A meta-analysis published in the Journal of Advertising Research reviewed 53 studies involving more than more than 11,000 participants to determine whether personalised ads actually outperform their generic counterparts and what makes them work — or not.
The researchers found that perceived relevance fully mediated the persuasive power of personalisation. That is, personalised ads worked because people felt the message was genuinely intended for them. By contrast, perceived intrusiveness — the idea that personalisation might feel creepy or overstep boundaries — had no significant effect on ad persuasiveness. Consumers noticed it, but didn’t change their behaviour.
The overall persuasive effect of personalisation was modest, but statistically significant. Across the 114 effect sizes analysed, personalised ads led to more favourable attitudes and stronger behavioural intentions than non-personalised ones.
The implication for media planning is that personalisation isn’t risky, as long as it doesn’t shout about itself. The data suggests advertisers are more likely to gain than lose when they personalise, provided they do so with subtlety and intent.
There are, however, caveats. The most effective ads were those that used actual personal data, instead of imagined scenarios or generic demographic cues. Studies using real browsing behaviour or search histories delivered significantly stronger effects than those asking participants to imagine a personalised ad (which had no significant effect at all).
This distinction matters for campaign design. Scenario-based testing, which is often used to simulate personalisation in concept testing or academic studies, may underrepresent its true impact in market.
Similarly, covert personalisation — the kind that draws from online behaviour without making it obvious — was more effective than overt strategies that called out a person’s name, location, or status. And while real brands outperformed fictitious ones, the difference wasn’t statistically significant.
Personalisation also worked better in commercial advertising contexts than in non-commercial ones. Ads promoting products or services were more likely to benefit from personalisation, while those in political or charity contexts showed no persuasive advantage, possibly due to greater sensitivity to perceived manipulation or autonomy loss in those domains.
The study adds weight to the theory that effective personalisation is about resonance, not just recognition. It supports the congruity and self-referencing theories, which suggest people respond better to messages that feel aligned with their sense of self.
The analysis also found no evidence of publication bias across the 114 effect sizes tested, strengthening the reliability of its findings. The effects of personalisation were consistent across demographics, as age, gender and whether participants were students didn’t change the results.
For marketers, this means reframing personalisation as a relevance engine, not a targeting tactic. The goal isn’t to prove the ad knows who you are, but to feel like it was made for you.
How Persuasive Is Personalized Advertising? A Meta-Analytic Review of Experimental Evidence – Tien Ee Dominic Yeo, Tsz Hang Chu & Qiqi Li.