We can’t argue that AI, or artificial intelligence, has been heavily infiltrating all sectors of marketing. It’s changed the discipline and arduous research in media planning, the structure and management of search engine marketing, the creation of ads, the optimization of media investments, and even the surfacing of insights. Here are just some of the ways we’ve been using AI in the media department, looking at three of our largest channels of investment:
SEM: Long gone are the notions of how we used to set up campaigns, ad groups and match types – especially the long hours analyzing positions and costs-per-click to maximize spend. Keyword creation is automated, and even adding new keywords with no historical performance is informed by AI: By teaching the machine concepts like campaigns, ad groups and keywords, plus how these relate, we are providing Google with the same thinking we use to make reasonable guesses. So the system can now automate bid management and might set a similar bid to other keywords in the campaign, because it knows that campaigns tend to have keywords with something in common. By having a score for how likely it is that each search will translate into a conversion, automated bidding products inside AdWords can “think” through many more dimensions (like geolocation, hour of day, device or audience) that might impact the likelihood of a conversion than a person could. Google’s AI also surfaces optimization suggestions and additional keywords based on machine learning.
Paid Social: The myriad audience targeting options in paid social can be overwhelming, but not with AI. Through a pixel and machine learning, AI will identify the best possible interests and audience affinities, including the not so obvious. By placing this pixel in key areas of your website, especially the cart completion process, you can form look-alike models against your best customers and then analyze the segments to extract audience insights. AI also informs bidding and budget allocation to maximize spend and conversions across targeting type, target audience, creative and creative sets, device, and numerous other signals. Better yet, this is all done in real time. So no more are the days of deep analysis and hours of hand-optimizing campaigns.
Programmatic Media: Media planning used to be a strenuous process: We would spend hours researching and then analyzing audience demos, psychographics, websites visited and media consumed. From there, we would plan our media to include reach and frequency goals, and specific allocations across tactics. This process took hours, if not weeks. Once the plan was done, it would take another couple of weeks to RFP vendors, negotiate proposals, then place the media, including the trafficking of ad creative. This whole process would take a month or two. With AI, this whole process can literally take a couple of days [if we wanted it to]. Using a pixel placed on the website, machine learning will surface every single audience attribute we need to target based on analysis of website traffic and conversions. The audience segments are automatically available for targeting. Media planning? Done. Using AI, specific tactics are also recommended, including spend suggestions. As more media is driven to the website, the pixels continue to learn and recommend additional audience strategies, tactics, and of course, optimization suggestions. The optimizations in programmatic can be tedious if done manually, with hundreds of signals and data points that can be bid up or down in various combinations – this would take a human days or weeks to implement if leveraging all signals. Using AI, optimizations happen both in real time and also in suggested form, which traders can accept or reject. Creative is also optimized to hit specific goals, whether reach, engagement or conversions are desired. We’ve even automated the creation of the ads themselves. Using a bank of images, text and calls to action, creative builds on itself with audience signals and optimizes variations based on multivariate testing – all with the end goal of exceeding performance. And who doesn’t want that?
It’s no doubt that our world continues to evolve, and AI will not stop. There are even more ways we’ve been experimenting with AI, and even some forms of AI we’ve strayed from [insert job security concerns here]. Kidding aside, with all the AI at our fingertips, and even more AI capabilities in the future, what will our role be? An article from Search Engine Land sums it up nicely:
1. We will teach the machines to learn.
Now that machines can learn, they certainly will surpass humans, right? The reality is that machine learning is still very dependent on humans. We program the algorithms, we provide the training data, we even manipulate the training data to help the machine get it right.
2. We will provide the creativity machines lack.
Humans are still good at creative strategy – putting old ideas together in new ways and testing the results. The reason we don’t have Google’s computers writing all the ads for AdWords is that they all would end up looking the same – and then they would stop evolving because the machine would no longer have any variations to test.
3. Agencies will be the pilot who averts disaster.
The problem with many systems built today is that they have narrow goals that can fail due to self-reinforcing feedback loops that can cause a downward spiral:
bad performance → bid down a bit → even worse performance → bid down some more → doom!
And sometimes your job as copilot is to see something that’s not there but that should have been. When you look at what leads to a conversion because you want to do more of that, maybe you should also ask what doesn’t lead to a conversion and do less of that. For example, high shipping fees may tank your conversion rate, but you wouldn’t find this out if you asked the wrong question (and this data is not in the AI to learn from, like a lot of other variables that exist outside the machine).
4. Agencies will have the empathy machines lack.
Empathy is understanding the nuances of your client’s business (which will help you come up with new ideas to test), understanding their fears about PPC, understanding their frustrations with the last account manager and so on. All this will help you have a more productive relationship with them.
A couple of parting thoughts, from two experts discussing AI and its implications for humans in the very near future:
Marina Gorbis, executive director of the Institute for the Future, said, “Without significant changes in our political economy and data governance regimes [AI] is likely to create greater economic inequalities, more surveillance and more programmed and non-human-centric interactions. Every time we program our environments, we end up programming ourselves and our interactions. Humans have to become more standardized, removing serendipity and ambiguity from our interactions. And this ambiguity and complexity is what is the essence of being human.
Baratunde Thurston, futurist, former director of digital at The Onion and co-founder of comedy/technology startup Cultivated Wit, said, “For the record, this is not the future I want, but it is what I expect given existing default settings in our economic and sociopolitical system preferences … The problems to which we are applying machine learning and AI are generally not ones that will lead to a ‘better’ life for most people. That’s why I say in 2030, most people won’t be better due to AI. We won’t be more autonomous; we will be more automated as we follow the metaphorical GPS line through daily interactions. We don’t choose our breakfast or our morning workouts or our route to work. An algorithm will make these choices for us in a way that maximizes efficiency (narrowly defined) and probably also maximizes the profitability of the service provider. By 2030, we may cram more activities and interactions into our days, but I don’t think that will make our lives ‘better.’ A better life, by my definition, is one in which we feel more valued and happy. Given that the biggest investments in AI are on behalf of marketing efforts designed to deplete our attention and bank balances, I can only imagine this leading to days that are more filled but lives that are less fulfilled. To create a different future, I believe we must unleash these technologies toward goals beyond profit maximization. Imagine a mapping app that plotted your work commute through the most beautiful route, not simply the fastest. Imagine a communications app that facilitated deeper connections with people you deemed most important. These technologies must be more people-centric. We need to ask that they ask us, ‘What is important to you? How would you like to spend your time?’ But that’s not the system we’re building. All those decisions have been hoarded by the unimaginative pursuit of profit.”