The positioning line the room loved never got tested, and the brand was better off for it

The situation

A household cleaning brand was preparing its annual relaunch and had three candidate taglines on the table, all built around the idea of speed: "5-minute clean," "done before your coffee's ready," and "less scrubbing, more living." In the internal workshop, the marketing team gravitated toward the third option. It tested well in the room, and more than one person there had personally complained about scrubbing being a chore.

The insights team had two existing research studies sitting in a shared drive: a usage and attitude study from the year before, and a more recent brand tracker. Both were full of relevant consumer data that nobody had revisited since the original presentations. Before committing budget to a full quantitative ad test the following month, which would take weeks and cost real money, the team wanted a faster way to check whether the room's favorite line actually held up with real customers, or whether it was simply the line that sounded best to the people who'd spent a month writing it.

What a "digital twin" means here

PersonaAI works by building AI-driven profiles, called digital twins, from a brand's own existing research data. Each twin represents a specific group of real consumers found in that data, built from how those people actually answered survey questions, not from assumptions about who they are. A researcher can then have a conversation with a twin almost like interviewing a focus group, asking it questions and getting answers grounded in the original survey responses behind it.

This matters because it let the insights team test all three taglines against real customer data within days, instead of waiting for a traditional study to field, close, and get analyzed.

What they put in

The team fed PersonaAI two existing studies:

A household cleaning usage and attitude study covering cleaning occasions, routines, which brands people already used, and what made them trust a cleaning product

A more recent brand tracker covering brand consideration, brand image, and whether a product felt like it was "made for people like me"

They described the objective directly to PersonaAI, specified the category as household surface cleaners, focused the analysis on the people who actually do the weekly cleaning in their household, and entered all three tagline options as the thing to be tested.

What PersonaAI created

From that existing data, PersonaAI built three digital twins, each one representing a distinct group of real respondents found in the survey data, complete with how many respondents backed it up, how well it matched the brief, and how much relevant data existed to support its answers.

The team moved forward with two of them:

• Time-Squeezed Cleaners: juggling work and household responsibilities, fitting cleaning in around everything else, with strong underlying data on their routines and when they clean
• Standards Keepers: clean thoroughly no matter how long it takes, judge a cleaning product by the result it delivers rather than how fast it works, with strong underlying data on what makes them trust a product

A third twin, representing younger, first-time buyers of the category, was also created but set aside. It matched the brief reasonably well, but there wasn't enough underlying data about this group's established cleaning habits to trust its answers, and the brand wasn't targeting younger first-time buyers in this relaunch anyway.

What they asked

Rather than simply asking each twin "which tagline do you like best," which tends to produce shallow, surface-level preferences, the team had a real conversation with each one. They presented each tagline individually and asked what it implied about the product, what it assumed about the person's week, and what they would expect the product to actually do once they tried it. Then they brought both twins into the same conversation to see exactly where their reactions diverged.

What the conversations revealed, and what they couldn't answer

The line the marketing team loved was actually a liability with one key group.

Standards Keepers read "less scrubbing, more living" as a warning sign, an implication that the product might not clean as thoroughly as they needed. The data behind this twin showed that, for this group, the physical effort of scrubbing was tied to visible proof that a surface was actually clean, not something they wanted removed from the experience. Time-Squeezed Cleaners liked the emotional appeal of the line but couldn't articulate what the product would concretely do differently if they bought it.

"5-minute clean" made a specific, testable promise the other two lines didn't.

Both twins responded to the number itself as a concrete commitment rather than a vague feeling. Time-Squeezed Cleaners saw it as a direct solution to the time pressure that showed up throughout their underlying survey data. Standards Keepers wanted that speed claim backed up by a result-based claim alongside it, not offered as a standalone promise.

"Done before your coffee's ready" fell flat with both groups.

Neither twin connected this line to anything in their actual daily routine. It came across as a cute turn of phrase to Time-Squeezed Cleaners and did essentially nothing for Standards Keepers.

Whether "5-minute clean" would actually be believed on the shelf remained an open question.

The existing brand tracker had measured general brand perception, not reactions to this specific numeric claim. PersonaAI was clear about this limitation rather than guessing at an answer the underlying data couldn't support. That gap became the exact question the team decided to answer in the quantitative test that followed, rather than something PersonaAI could resolve on its own.

What changed

The team's favorite line was removed from consideration before any creative production began. The relaunch plan was rebuilt around "5-minute clean," paired with a result-based proof point specifically designed to satisfy Standards Keepers, rather than producing two separate commercials for two different audiences.

• Before: Test all three positioning routes in a full quantitative ad test and see which one scores best.
• After: Test a single, sharper question among primary weekly cleaners: how believable is the 5-minute claim at the shelf, and does adding a result-based proof point win over Standards Keepers without watering down the message for Time-Squeezed Cleaners?

The quantitative ad test still went ahead as planned. It ran with one strong, data-informed route and a focused question to answer, instead of three options and a simple popularity contest.