This article was originally published on MarTech Today by Sean Zinsmeister, Senior Director of Product Marketing at Infer.
The sales and marketing industry has been abuzz with talk of predictive analytics, machine learning and artificial intelligence (AI) this fall, especially on the heels of a flurry of AI updates from Microsoft and Oracle, Salesforce’s recent Einstein announcement at Dreamforce, and Google’s unveiling of its efforts in machine learning and AI yesterday.
In all of this hype, I’ve encountered several conflicting definitions and explanations of what AI really means.
Those of us close to the space know that AI, at its core, is actually foundational technology that’s been around in the consumer world for over a decade.
Think of the amazing intelligence behind Google Photos, which uses facial recognition technology to organize your images for you, as well as the highly accurate music recommendations that you get from Pandora based on your likes and dislikes. We see similar examples in major league baseball (remember Moneyball?) and, of course, the fast-evolving world of Uber, Waze and self-driving cars.
As AI enters the enterprise realm — in what Constellation Research predicts will be a $100 billion market by 2025 — it’s important to shift our focus away from science fiction perceptions, and instead look toward the specific business outcomes that AI can produce.
To help cut through the noise, I’ve outlined below four key roles that analytics plays in the sales and marketing landscape. Keep in mind that each of these approaches delivers insights based on sophisticated data processing, modeling and other scientific techniques — all of which are important aspects of AI.
1. Descriptive analytics tells us what happened in the past
These applications look at this information and extract insights about what’s happening in your business and what you need to know (about your customers, prospects, reps and so on).
By shortening the distance between data and decisions, companies can more quickly make course corrections or take advantage of new opportunities.
2. Predictive analytics tells us what is likely to happen next
For example, predictive scoring models analyze thousands of current and historical data signals (both internal and external) about a lead or customer in order to predict whether they’re a good fit to buy your product, as well as whether they’re exhibiting behavior that indicates an impending purchase.
Basically, predictive analytics leverages a bunch of AI and distills it into the form of a simple score that any business user can leverage to understand what they should expect to happen next.
3. Prescriptive analytics tells us what we should do next
The next category of sales and marketing analytics that’s emerging looks at your data to tell you what your business should do about specific insights and what you might want to avoid.
This typically comes in the form of recommendations and can be situational based on a particular workflow or perhaps a certain stage in the customer funnel.
Some examples of prescriptive tools are chatbots and natural language processing. These technologies help businesses handle large volumes of inbound support requests by routing them intelligently, or run programmatic advertising campaigns that adjust bids based on relevant real-time factors.
4. Self-driving workflows proactively execute for us
Finally, the type of analytics most often associated with AI is all about reimagining automation through data science. Many people who have a science fiction perception of AI — as seen in the futuristic HBO hit, “Westworld,” — are actually thinking about artificial general intelligence (AGI): “the intelligence of a (hypothetical) machine that could successfully perform any intellectual task that a human being can.”
This type of technology is still pretty far off, and it’s not what most folks in the business world mean when they talk about AI today, which has more to do with using software to accomplish specific problem-solving or reasoning tasks.
AGI aside, there is still a lot of routine, repetitive work in sales and marketing that we really don’t need humans for. Instead, businesses can apply automated conditional logic to run based on their data sets.
These systems look at workflow patterns to figure out limited actions that should always happen in clear scenarios, and then actually trigger and execute those tasks for us. With these solutions, the sales rep or marketer is just there to supervise and manage exceptions.
Don’t get caught up in the hype
While all of these technological advances are certainly exciting, it’s important for marketers and salespeople to avoid getting caught up in the AI over-hype and instead think about the results that matter to them.
Does your team need to find white space in the pipeline where you can generate revenue this quarter? Optimize marketing campaigns to ensure you’re generating the most MQLs at the lowest cost? Free up time to call more top leads faster?
As Ray Wang, founder of Constellation Research, said in a recent article on artificial intelligence, “Just enabling AI for AI’s sake will result in a waste of time. However, applying a spectrum of outcomes to transform the business models of AI-powered organizations will indeed result in a disruptive business model and successful digital transformation.”
Don’t grasp at straws with AI. Make sure you’re focusing on your business needs and applying the appropriate martech solutions to achieve real impact.