This is the first blog in a two-part series that explores how to use advanced digital technologies to create revenue lift. Part two will take a deeper look at technical considerations, use cases, and benefits.
The first and only real consideration for any business looking to deploy artificial intelligence (AI) or machine learning is whether the application or model will produce revenue. Will inserting advanced digital technologies into the continuous business process cycle of data, insight, and action monetize opportunities that arise from turning deeper insights into action?
Many companies deploy machine learning models to improve the customer experience, but most fall short of becoming true revenue-generating engines. Fast food restaurants, for example, are racing to introduce AI-powered menu boards that recommend add-on items based on current selection, restaurant traffic, or conditions such as the weather, time of day, or trending items in the area. Upsell may be a noble pursuit, but this use case in present form is more of a novelty than a mission-critical system that will boost the bottom line. Likewise, AI-powered chatbots are exponentially more intelligent than even a few years ago, but other than providing a more pleasant customer experience they are unlikely to move the revenue needle.
Move Beyond Personalization for Personalization’s Sake
For AI and machine learning to truly ascend as a top revenue-generating engine for the business by providing a differentiated customer experience, advanced analytic models must be embedded across the complete customer lifecycle and every channel or touchpoint.
Otherwise, models do little more than scratch the surface of possibilities. Personalization for personalization’s sake, such as seeing your name on a menu board when you order a hamburger, is vastly different than personalizing a customer experience for a segment of one, in real time, based on a customer’s behaviors, interests, and intent across an omnichannel buying journey.
Consider again the AI-powered chatbot that helps a customer resolve a service issue. The customer may be pleasantly surprised by the user-friendly experience and awed by the technology, but the experience will not translate into direct and measurable revenue gains because it is restricted to a single channel.
While the chatbot can intelligently respond and interact with a customer, it will not know anything about the customer beyond the specific issue, or beyond the available data that resides in that channel. Perhaps it goes so far as to recommend a product in the color and size that matches previous online transactions. The customer may later purchase the item in-store, but if the data is siloed by channel the brand will not have visibility into the journey, clouding any direct impact of the chatbot conversation and depressing the value of the advanced technology. A one-off sale is not akin to direct revenue lift, which can be significant.
According to Boston Consulting Group, there will be an $800 billion revenue shift to the 15 percent of companies that get personalization right over the next five years in three sectors alone – retail, healthcare, and financial services. Personalization also drives retention, which astute brands know is more profitable than acquisition. According to the Aberdeen Group, companies that have an extremely strong omnichannel engagement strategy have an 83 percent customer retention rating, vs. 53 percent for companies that do not. The “Profiting from Personalization” article states that “brands that create personalized experiences by integrating advanced digital technologies and proprietary data for customers are seeing revenue increase by 6 percent to 10 percent … two to three times faster than those that don’t.”
Unlock Channel Constraints to Move the Revenue Needle
To produce revenue lift that makes a difference, advanced technologies must automate intelligence to dynamically engage with a customer across every interaction and touchpoint in an omnichannel buying journey. Moreover, self-training models must be built on a unified customer profile or single customer view that captures customer data of every source and type in real time. Embedded intelligence supported by a 360° view of the customer has the power to recommend an algorithm-produced next-best-action to a segment of one the moment the customer appears next in a dynamic customer journey.
This is transformative. Consider a traditional marketing strategy bound by channels. A message or offer is sent to a segment of customers, with success measured by click rates and conversions. A customer who doesn’t respond may then show up anonymously on the website; without any link between the email and the cookie, the customer will likely receive an inconsistent message. Like the chatbot example, without synergy between channels across a complete buying journey, the power of AI is muted because it will fail to recommend a next-best-action in the context and cadence of the customer.
With embedded advanced analytic capabilities unbound by channels and data siloes, marketing will know in real time everything there is to know about the customer – a call center agent or chatbot will know that the customer who received a specific offer then went to the website anonymously before making the call. Within milliseconds, a machine learning model will not only produce real-time information about the customer, it will also produce a next-best-action based on the consistently updated single customer view.
Ensure Consistent Messaging with Lights-Out Modeling
Advanced digital optimization ensures a consistent, personalized message and journey for a customer irrespective of channel or any other variable or condition such as day or time. Embedded AI doesn’t have to wait for data scientists to build models. Rather, lights-out modeling runs 24/7 looking for opportunities in the unified customer profile. Simulation engines constantly watch models, and will move new models into production that predict better outcomes based on predetermined metrics.
Linking predictive rules, despite what many marketing automation tools claim, is not AI. The fact is, predictive rules are not dynamic; models built on them will become stale over time. An in-line optimization engine that works with real-time customer data is immune from this problem, and thus does not require human intervention to refresh models – a common practice that is often only done after many missed opportunities at monetizing data.
Automated embedded intelligence enables hundreds or thousands of models to run concurrently, all with a single-minded purpose of exploiting revenue opportunities according to any metric the business proposes. Personalization is the key that unlocks any opportunity to monetize customer data with a differentiated customer experience.