Five AI prompts to stress-test your Gen Z thinking

Audience strategy is built through a series of decisions. The audience you define. The method of targeting them. The region you prioritize next quarter. Each one is only as sound as the consumer insights behind it.

As our 2026 Gen Z Report shows, generation only tells you part of the story. The rest comes from the layers underneath it: geography, life stage, behavior and more. When those aspects aren't thoroughly investigated, assumptions can fill the gaps and distort the audience picture without anyone noticing.

These five prompts surface an Agent Spark finding in 10 seconds. Each one stress-tests one assumption that generational thinking tends to carry. Pick the one closest to a brief on your desk this week.

Explore the prompts: 

  1. Cross-market audience comparison
  2. Cross-generation behavior check
  3. Audience bias check
  4. Audience label stress test
  5. Life stage audience split

 

 

1. Cross-market audience comparison

When you want to compare attitudes and behaviors across multiple countries.

Example prompts

Using Agent Spark:

For Gen Z in UAE, Brazil, Australia, and Portugal (latest year), scan personal finance attitudes and list the top 5 cross-market gaps.

Using LLMs via GWI data connector (including ChatGPT, Claude, Gemini, more): 

Using the GWI Data Connector, analyze Gen Z (born 1997 to 2012) in UAE, Brazil, Australia and Portugal across personal finance attitudes using the latest available data. Identify the five biggest cross-market gaps and return the percentage in each market for each attitude alongside the over-index difference.

What Agent Spark returns

Here are the top 5 cross‑market gaps for Gen Z personal finance attitudes across the UAE, Brazil, Australia, and Portugal (latest year), using comparable metrics where we have country overlap.

  • Finance app usage: Australia vs UAE – 34% used banking/investing/insurance apps last month vs 20%; gap: 14 pts.
  • Money‑driven mindset: Portugal vs UAE – 39% say they’re money‑driven vs 28%; gap: 11 pts.
  • Prefer experiences over things: Portugal vs Brazil – 62% prefer to spend on experiences vs 53%; gap: 9 pts.
  • Financial optimism (next 6 months): UAE vs Portugal – 77% expect finances to improve vs 69%; gap: 8 pts.
  • Investment interest: Brazil vs Portugal – 49% interested in investments vs 43%; gap: 6 pts.

Context on distinctiveness: On “money‑driven,” Gen Z in Australia are 60% more likely than average, while Gen Z in the UAE are 20% more likely, underscoring meaningful variance in financial motivation.

What this means commercially

Four markets, four different financial conversations worth having. The 14-point gap on app usage is interesting because it points to a channel decision more than a creative one. The money-driven mindset gap between Portugal and the UAE is the one that probably shifts your creative. A single global brief will likely overspend in some of these markets and underspend in others — and running this same prompt against your current global financial campaign would be a useful sanity check.

Adapt it for your audience

The Agent Spark prompt:

For [your audience] in [four to six markets you operate in] (latest year), scan [the attitude or behavior area you want to explore] and list the top 5 cross-market gaps.

The LLM prompt:

Using the GWI Data Connector, analyze [your audience] (specify the age range) in [four to six markets you operate in] across [the attitude or behavior area you want to explore] using the latest available data. Identify the five biggest cross-market gaps and return the percentage in each market for each attitude alongside the over-index difference.

 

2. Cross-generation behavior check

When the brief frames Gen Z as the opposite of every other generation.

Example prompts 

Using Agent Spark: 

Compare Gen Z, Millennials, Gen X and Baby Boomers on saying their smartphone is their most important device.

Using LLMs via GWI data connector (including ChatGPT, Claude, Gemini, more): 

Using the GWI Data Connector, compare Gen Z (1997 to 2012), Millennials (1981 to 1996), Gen X (1965 to 1980) and Baby Boomers (1946 to 1964) on the survey question "smartphone is the most important device for accessing the internet." Return the percentage for each generation.

What Agent Spark returns

How generations differ (likelihood vs others)

  • Gen Z are 6% more likely than Gen X to say their smartphone is their most important device, underscoring a stronger mobile-first mindset among younger users.
  • Versus Baby Boomers, the gap is pronounced: Gen Z are 53% more likely, Millennials 51% more likely, and Gen X 44% more likely to prioritize smartphones as their primary internet device. 
  • Between the two youngest cohorts, Gen Z are just 1% more likely than Millennials to put smartphones first, indicating near‑parity at the top end.

What this means commercially

The mobile-first audience is wider than most briefs assume: three generations sit within 6 percentage points of each other on this. If your media plan is paying a Gen Z premium for behavior that is genuinely shared across Gen Z, Millennials and Gen X, there is room to broaden the audience definition without changing much else. The same prompt run against any behavior your team currently treats as generationally distinct will tell you where else this might be true.

Adapt it for your audience

The Agent Spark prompt: 

Compare Gen Z, Millennials, Gen X and Baby Boomers on [a behavior you want to check is actually about age].

The LLM prompt: 

Using the GWI Data Connector, compare Gen Z (1997 to 2012), Millennials (1981 to 1996), Gen X (1965 to 1980) and Baby Boomers (1946 to 1964) on [a behavior you want to check is actually about age]. Return the percentage for each generation.

 

3. Audience bias check 

When you suspect the AI-generated audience profile is just US data wearing a global label.

The prompt

Using Agent Spark:

Compare US Gen Z and global Gen Z on wanting brands to be young.

Using LLMs via GWI data connector (including ChatGPT, Claude, Gemini, more): 

Using the GWI Data Connector, compare Gen Z (born 1997 to 2012) in the United States against global Gen Z on the brand attribute "want brands to be young." Return the percentage in each audience. 

What Agent Spark returns

Global Gen Z are more likely than US Gen Z to want brands to be perceived as young. Over the last 12 months, 29% of Gen Z globally said so, compared with just 18% of US Gen Z — an 11-point gap.

What this means commercially

An 11-point gap on one brand attribute is one data point, but it is a useful one. The more revealing exercise is running the same comparison across the other pillars in your global brand framework. The ones that travel will be reassuring. The ones that do not will be the ones worth knowing about before the next global campaign. Most US-built global playbooks have one or two attributes that quietly do not translate, and finding them now is significantly cheaper than finding them in market.

Adapt it for your audience

The Agent Spark prompt:

Compare US [your audience] and global [your audience] on [the brand attribute you're testing].

The LLM prompt: 

Using the GWI Data Connector, compare US [your audience] and global [your audience] (specify the age range) on [the brand attribute you're testing]. Return the percentage in each audience and the over-index difference.

 

4. Audience label stress test

When the brief defines the audience with one label and you suspect the label is doing too much work.

The prompt

Using Agent Spark:

What are the top 3 contradictions and myth busters about price conscious Gen Z consumers?

Using LLMs via GWI data connector (including ChatGPT, Claude, Gemini, more): 

Using the GWI Data Connector, identify the top 3 contradictions or unexpected behaviors within Gen Z (born 1997 to 2012) who identify as price-conscious. Look across categories like luxury purchases, premium products, brand loyalty and discretionary spending. Return the over-index for each unexpected behavior and a one-line note on why it contradicts the price-conscious label.

What Agent Spark returns

Here are the top 3 myth busters about price‑conscious Gen Z (global, last year).

  • Myth: They won’t pay extra for anything. Reality: 53% are willing to pay more for a brand they know. They trade up when value is clear.
  • Myth: Being price‑driven means they’re not into fashion or luxury. Reality: 41% say they’re fashion‑conscious, signaling aspiration alongside value scrutiny.
  • Myth: Deal seekers aren’t brand loyal. Reality: 51% say they’re loyal to brands they like, yet they still hunt value with 52% using discount codes/coupons—loyal but price‑conditional.

What this means commercially

Half of this segment will pay more for a brand they trust. Eighteen percent are using buy-now-pay-later. Forty-one percent describe themselves as fashion-conscious. The price-conscious label captures one financial behavior but does not capture the whole customer. If your category is currently competing on discount alone against this segment, there is room for a premium proposition to reach part of the same audience without competing in the same bidding war.

Adapt it for your audience

The Agent Spark prompt 

What are the top 3 contradictions and myth busters about [the single-label audience in your latest brief], who are [behavior the label would predict against].

The LLM prompt 

Using the GWI Data Connector, identify the top 3 contradictions or unexpected behaviors within [your audience] (specify the age range and any defining attribute). Look across categories that relate to your brand [enter name of your brand]. Return the over-index for each unexpected behavior and a one-line note on why it contradicts the assumed profile.

 

5. Life stage audience split

When the brief still treats one generation as one life stage.

The prompt

Using Agent Spark:

Within Gen Z, give me the 5 distinct examples of how does life stage influence what people are actually buying right now? Using different types of life stages, not just focus on one.

Using LLMs via GWI data connector (including ChatGPT, Claude, Gemini, more):

Using the GWI Data Connector, profile Gen Z (born 1997 to 2012) across five different life stages. Use household composition, employment status and relationship status to define the stages. Return five distinct examples of how purchasing behavior differs across those stages, with the percentage and over-index for each behavior.

What Agent Spark returns

Here are 5 clear, life‑stage examples within Gen Z showing how what they buy shifts “right now” (latest year), using different life stages.

1. Students (often not the household shopper) skew away from household staples and toward easy wins

  • A sizable share don’t handle the household shop 26% do not do the shopping and are 375% more likely not to shop.
  • They underindex on staples like laundry detergent (12% less likely), cleaning products (8.3% less likely), and toothpaste (9.2% less likely), while still buying quick eats like 76% snack foods.

2. Entering full‑time work (new household responsibility = broader baskets)

  • Full‑time workers are often the household buyer: 67% are the main food shopper.
  • Their baskets routinely include maintenance items at scale, e.g., cleaning products (73%) and laundry detergent/washing powder (70%).

3. Parenthood (infant care dominates incremental spend)

  • Gen Z parents’ baskets shift heavily into baby care, e.g., 26% diapers, 20% baby wipes, and higher‑care SKUs like 18% baby moisturizer.

4. Pet ownership (home care and fresh food needs rise)

  • Gen Z dog owners frequently buy home‑care and food categories: cleaning products (70%), laundry detergent (65%), and meat (70%), alongside everyday personal care like shampoo (73%); they even show a small lift on coffee 3% more likely than Gen Z overall.

5. Responsibility shift inside Gen Z (not just age, role matters too)

    • Compared with Gen Z in full‑time work, the broader Gen Z cohort is 34% less likely to be the main food shopper and 381% more likely not to do any household shopping—a strong sign that moving into full‑time roles triggers more routine household purchasing.

What this means commercially

Life stage produces bigger basket differences than generation does. The Gen Z student and the Gen Z full-time worker live in the same category and shop in completely different ways. If a current brief has not specified which life stage within Gen Z it is targeting, the media and creative teams are likely reading it as all of them — which usually lands as none of them. One of the fastest improvements to most underperforming Gen Z campaigns is simply adding a life stage filter at the audience definition stage.

Adapt it for your audience

For Agent Spark:

Within [your audience], give me [insert the number you want] distinct examples of how life stage influences what people are actually buying right now. Use different types of life stages, not just one.

For LLMs: 

Using the GWI Data Connector, profile [your audience] (specify the age range) across [specify the number you want] different life stages. Use household composition, employment status and relationship status to define the stages. Return five distinct examples of how purchasing behavior differs across those stages, with the percentage and over-index for each behavior. 

What to do next

Pick the prompt that maps to the assumption you most often see in your team's briefs. Run the example in your preferred environment (Agent Spark, the GWI platform, or your LLM via the Data Connector). Then run the adapted version with your own variables. The exercise takes five minutes and will tell you whether your last campaign brief held up against the data.

 

 

Step into the future of consumer research