has been created. That consumer is more likely to consider your product, visit your website, and ultimately purchase. But no click connected the AI conversation to your analytics. No referral string passed through. No session identifier linked the AI user to the eventual purchaser.

The influence was real. The measurement path doesn't exist.

This creates a mystery at the heart of modern commerce. ChatGPT is almost certainly influencing your sales—the question is whether that influence is positive, negative, or neutral for your brand specifically. And the available data cannot tell you directly.

What We Know About AI Commerce Behavior

While direct measurement eludes us, research and observation reveal patterns about how consumers actually use AI assistants for commerce.

Frequency of commerce-related queries. Surveys consistently show that significant percentages of ChatGPT users ask commerce-related questions. Product recommendations, comparison shopping, feature research, problem-solution matching—these use cases appear commonly across user bases. Your customers are likely among them.

Trust in AI recommendations. Studies suggest that AI recommendations carry substantial weight with users. Many consumers view AI as more objective than traditional sources—free from the paid placements and review manipulation that have eroded trust in conventional channels. This perceived objectivity amplifies the influence of AI recommendations.

Journey stage patterns. AI assistants are particularly influential in early journey stages: need recognition, initial research, and consideration set formation. By the time consumers reach comparison and purchase stages, AI has often already shaped which options they're evaluating. This early-stage influence is especially difficult to attribute because it precedes measurable touchpoints.

Category variation. AI commerce influence varies significantly by category. Complex, considered purchases—electronics, appliances, business software, financial products—show high AI research rates. Habitual, routine purchases show less AI involvement. Understanding your category's profile helps estimate potential AI impact.

Demographic patterns. Younger consumers, tech-forward users, and early adopters show higher rates of AI commerce usage. If your customer base skews toward these segments, AI influence on your sales is likely more significant.

These patterns suggest that AI is a real factor in commerce—but they don't tell you specifically how ChatGPT affects your brand's sales. For that, you need to develop more tailored approaches.

Indirect Signals and Indicators

While you can't directly attribute sales to ChatGPT, you can look for indirect signals that suggest AI influence might be at play. These signals are imperfect—each has alternative explanations—but patterns across multiple indicators can build confidence.

Direct traffic anomalies. When consumers research via AI and then visit your site by typing your URL or searching your brand name, it appears as "direct" traffic in analytics. Unusual increases in direct traffic, especially among new visitors, might indicate AI-influenced arrivals. This is noisy—direct traffic fluctuates for many reasons—but persistent changes warrant investigation.

Branded search patterns. AI recommendations often prompt branded searches as consumers verify information or seek additional details. Changes in branded search volume, especially for queries that suggest comparative shopping ("brand X vs alternatives"), might reflect AI influence. Again, many factors affect brand search, but patterns can be suggestive.

Consideration pattern shifts. If AI recommendations favor certain products, you might see changes in which products customers consider together. Analytics showing altered comparison patterns, product page co-visitation changes, or shifts in competitive consideration might reflect AI influence on consideration sets.

Geographic or demographic signals. If AI commerce usage is higher among certain demographics or in certain geographies, you might look for performance differences across segments. Stronger performance among AI-heavy user segments could suggest AI influence—though again, many factors differ across segments.

Temporal patterns. Did something change in your metrics around the time of major AI adoption waves? Correlation isn't causation, but timing patterns can be suggestive. If performance inflections coincide with AI commerce growth, it's worth investigating.

None of these signals alone is conclusive. Each has explanations unrelated to AI. But when multiple indicators align, they can suggest a pattern worth taking seriously.

Survey and Research Approaches

The most direct way to understand AI's influence on your customers is to ask them. Survey and research approaches have limitations, but they provide ground truth that analytics cannot capture.

Post-purchase surveys. Adding questions about AI usage to post-purchase surveys can reveal how many customers used AI in their purchase journey. Simple questions work: "Did you use ChatGPT, Claude, or other AI assistants while researching this purchase?" Follow-ups can explore what they asked and how it influenced their decision.

Consideration stage research. Intercept surveys during the shopping process—on comparison pages, in cart abandonment flows, or during account creation—can capture AI influence while it's fresh. Customers may more accurately recall recent AI interactions than past ones.

Qualitative research. In-depth interviews or focus groups can reveal how AI actually influences purchase decisions. These methods sacrifice scale for depth, revealing the mechanisms of influence rather than just its presence. Understanding how customers use AI helps interpret quantitative signals.

Panel studies. Ongoing research with consumer panels can track AI commerce behavior over time. This reveals trends in how AI usage evolves and how it relates to purchase behavior. Longitudinal data is particularly valuable for understanding trajectory.

Competitive perception research. Surveys can explore whether consumers perceive your brand as well-represented in AI recommendations compared to competitors. Even if customers can't precisely recall AI interactions, their general impressions can suggest whether you're benefiting from AI visibility.

Self-reported data has known limitations. Customers may not accurately recall AI interactions, especially if significant time has passed. They may not connect AI research to eventual purchases. And stated importance may not reflect actual influence on behavior.

But survey data provides something analytics cannot: any visibility at all into AI-influenced behavior. Imperfect data about a hidden channel is more valuable than perfect ignorance.

Correlation-Based Estimation

Another approach involves looking for correlations between AI visibility and business outcomes. This requires the ability to monitor your presence in AI responses—how often you're recommended, how you're described, and how you compare to competitors.

With visibility data and business outcome data, you can explore correlations:

Cross-sectional correlation. Across product lines, do products with stronger AI visibility show stronger performance? This requires comparing AI visibility and business metrics across your portfolio, controlling for other factors like marketing spend and price positioning.

Temporal correlation. When AI visibility changes—whether through deliberate optimization or external factors—do business metrics follow? Lag patterns between visibility changes and outcome changes can suggest causation, though many confounds remain.

Competitive correlation. Across competitors in your category, do those with stronger AI visibility show stronger market performance? This is limited by data availability but can reveal category-level patterns.

Natural experiments. Sometimes external events create quasi-experiments. If a competitor's AI visibility suddenly drops due to a product issue or negative publicity, do you see corresponding improvements? If your visibility increases after a product launch, do metrics respond?

Correlation-based estimation has clear limitations. You cannot control for all confounds. Correlation doesn't prove causation. And the data required—accurate AI visibility measurement across time—is itself challenging to obtain.

But in the absence of direct attribution, correlation analysis offers one path toward quantified estimates of AI impact. Even wide confidence intervals are more useful than no estimate at all.

Building Confidence Without Certainty

The goal isn't to answer "Is ChatGPT driving your sales?" with absolute certainty—that's likely impossible with current technology. The goal is to develop a well-reasoned perspective supported by multiple lines of evidence.

Consider building a confidence assessment that synthesizes available evidence:

Category analysis. Is your category one where AI commerce research is common? Complex, considered purchases with many options are more likely to involve AI research. If you're selling mattresses, laptops, or enterprise software, AI influence is more plausible than if you're selling gas station snacks.

Customer profile analysis. Is your customer base likely to use AI for commerce? Younger, tech-forward customers use AI more heavily. If your customers match this profile, AI influence is more likely.

Visibility assessment. What is your brand's presence in AI recommendations? If you're frequently recommended, you're likely benefiting from AI influence. If you're absent or poorly represented, AI might be working against you. Monitoring visibility provides a leading indicator.

Survey evidence. What do your customers report about AI usage? Even rough survey data provides ground truth that analytics cannot. If 25% of your customers report using AI in their purchase journey, AI influence is clearly significant.

Signal pattern analysis. Do indirect signals suggest AI influence? Changes in direct traffic, branded search, consideration patterns, and other indicators—especially when they correlate with AI visibility changes—can suggest AI impact.

Competitive context. How do competitors view AI commerce? If competitors are investing heavily in AI visibility, they likely see evidence of impact. Their actions provide external validation of the channel's importance.

No single line of evidence is conclusive. But when multiple indicators align—when you're in a high-AI category, with AI-forward customers, maintaining strong AI visibility, seeing suggestive signals in your analytics, and observing competitor focus on the channel—the case for AI influence becomes compelling even without direct attribution.

Communicating Uncertain Findings

Armed with a confidence assessment, you can return to that board member's question with a nuanced response.

"We can't directly measure ChatGPT's impact on our sales—the attribution paths don't exist. But based on our analysis, we have moderate-to-high confidence that ChatGPT is positively influencing our sales. Our customers fit the profile of AI commerce users. We're generally well-represented in AI recommendations. Our surveys suggest a quarter of customers use AI in their purchase research. And several indirect signals suggest AI is playing a role in our traffic patterns."

This response is honest about limitations while providing actionable insight. It acknowledges that certainty isn't available while demonstrating that informed analysis is.

The board member might follow up: "So how much revenue is ChatGPT driving?" Here, appropriate humility is essential. You might provide a range based on your evidence—"somewhere between X% and Y% of our revenue is likely influenced by AI recommendations"—while being clear about the substantial uncertainty in that estimate.

What executives actually need isn't false precision. It's a defensible perspective that can guide strategy. Should we invest in AI commerce visibility? The answer depends on whether AI is likely influencing our sales. And that question, while not answerable with certainty, can be approached with evidence and reason.

Living with the Unanswerable

The question "Is ChatGPT driving your sales?" may never have a definitive answer. The fundamental architecture of AI commerce—private conversations, click-free influence, cross-platform journeys—makes traditional measurement impossible.

But unanswerable doesn't mean unimportant. Companies need to develop perspectives on AI commerce impact even without perfect data. Waiting for measurement certainty that may never arrive means ceding advantage to competitors willing to act on incomplete information.

The frameworks described here—indirect signals, survey research, correlation analysis, confidence assessment—don't solve the measurement problem. They work around it. They provide enough insight to guide strategic decisions without requiring the impossible precision of direct attribution.

This is uncomfortable for organizations accustomed to data-driven decision-making. But the alternative—ignoring AI commerce because it can't be measured—is worse. As AI commerce grows, the companies that develop competence in navigating measurement uncertainty will outperform those that remain paralyzed by it.

Is ChatGPT driving your sales? Probably, if you're well-represented in its recommendations. Probably not, if you're absent. The exact magnitude remains uncertain. But you know enough to act strategically—and in today's AI commerce landscape, that's the most anyone can reasonably expect.

---

The challenge of measuring ChatGPT's revenue impact extends across the AI commerce landscape. Learn more about the broader AI attribution problem facing executives and explore why tracking AI-influenced purchases remains impossible with current technology.

Ready to understand your ChatGPT visibility? While revenue attribution remains uncertain, visibility monitoring can reveal how often you're recommended. Explore how platforms like Noema approach this challenge.

---

Want to see how your store scores? Run a free AI readiness scan and get your store's AI visibility report in 60 seconds.

---

About the Author: Josh is the founder of Noema, an AI commerce observability platform that helps e-commerce brands understand how AI shopping agents see their products. Noema has scanned 80,000+ Shopify stores to build the industry's most comprehensive AI readiness benchmarks.