Horizontal image of a crowd of people with one fist in the air | iProspect Global

One of the most important considerations in building experiences that truly capture attention is to make sure everyone feels included. It may sound obvious but is yet to be a reality for many population groups, despite their growing economic influence (in the US alone, the buying power of racial and ethnic minority groups is close to four trillion dollars). Like attention, diversity and inclusivity should not be taken for granted and require continuous effort. Brands have a critical role to play in that space. According to dentsu and SeeHer, 81% of US consumers agree media plays a crucial role in shaping gender roles, but less than a third (32%) think media usually portray women accurately.

 

A pivotal moment in time

2020 has been pivotal in allowing many to realise for the first time the disparities endured by minorities. The COVID-19 pandemic has hit minorities hardest and several tragedies in the US such as the murder of George Floyd have spotlighted systemic discrimination against people of colour. The global public outrage that followed made waves across all aspects of society, including consumption, through movements such as #BlackoutDay2020, and brands, with many major companies publicly confronted for the lack of diversity in their boards.

Brands can no longer neglect diversity and inclusivity. It may lead to uncomfortable realisations, difficult conversations and mistakes along the way, but embracing inclusivity is now an imperative for brands to become or stay relevant to population groups that have been overlooked for too long.

 

There has been progress, but the road to diversity and inclusivity is still long.

According to the iProspect 2020 Global Client Survey, 96% of marketers now believe inclusive marketing is important, most of them (63%) seeing it both as a moral imperative and a business opportunity. Interestingly, the percentage of marketers who believe it is only important for the business potential it represents is decreasing (-7% YOY), while the percentage of respondents seeing it as important solely on moral grounds is increasing in the same proportion. It could be a sign that more companies are growing more comfortable with embracing a societal role.

Reflecting on the campaigns they ran over the last six months, 55% of marketers declare they featured people from a different ethnicity/caste than the dominant one in their market at least once in a positive way. This increasing figure (+3% YOY) could hint that the increasing visibility of movements like Black Lives Matter progressively change perceptions and actions around social justice.

However, not all population groups follow the same trend. For instance, we observe the percentage moving down for people with disabilities (-4%), which makes them underrepresented compared to the global share of people living with some form of disability (15%).

These figures show that, although inclusive marketing is gaining traction, there is still a lot left to do to improve the visibility of minorities in media and advertising and to build authentic stories that accurately reflect minorities’ experiences. Although there is no inclusive silver bullet, there is definitely a starting place: the diversity within the organisation. Building inclusive marketing campaigns requires inclusive marketing teams wherein diverse voices can be heard. This is a sine qua non condition for brands to resonate with all their potential customers.

 

The fight against bias

Bias is everywhere in society, and technology and data are no exception. Far from being neutral, technology and data can perpetuate and exacerbate disparities, prejudice, and discriminatory patterns. Many fields at the core of data marketing can be subject to bias, from insights to algorithms to targeting and performance analysis.

 

However, tackling bias is a difficult challenge as it can appear in many forms and on multiple occasions.

  • It can be intentional (e.g., a decision to exclude a certain ethnic group from a campaign) or unintentional (e.g., combining multiple data targeting dimensions such as income and location that could de facto exclude minorities living in certain areas).
  • It can play out at an individual level (e.g., due to the personal values of an employee) or at a company level (e.g., due to a lack of diversity in a product development team).
  • It can stem from organizational inertia, negligence, or ignorance. (e.g., if a community was never targeted by the brand, a predictive model using historical data can incorrectly conclude the community is less likely to purchase its products and thus that the brand should not target this community in the future).
  • It can emerge from within or be imported into the organisation (e.g., by using incomplete or poor-quality data from third parties).
  • It can be a combination of the factors listed above and more, which makes bias detection even harder.

Because of the omnipresence of bias across organisations, it is crucial for companies to actively seek to identify and eliminate bias through various, overlapping strategies.

 

According to the iProspect 2020 Global Client Survey, a majority (52%) of marketing teams are now diverse and inclusive (+9% YOY). This is an important milestone as minority groups should not only be visible in consumer personas, but also around the marketing table.

 

Guidelines and best practices on diversity and inclusion are the second option favoured by brands (44%), followed by multidisciplinary bias training (32%, +8% YOY).

From the results, it seems that brands are increasingly concentrating their efforts on the people within the organisation (e.g., through recruitment, training, committees) to drive change, rather than relying on external partners or focusing on processes. However, the latter should not be overlooked. An external perspective can be useful for organisations to look beyond their filter bubble (e.g., through advanced analysis of data and algorithms, hiring process testing, or consulting on website design accessibility), and a systematic audit of product development and marketing campaigns can prevent oversights.

 

Marketers should keep in mind these additional considerations to reduce bias specifically in data marketing:

  • As with the move toward privacy and Intelligent Tracking Prevention (ITP), the fight against bias calls for better data. Question the quality of the data you use in your marketing campaigns to prevent bias from spreading into data-powered activities.
  • Your teams are not the only ones who should have frequent bias training. Your predictive models should be retrained regularly as well to learn from new, real-world data. Tools such as the AI Fairness 360 by IBM and The Linux Foundation can help you understand the bias in machine learning models.
  • Keep investing in diverse teams. Human critical thinking is an essential safeguard to remove blind spots automation can generate.

 

This article is excerpted from the report Future Focus 2021: Brands Accelerated.

Download it now  for key insights on how brands can make the most of brand and performance to accelerate their growth.