Outdated demographics 'waste budget and hit privacy'

Marketers and their agencies have been warned they are placing far too much reliance on outdated demographic segments that not only waste their budgets by targeting the wrong audiences, they risk fuelling concerns over data privacy to boot.

That is according to a study conducted by Adlook, which claims to have exposed major flaws in traditional targeting techniques that group audiences based on a combination of social and demographic traits, like age, gender, income, and lifestyle, highlighting the need for more accurate solutions that also align with the shift towards privacy-conscious marketing.

Firstly, 1,325 online consumers survey respondents were polled to determine whether they identified with the socio-demographic segments they were targeted for, based on third-party cookie-derived signals present in bid request data.

The study then compared the survey results to the bid request data to examine whether segments designed to be mutually exclusive (e.g., “Men” vs. “Women” or “Age < 34” vs. “Age > 55”) were accurately delineated, or if significant overlap occurred.

Control questions were also included to ensure data integrity by filtering out respondents who provided random or inconsistent answers.

The study highlights substantial challenges in the accuracy of socio-demographic targeting, with a key insight being the disconnect between targeting assumptions and actual audience composition.

In fact, for the commonly targeted segment of “Women 18-24,” for example, precision was found to be less than 20%. Among those targeted, 43% were men, 61% were over 24 years old (35% were above 55), and only 18% were women aged 18-24.

Meanwhile, in the “Mums” segment, 52% of the targeted users identified as men, and 62% reported not having children. Similarly, 67% of users in the “Parents” segment declared they did not have children.

Inaccuracies were also found across broader categories, with 40% of users categorised as primary residence owners actually being renters, and vice versa.

In addition, 67% of those targeted at a secondary school education level reported having a college or university degree, while half of the “Women” segment were men, and 76% of users targeted as “Married” declared they were not hitched.

Adlook chief product officer Mateusz Jedrocha said: “Legacy media-buying strategies and limited offline tools, like panels, force complex consumer profiles into broad categories such as ‘Women 20-44.’ In today’s digital age, this is unnecessary – brands can now target consumers based on real interests and behaviours, reducing wasted spend and reliance on outdated assumptions.

“This is the inevitable outcome of flattening nuanced audience insights into simplistic demographic assumptions that rarely align with real-world behaviour. The result is wasted ad spend and reduced campaign effectiveness.”

Even simple socio-demographic segments that should be mutually exclusive revealed significant overlaps, meaning the same users were inaccurately classified into multiple, conflicting categories.

For example, 35% of impressions were simultaneously eligible for both the “Women” and “Men” segments, while 55% fell into two or more age groups, demonstrating significant classification errors.

In addition, 28% of impressions were eligible for both the “Age < 34” and “Age > 55” segments.

Jedrocha added: “These findings expose a critical issue in digital advertising that too many are scared to call out around the lack of accuracy in socio-demographic targeting.

“But it’s not just about data accuracy; it’s about moving beyond outdated, simplistic audience definitions. Brands must adopt solutions that embrace the complexity of modern consumer behaviour while improving transparency, reducing costs, and being privacy-centred.”

TOPICS:data strategy data-driven marketing online advertising