Practical Uses of AI
Artificial intelligence is exploding all around us. Here's what's important for marketers.
The artificial intelligence space is booming. Just today, Gartner Group reported that enterprise adoption of AI has grown 270% over the last 4 years. It's estimated that AI already has amounted to $5 Billion in 2015, and is expected to achieve 20% growth rate each year. Of course, projections of AI valuation are all over the place - check out this comprehensive evaluation of the leading market forcasts here.
Still, the hype around AI in many ways is greater than the reality of meaningful value. In this case, hype equals promise, as the potential for AI continues to be discussed and pioneered across multiple industries. The failures are real too. So how should a marketer approach AI in 2019, and what should they focus on? For starters, the focus should be on AI tech that is already in the marketplace. Also, the AI tech has several vendors that are established and credible. And finally, this real AI tech is tied to business goals (sales, effort, efficiency, etc) and is already reporting true value. Using this short list of criteria, I would suggest there are at least three avenues that offer real benefits - here and now - not just for tomorrow.
Just last week, it was reported that beauty manufacturer and retailer, L’Occitane en Provence is adding artificial intelligence alerts to its website. L'Occitane partnered with Contentsquare,an AI-powered digital experience company, to add proactive monitoring of the website which allows the brand to detect technical or usability anomalies. For instance, if there are content modules or website technical issues that prevent customers from completing a transaction, the system will trigger an alert the ecomm team, and make the appropriate fixes. The AI system can monitor the full range of web page elements, determing engagement (or hesitation) across merchandising images, form fields, call-to-actions or buttons. It seems in this case, almost any ecommerce platform may want to take a hard look at deploying an AI alert system.
2. Customer Service Automation
Automating repetitive tasks is another major promise of AI, and in today's customer service world, the opportunity is now a reality. The business case is this: typically a good portion of customer questions or issues are repeat questions. So, why tie up a customer service agent, or a social media manager, to copy/paste the same response over and over again? Instead, train a smart AI agent to answer these questions. Done right, these agents will provide responses faster, in real-time, with more accuracy than a human ever could - 24/7/365. Already, the main form of automation is chatbots, with over 100,000 businesses using chatbots to deliver answers to customers faster. And the main AI trained architecture can be plugged into any digital channel, not just chatbots. This includes Facebook Messenger and Post comments, Twitter, webchat, phone banks, etc. Once you have the platform built, you can power any customer-facing channel to provide a ubiquitous and consistent customer experience.
An emerging phase of AI-powered customer service is around image detection which can better respond and analyze product issues posted by customers. For retailers and product marketers, the AI drivent image detection space offers a valuable test and learn opportunity.
3. Predictive Analytics
Say a marketer is launching a new campaign, but doesn't know which two different messaging strategies might work best. Utilizing a predictive technology platform, such as Optimove or Segment, that marketer can see which campaign strategy is most likely to be more successful. The opportunity here rests on analyzing multiple variables which will determine success factors and can guide the marketer accordingly, avoiding a wrong direction, or time tied up in testing in a live environment. Predictive technologies can be applied to really any marketing channel, or format. Think about optimizing banner messaging, search marketing copy, email content, or social media posts. Understanding what has the best likelihood of success before you launch is a game changer for today's (and tomorrow's) marketer.
This only scratches the surface of AI's potential for marketing, but again, the focus is on the here-and-now. I know I've missed other avenues that may be just as real, and not just hype. What other ones have I missed? Feel free to leave a comment.