3.3: Big data, analytics, and audience targeting

Figure \(\PageIndex{1}\): Data analytics graphic illustration. (Free-to-use from Unsplash; Choong Deng Xiang)

Media writers and producers have access to vast amounts of data and the ability to analyze that data for all sorts of creative purposes.

Journalism

For journalists, this means they have more powerful tools to practice investigative journalism. This article about the use of AI in analyzing big data to facilitate investigative journalism notes how journalists tracked Israeli bombings of Gaza using AI. "The paper’s journalists used artificial intelligence to track satellite images that showed more than 200 craters in densely populated civilian areas, which experts say are likely caused by 907-kilogram (2,000-pound) bombs." The article goes on to state that AI might enable more journalists even in smaller news organizations to do the types of investigative journalism only previously possible in large newsrooms with large staffs.

Investigative Reporters and Editors, a leading group that supports the practice of using data to report on impactful stories, breaks the process down to four essential steps including asking a question, gathering the appropriate data, cleaning and analyzing the data, and doing further reporting and writing of the story.

For more information about how data journalists are working to effectively and ethically use AI, read this article from the Global Investigative Journalism Network.

The data gathering and analytics tools available to journalists are too numerous to mention here, and new ones are being developed all the time. Suffice it to say that students interested in data journalism will likely want to add learning the computer programming language Python and the R statistical package to their repertoires.

A good list of resources for journalists interested in doing data journalism can be found at this page by the Public Media Alliance.

Advertising

For advertisers, access to "big data" and the tools to analyze much of that data means they can target audiences not just by demographic characteristics but based on behavioral tendencies and likelihoods. They have the ability to target not just women in their late 20s who own a home but women in their late 20s living in certain ZIP codes with certain income levels, with one or more children, who have visited an upscale clothing store in the past two weeks, who use Pinterest, who appear to be renovating their kitchen and are likely to buy new appliances in the next 4-6 weeks. That level of hyper-targeting is valuable in terms of reaching those at the opportune time when they are making a purchase decision, but that same level of persuasive power can be disruptive, even destructive in electoral politics, which is, of course, heavily influenced by advertising. 

For example, Cambridge Analytica shared Facebook user data with the Trump campaign in 2016. Before falling back on trope that "all's fair in love and politics," consider this statement from a Vox article on the scandal: "Facebook allowed a third-party developer to engineer an application for the sole purpose of gathering data. And the developer was able to exploit a loophole to gather information on not only people who used the app but all their friends — without them knowing." The data was used to build profiles for millions of voters, which could be used to target their behavior. Cambridge Analytica also facilitated an illegal relationship between a Super PAC and the Trump campaign, according to this article from Campaign Legal Center.

If journalists seek to use big data to conduct investigations and tell stories that inform voters, their efforts may pale in comparison to the persuasive influence of big data and psychological profiling in politics. Advertisers in the coming decades can expect to continue to grapple with the ethical challenges of audience targeting which often stand in opposition to the power of targeting audiences by gathering, analyzing, and acting on incredibly personal data.

Audiences appreciate relevant messages. As the Center for Media Engagement at the University of Texas at Austin puts it, "if advertising is unavoidable, it makes sense to at least have advertising that is relevant to us. In fact, such targeted advertising can be helpful in many ways, rather than just tolerable." But the same article sums up our fears as consumers that the entire practice may not be ethical simply because it requires deeply personal, previously private information to work.

Public relations

For public relations professionals, access to vast amounts and types of consumer data and the ability to conduct advanced data analytics presents the ability to not only test messages but to target that testing like never before. Brands and institutions want to know everything from which platforms their intended audiences use most to which word choices are most likely to be effective.

Definition: Message Testing

In marketing and public relations, message testing refers to the practice of presenting different versions of a persuasive message to targeted audiences in order to gather feedback about how receptive they are to the premise, structure, wording, length, and platform of the message as well as other more detailed qualitative responses in each case. In simple terms, message testing measures if audiences understand the message the organization is trying to convey and what they may like or dislike about the message and how it is presented.

The way messages from brands and institutions meant for mass audiences are written is important. Creative success in this area can make a significant difference in what people think about a brand, industry, or institution, and this can equate to greater power, influence and success in society, but there is more to messaging than creative writing and testing messages with targeted focus groups in person, on the web or on mobile devices.

Consider this quote from a paper put out by the Institute for Public Relations: "A common misconception holds that 'messaging' is a purely creative endeavor based on clever phrasing, brilliant visuals and edgy disruptive execution. While creativity certainly plays an important part, successful messaging is as much science as art; and the required science actually enhances the creative process by focusing resources on those messaging opportunities with the highest potential. All of these tools can mean a more catered, personal media consumption experience, but they can also mean the death of privacy as an individual's media use data can easily be tied to their movement, shopping, and purchase habits as well as their proclivities."

The ethical concerns in public relations are similar to those in advertising with an added caveat. In addition to working with consumer data, PR professionals often gather large amounts of mass media and social media content to analyze how much their organization is being talked about and what is being said. While it is possible to track the success of a message or a campaign, PR professionals should not mislead their partners, internal or external, into thinking that they can control the narrative at all times. 

Human communication is influenced by numerous variables, and no PR writer or firm can predict them all.

Ethics of audience targeting

The ethics of targeted advertising, as discussed in this ethics case study, have been discussed at length in recent years and touched on in the context of journalism, advertising, and PR in this chapter section. From the consumer's perspective, it comes down to the convenience and usefulness of targeted advertising versus the potential damage that could be done if their privacy is breached. There have been astonishing cases of trauma made worse by hyper-targeted ads. The example in the case study linked just above mentions a woman who had a miscarriage but continued to see maternity ads on her social media feeds. This type of targeted messaging was both too aware and not aware enough at the same time. Advertisers had just enough information about this individual to enhance and extend her trauma.

Not only are circumstances like this bad for business, in the long run they are likely to be bad for society. Media writers must exercise extreme care and caution when working with data to target messages. The people on the receiving end are not amorphous collections of audience members. They are individuals who can be helped or hurt by what you write.

Thought exercise

Do social media apps seem to be able to predict your physical or mental state, your behavior and preferences? This is because cross referenced data including essential demographic information about you, to behavioral data about what you do online, to data about where you go in the real world, what you believe in, what you do for recreation, what you buy, and even how you sleep can all be used to predict your behavior. Consider carefully what you share and how you interact online.

The University of West Virginia's Information Technology Services has a list of suggestions for individuals modified to fit this space:

1. Don’t overshare information on social media.

2. Collect only essential information in your professional capacity as media workers. 

3. Use incognito or private browsing in your personal internet use.

4. Limit app access to personal data by opting out of inessential data sharing choices.

5. Change your privacy settings and check them often for apps, websites, and devices you use often. 

For more help, ITS at West Virginia writes: "The National Cybersecurity Alliance has a helpful Manage Your Privacy Settings page with links to dozens of popular services."