Like a lot of industries, artificial intelligence (AI) is changing advertising right before our eyes. Today, AI is getting attention for writing emotive TV scripts, targeting smart ads and using facial recognition to recommend products based on personal preferences.
In each of these scenarios, it’s helping advertising professionals do their jobs with intelligence backed by solid data; much more than humans have the time or capacity to analyze. And while there’s some debate over how far some of these technologies should go and whether they violate privacy and keep customer data secure, there’s a lot of excitement over its potential.
So how is AI making an impact on the television advertising industry – and the people in it – today?
Digital teams at agencies were the first to adopt machine learning and AI for automated bidding for digital ads.
The August 2019 Nielson Local Watch Report said 65% of U.S. homes have TVs that are connected to the internet which means the growth of connected devices and smart TVs is skyrocketing. With that change comes AI allowing for more targeted, relevant programming that promises to improve audience experiences.
Here are two examples:
In perhaps the most publicized AI news in our industry, Lexus claimed to produce the first commercial written by IBM Watson, an AI tool, and directed by an “award-winning human.” Before writing the script, Watson had to learn a few things:
Lexus’ director for the spot said he was surprised when he saw the completed script: a cohesive, melodramatic and data-based storyline. Every line of script and sound effect was accompanied with a footnote that explained why it was included based on enormous amounts of data – more than a human could probably retain or consider. You can see the full commercial with scene-by-scene explanations of what AI discovered at Adweek.
While this is exciting, it can be unsettling to consider the negative side of AI. Will artificial intelligence replace creatives or writers? While the completed ad was good considering it was written by AI, the advertising world didn’t consider the completed ad perfect.
The plot line takes a vehicle from its engineering environment, into a beautiful world and then into a test dummy environment. It’s peppered with shots of news reporters, media hype and people watching what is sure to be the vehicle’s destruction on TV.
However, having a script written by AI based on communication goals and backed by more data than humanly feasible to produce in a short time-frame can prove helpful to creatives who ultimately have the final say about what people want to watch.
While creative by AI makes for exciting headlines, some of the best technology is the AI that works quietly in the background. Technology advances promise to help media planners with budget allocations based on their media mix modeling.
When media planners can use technology to aggregate a variety of data sets from different channels – both traditional and digital. Once data on internal (product launches, promotions) and external factors (seasons, trends) becomes more widely available across all campaigns, AI will be able to optimize and reoptimize allocations based on attribution data.
With technology powered by AI, media planners will be able to free themselves up for other high value activities.
Recent advances made in 2019 are now helping media buyers with their biggest pain points: ratings forecasts.
In a month, buyers can spend a full week doing the research to manually forecast ratings. Leveraging artificial intelligence loaded with data from Nielsen and Comscore, they can get accurate ratings – and a big portion of that week back.
AI allows buyers to efficiently forecast ratings without sacrificing accuracy. And of course, more accurate ratings save agencies money by reducing lost opportunities and the need to go after makegoods.
Ad tech powered by machine learning also provides buyers with greater transparency and automated advertising transactions. Buyers and their television sales reps don’t have to spend weeks availing, estimating, stewarding, chasing under-delivery weight, managing makegoods, and clearing up billing discrepancies. Instead, buyers harness automation to find more time to focus on higher value activities.
While I haven’t seen the application of facial recognition in the TV world yet, and this application of AI gets a bit futuristic in application, it deserves a mention. Facial recognition is making waves as brands scramble to use it in retail environments.
According to a recent piece in Forbes, retailers are leveraging AI to recognize shoppers and show them products they might like, based on what they’ve looked at during past visits or “sessions.” While privacy and security are serious concerns, AI may be used to show specific products in digital store displays.
Or, AI could provide store associates with customized suggestions for individual shoppers through eye-wear to help them make better product recommendations – and sales.
Chains like CaliBurger are using facial recognition to show favorite meals to people as they walk up to its ordering kiosks. The brand says this is a step toward replacing credit card swipes with face-based payments in the future.
While we can only imagine all the applications for AI in the advertising industry, advances are being made every day to make our lives better. Sometimes, we’re using tools powered by AI and we don’t even realize it. Think about any new technology solution you’ve tried or come face-to-face with lately – is it powered by artificial intelligence?
As ad industry professionals, it only benefits us to stay curious about new tools being developed for our use. And, to consider how the technology will help the consumer, as well as the implications for how AI could affect their privacy as well as the security of their data.
With AI and machine learning, we have access to information that we traditionally could not process at a higher level. As we continue to pair insights from AI with our own expertise and human experience, the advertising industry will only become stronger and more effective.