Advize is an AI-powered performance marketing agency that treats creative research as a non-negotiable step before production, because the most expensive mistake in Meta advertising is producing a polished asset nobody asked for. The question this post addresses directly: how do you actually know what your customer wants to see in an ad before you spend money making it? The answer is not surveys, focus groups, or trend decks. It is going to where your customer is already talking.
Why Most Creative Teams Are Answering the Wrong Question
Most creative processes start with a brief that describes the product, the audience, and the message the brand wants to send. That is the wrong starting point. The relevant question is not what message do we want to send but what is the customer already thinking about this problem before they ever see our ad. These are different questions with different answers, and [the gap between them is where most creative budgets disappear](internal-blog://182).
Consumer psychologists have established that emotional response to an ad predicts purchase intent more reliably than the ad's actual informational content. Kantar's BrandZ Global Study found that ads achieving an emotional response score above 75 converted viewers at 73.2%, while ads scoring below 40 converted at only 19.8%, a 3.7x difference driven entirely by emotional resonance rather than product logic. This means the most important creative decision is not what to say but which emotional state to meet the customer in. You cannot make that decision without understanding where the customer's head is before the ad appears.
The Three Places Your Customer Is Already Telling You What They Want to See
The first place is your own product reviews and your competitors' reviews. Read the one-star and five-star reviews separately. One-star reviews tell you the exact disappointments people carry into the buying decision, which are the objections your creative needs to address. Five-star reviews tell you the specific outcomes people were hoping for and received, which are the promises your creative can make with credibility. The language people use in reviews, not polished marketing language but the actual phrases they type when describing their problem, is [the raw material for hooks that work](internal-blog://8).
The second place is Reddit. A brand selling any product with a community around the problem category will find threads where people describe their situation in unguarded, specific language. A creative strategist researching a sleep supplement will find threads where people describe lying awake at 2am running through tomorrow's schedule rather than describing poor sleep quality in clinical terms. That specificity is the hook. Not 'improve your sleep' but 'you know that feeling when your brain won't stop at 2am.' The second version came from a real person describing a real experience.
The third place is your competitors' comment sections. The comments on a competitor's Meta ad are a live research feed of what the target audience is skeptical about, what they genuinely want to know, and what would tip them toward purchase. Most creative teams have never looked here. The ones who have found ready-made objection handling and social proof angles they could not have invented internally.
Why Specificity in Customer Language Outperforms Generic Claims in Ad Creative
The leading creative operators in 2026 share one consistent behavior: they research before they produce, and they specifically look for [verbatim customer language](internal-blog://183) rather than summarized insights. According to ad creative researchers at Motion, the most common research failure is taking a customer insight and translating it into marketing language before it reaches the creative. The original phrasing is the asset. 'My skin was literally peeling off in November' is a better hook than 'experience severe dry skin in winter.' The first came from a real person. The second came from someone who processed the real person's words through a marketing filter and lost what made them work.
This is why the best performing ads often feel uncomfortably raw to the brands making them. They use the customer's own words rather than the brand's preferred language, which can feel off-brand to the marketing team even as it dramatically outperforms polished alternatives. When UGC-style creatives outperform polished brand content by 27% on CTR and 19% on conversion rate according to Motion's 2026 benchmarks, a significant portion of that advantage is specifically the language, not just the production style.
What This Looks Like in Practice: From Research to Winning Hook
Consider a brand selling a back support product targeting office workers. The internal creative brief describes the product as ergonomic lumbar support for professionals. The research phase tells a different story: Reddit threads about back pain from sitting all day are full of people describing 'that specific ache that starts around 3pm' and 'dreading the commute home because sitting in the car is worse than sitting at the desk.' Nobody on Reddit described their problem as lumbar support deficiency.
The ad that runs from internal briefs opens with 'Struggling with back pain? Our ergonomic lumbar support can help.' The ad that runs from customer research opens with 'that 3pm back ache that makes the rest of the day feel impossible.' The second version addresses the customer at the exact moment and in the exact language they have used to describe their own experience. It does not need to explain what the product is in the hook because the customer has already recognized their situation. The research took two hours. The performance difference between the two approaches is the difference between a creative that converts and one that informs.
The Research Mistakes That Produce Generic Creative
Three specific research shortcuts produce creative that is technically competent and emotionally irrelevant. The first is [reading reviews for themes rather than language](internal-blog://183). A team that summarizes reviews as 'customers value quality and fast delivery' has lost every piece of information that makes that insight usable in creative. The useful version is reading reviews for the exact phrases people use, copying them verbatim, and tagging them by emotional state.
The second mistake is researching the product instead of the problem. Teams spend significant time understanding their product's features and comparatively little time understanding the customer's experience of the problem the product solves. The customer does not think about your product category until they are searching for a solution. Understanding what triggers that search, what the problem feels like from the inside, is where the creative angle lives.
The third mistake is conflating what customers say they want with what makes them act. Customers will say in surveys that they want clear product information and transparent pricing. What actually produces the emotional response that drives purchase is almost always something more specific and less rational, a recognized situation, a shared frustration, an unexpressed desire named out loud. The research task is getting to that layer, not stopping at the surface.
A Practical Research Process Before Any Creative Goes Into Production
Read 50 reviews for your product and 50 reviews for the top two competitors. Copy verbatim the most emotionally charged phrases you find and tag each with the emotion it carries. Do not paraphrase. Find three to five Reddit threads where people describe your problem category in their own words. Pull the specific phrases that appear across multiple threads, because repetition across strangers indicates shared experience rather than individual quirk. Check the comment sections of three competitors' Meta ads. Note every objection, every question, and every expression of skepticism. These are the doubts your creative needs to address. From this research, identify the two or three most specific emotional moments in the customer's experience of the problem. These become the hook candidates. The hook that tests best almost always comes from direct customer language rather than internal ideation, and the best performing Meta creative teams have inverted their time allocation: spending more time on research than on production.
The Short Version
You know what your customer wants to see in an ad by going to where they already talk about the problem: their own reviews, competitors' reviews, and Reddit threads describing the experience. The goal is verbatim customer language, not summarized themes, because the original phrasing is the hook. Emotional response to an ad predicts purchase intent more reliably than the informational content, and the emotional resonance comes from meeting the customer in the specific language of their own experience. Advize treats this research as a prerequisite before any creative production, not as optional context.
Conclusion
The most common creative failure in performance marketing is not bad execution. It is good execution of the wrong insight. Producing a well-shot video with a polished hook around a message the customer does not recognize as their own experience is expensive and invisible. The research to prevent it costs a few hours and significantly changes what gets produced. Advize builds this into every creative engagement because the performance data consistently shows the same thing: the ads that win are the ones that started from customer language, not brand language.