Should AI have a role in your digital marketing stack? With a wide variety of promising AI-powered tools on the market, more digital marketers are re-examining their priorities. So whether you’re new to the concept of AI, or are already exploring the right AI-powered digital marketing tools for your organization, here’s some expert guidance from Maria Flores Portillo, General Manager for Amobee.
Automation and better decision-making can hopefully free us all to spend more time doing more creative and healing work.
Flores Portillo oversees growth in the EMEA region (Europe, the Middle East and Africa) for advertising platform Amobee, a leading provider of forward thinking AdTech solutions. She brings a wealth of AI technology experience from working with companies like Google and Videoplaza. Most recently, she was instrumental in the launch of Persado in the UK, bringing AI to brands to personalize creative. A self-described nerd at heart, she holds an MSc in Telecommunications Engineering from Universidad Publica de Navarra and University of Surrey.
Can you give us your perspective on the current landscape of AI technology for digital marketers? The broad use case categories seem to be customer experience, data insight and personalization.
I agree that those could be the broad categories, though we could probably make a case against them, as they are all very much interlinked.
The basic thing to understand, before entering the specific use cases, is that what AI-related technologies have allowed most vendors to do is to make decision-making more effective and efficient. In the past, a lot of the decisions around personalization or ad placement would be based on rudimentary rules that humans would create on the back of limited data analysis. Today, algorithms can automate and massively enhance this decisioning, and the number of front-end use cases on top of that is really limitless.
Better data insights through AI is a natural consequence due to the sheer amount of data processed through new techniques. Things like customer churn prediction or gamification suggestions for customer experience are extremely useful to marketers.
AI Hype
However, a lot of the AI hype in marketing has been around personalization, which in itself is a broad category. Companies such as Emarsys, Tinyclues, or Dynamic Yield are helping marketers better define segmentation, targeting, packaging, content generation, etc. Adtech vendors are leveraging machine learning algorithms to make much better predictions of where to find the right user for a relevant ad, which increases media effectiveness. I now work in this category at Amobee. All of these use cases are powerful and marketers are noticing; Dynamic Yield was acquired last year by McDonald’s, for instance.
Customer experience, which is really what should underpin our motivation as marketers, is very much related to this notion of personalization, though it tends to be focused on more interactive transactions between a brand and a customer. As a category, it typically also makes use of Image and Natural Language Processing/Generation. A very simple chatbot on a website can be an extremely effective way to both apply cost savings and provide a phenomenal user experience if done well. As robotics, AR and VR keep advancing, arguably the most exciting category to watch for the years to come.
What hesitancies are you seeing among marketers when it comes to adopting AI tools? Are there any common myths that you have to address?
I would say the notion of AI-related technologies for marketing is a lot more accepted today than five years ago. There is, however, still a fair amount of hesitation. This tends to come from three different places.
The first one is that the term itself is so hyped that it almost puts a buyer on the defensive, not really trusting whether there is any truth behind the promise. Once they spend enough time with different vendors to better understand the different applications, and how a specific technology can resolve one of their very specific problems, progress can be made.
The second one is that many marketers believe that only very large organizations can make AI work, both because it is costly and because it demands a vast amount of first party data. While AI can be costly, you can also start small and rely on cross-industry data. POCs are very common to assess ROI, and most AI vendors can be great consultants on the data strategy itself.
AI control
AI in itself sounds scary both from the perspective of things potentially going wrong and losing our jobs. On the former, I am a strong advocate for controls around AI, either at a tactical level (i.e. a human needs to approve content before it gets sent) or at a broader regulatory level. Most commercial AI applications out there have a fairly significant element of human monitoring, and it is a matter of discussing the right level of control for your organization with your vendor. On the latter, such as machines replacing humans in the workplace, it is a very natural reaction to change. We will get over it, just like we did with programmatic advertising or CRMs. Automation and better decision-making can hopefully free us all to spend more time doing more creative and healing work.
My advice to marketers would be to embrace the change, embrace learning on these new tools and be experimental. There is a lot of value behind the hype!
A lot of the more well-known uses of AI are associated with massive brands, like Google or Amazon. Do AI-based solutions have a place for digital marketers with much more limited budgets for their tech stacks?
Technology doesn’t need to be expensive; it needs to be ROI effective and fit for purpose. Many small online retailers use chatbots, for instance, to reduce their need for a large customer service team. And most marketing tools will use a number of machine learning algorithms or NLP/NLG (Natural Language Processing/Natural Language Generation) disciplines to be more effective, even if they are not labelled “AI.” At Amobee, we serve a vast range of clients, from relatively small retailers to Fortune 500 companies. They are all leveraging AI-related technologies through us.
What is harder to achieve for smaller companies is the ability to leverage a vast amount of first party data. Algorithms are as good as the data you feed them, and someone like ASOS, for instance, will always have a scale advantage in terms of personalization, against a much more niche online retailer. The fact that scale is a business advantage is not new, obviously, as it creates tremendous leverage in buying/selling power, but first party data as a competitive asset is becoming more and more valuable.
What’s coming down the pipeline? What additional capabilities or offerings should we expect to see in the near future?
The current applications will continue to improve and become more pervasive. From better decisioning engines for advertising/marketing to all of the usability-related applications, such as chatbots, personalized UX, voice assistants, etc.
I am also quite excited at the creative field of AI, which is not so much about automation and efficiencies, but about actually letting an AI algorithm come up with something that no human could come up with. This is not just an artistic exercise, but also potentially very effective for marketers; AI creativity is, in the end, based on a much vaster amount of data than any human can store. It is very likely that for some creative, machines can be more effective than humans in generating a personalized response. I am not suggesting that an AI will win the next Cannes Lions; however, a combination of AI and humans could potentially bring creative effectiveness to a whole new level.
Outside of marketing, I keep an eye on AI applications on health and climate, of course. I do hope a lot more funding goes to the latter next year.
You have a long track record when it comes to emerging AdTech. Tell us more about your career up to now. What inspires you about the space?
I am a self-confessed nerd, so I gravitate naturally to technology and problem-solving. I am also someone who thrives under challenge, so I love to be involved with fast changing environments. Adtech ticks both, making it an exciting career path for me.
What makes it even more interesting is that adtech has actually been at the center of many societal changes. Google and its ability to generate profit from ads was a big contributor to the growth of investment in this industry, which in itself has also fueled the explosion of areas like user generated content, with severe impact on journalism or social networks.
Right now, we are just about to see a fairly significant shake-up of the space. The main currency that has been used to connect different adtech players, the cookie, has a very imminent expiration date. Either we figure out a fair-for-consumers, effective identity framework across this world, or what used to be a fairly democratic, interconnected industry where small challengers could come in to push the industry forward, we will just become a group of walled gardens. This will impact innovation and will make the power play a lot less fair for everyone, including users. One reason I joined Amobee is because we will be at the center of this transformation.
Is there anything else you would like to leave the readers with on this topic?
AI is not a futuristic word. It is just technology, and a massive accelerator for progress. It is on us to use this for a good purpose as a society. The more we educate ourselves on it, the better for us and of course for our careers. There are wonderful books, documentaries and podcasts out there, depending on the type of learner you are. So take your pick and embrace the change, with or without fear.
To learn more about Amobee, visit their website and follow them on Twitter at @amobee.
Image by Gerd Altmann from Pixabay