Susan Etlinger, an Industry Analyst at Altimeter, a Prophet Company is the featured guest on episode 165. Susan is a globally recognized expert in digital strategy, with a focus on artificial intelligence, big data, analytics and digital ethics. She conducts independent research and has authored a series of reports available for download at Prophet.com.
On this episode, Susan and I discuss the evolving role of big data, artificial intelligence, chatbots, and examples of how they’re used in business and where they’re heading. This conversation is grounded in practical reality for marketing and sales professionals to help you understand the impact of the age of the algorithms in 2017. Susan provides a sneak peek of her latest report titled The Conversational Business: How Chatbots Will Reshape Digital Experiences which is available to download for free.
The Role of AI in Business in 2017
Susan says that AI in business is in the very early stages. In her report The Age of AI, Susan reviews How AI is Transforming Organizations. We’re seeing lots of experimentation and pilots as organizations figure out how to create predictive analytics. Machine intelligence may be a better way to refer to AI. When people hear the term “AI,” most think of science fiction, but we use it every day with Facebook, Google, Amazon, etc.
It’s not size that matters when it comes to data – it’s context. In the early days of the Internet, big data was defined by volume, velocity, and variety. For example, today there is a high variety of images, emoji, text, and speech making it very important to understand context.
In a TED Talk Susan delivered in September 2014, she points out how in a Twitter search the word “smoking” has four categories of context: smoking cigarettes, smoking marijuana, smoking ribs, and smoking hot women. Marketers should look to large players like Microsoft Cortana and Google to help marketers overcome disambiguation – to take the ambiguity out of the data.
In addition to the major players like Google and Cortana, we’ll see AI created by startups. There are currently over 1,700 startups in the AI space tracked by Venture Scanner. Examples include applications of a chatbot that can order you a pizza or a tool that can predict the next action in a B2B sales process. Susan says that in the future, AI will be purchased as a service similar to software.
The Impact of the Age of the Algorithms
When piloting with AI, marketers should be cautious to avoid creating bad experiences. Take, for example, Facebook’s chatbots. They had a lot of hype behind them, but as people started to use them, they turned out to be pretty awful. When created well, chatbots can create an experience through language and image recognition, they can help to find new audiences and use predictive analytics to gather information during interactions.
Predictive sales capabilities of the past are different from today’s in that the algorithm can now update itself based on what it learns. Here’s a look at how Marketing, Sales, and HR can use AI:
Marketing
- Appeal to new audiences with conversational interfaces.
- Learn more about audiences based on their behavior.
- Solve customer service problems.
- Create a better experience that’s more innovative and differentiating.
- Learn and incorporate into predictive models for effective targeting.
Sales
- Predict high-quality leads.
- Predict better next steps, but remember that machines can be biased.
HR
- Allow the machine to find patterns in the data to determine a good candidate.
- Recruit top-quality candidates you otherwise wouldn’t have considered.
When experimenting with AI, Susan recommends looking to the leaders in the space, i.e. IBM, Microsoft, Google, Adobe, Amazon, Salesforce, etc. Dominos and Taco Bell are also using bots to create a good experience.
Susan points out that some of the companies that are deploying AI are not big innovators, they simply have a strong desire to improve the customer experience and are willing to experiment. The dividing line between research and skunkworks projects is much grayer than with traditional projects, the primary cause being that no one knows how an AI product will perform until it’s “in the wild.” Companies need to value what they learn as an outcome and celebrate their failures to move forward.
Over the next ten years, the way we interact with technology will dramatically change. Susan says we’re in the year 1993 of AI.