GM views car dealers as a primary source of customer insight, not just a sales channel. Dealers effectively run continuous A/B tests on messaging and can provide real-time feedback on what resonates with customers—what "makes their eye sparkle"—which is often more potent than formal research.
Related Insights
A GSB receptionist's casual chats with alumni revealed the program's long-term "fine wine" value—a strategic insight that formal surveys often miss. This shows how empowering frontline employees to listen can uncover profound user truths.
Marketing decisions should be driven by testing and data, not by the subjective opinions of internal stakeholders. The phrase "I wouldn't click on that" is a red flag for a poor marketing environment that lacks a culture of experimentation because you are not your audience.
Don't wait for large corporate campaigns to get audience feedback. Marketers should be "religiously" creating content on their personal social channels to micro-test messaging, language, and program ideas. This provides a direct, rapid feedback loop on what the audience actually cares about, enabling content-led innovation.
Conversational ads offer an unprecedented one-on-one channel for brands to interact with customers at scale. The resulting data—customer questions, complaints, and feedback—is a goldmine for product development and other business functions, potentially exceeding the value of immediate customer acquisition.
To truly understand customers, go to their natural environment—their home or shop. Observing their context reveals far more than sterile office interviews. This practice, internally branded "Listen or Die," ensures the entire team stays connected to the user's reality.
AI can't replicate insights gained from direct customer interaction. Methods like joining sales calls, reading product reviews, and one-on-one interviews provide "first-party data" essential for creating resonant content and differentiating your brand from competitors relying on public data.
Instead of asking AI for a final answer, use it as a sophisticated focus group. Prompt it to embody different customer personas (e.g., "a left-leaning feminist," "a conservative male") and provide feedback on your messaging from those perspectives. This helps refine copy before market testing.
Instead of guessing at marketing copy, build an AI model of your ideal customer. By feeding it internal data like call transcripts and external data like forum posts, this "digital twin" can review and rewrite your marketing materials using the customer's exact language.
Brands miss opportunities by testing product, packaging, and advertising in silos. Connecting these data sources creates a powerful feedback loop. For example, a consumer insight about desirable packaging can be directly incorporated into an ad campaign, but only if the data is unified.
Instead of vague sales correlations, GM marketing's success is tied to two specific outcomes: being in a car buyer's initial consideration set and winning the subsequent shopping journey. This provides clear, measurable goals and removes ambiguity about marketing's contribution to the business.