Jay Baer, Marketer, Author, Speaker and President of Convince and Convert, is known for saying: “Make your marketing so useful people would pay for it.”
In other words, every interaction our brands have with a customer should deliver value to that customer. One-to-one marketing is no longer good enough—even though we’re just getting there. To provide value with each and every interaction, we must understand who individuals are and speak to them contextually, one-to-one, in the moment in which they are situated.
There are many moving pieces that must come together for us to achieve this one-to-one in the moment marketing, and one of them is technology. An ecosystem of martech capabilities that shares data in real time allows us to begin to behave in the digital world just as we would in the physical one—in a more relevant, valuable, human manner. That real-time data sharing and the resulting insights are key, and signals in a Signal Hub make them exponentially easier to achieve.
Signals have been well-known in the IT world for some time—but not so much in the marketing world. Considering their value to enabling more human, one-to-one in the moment interactions, it’s time that changed.
What are signals?
Opera Solutions, the maker of Signal Hub, defines signals as follows:
Mathematical transformations of data that take the form of modular units of intelligence. Advanced analytics techniques, including machine learning, generate Signals. Signals can be consumed, blended with other Signals, shared, and stored. Signals reveal patterns that have applicability to business situations and that can facilitate accurate predictions of future activity.
The bold above is mine. It’s the real value of signals to marketers.
Why are signals important for marketers?
Let’s put this into marketing terms. As marketers, we’re trying to get at the small data that tells us who our customer is, why she is doing what she’s doing at any given moment and what she’s going to do next. What are those digital breadcrumbs she’s leaving behind that help us to figure this out and provide her with a message that fits her needs? In the physical world, we would just ask “How may I help you?” But in the digital world, we must discern the answer from these digital clues.
The raw data does not offer us any value. We must transform it, get it to tell us what to do, what will help us to improve our marketing performance, find a customer that will convert, etc. We need to take what we know about an individual and turn it into something that predicts what she’s going to do in the future.
Signals make this much easier for us than it has been in the past.
How do signals work?
Signals come from all of the data you’re generating with the various pieces of your martech stack. A Signal Hub takes raw data and helps you to transform it into descriptive signals about the buyer, then to predictive signals about what the buyer is likely to do, and then ultimately into prescriptive signals that recommend your next move.
This becomes the insight engine or the brain that brings all data together and helps us, as marketers, figure out what our next actions should be.
And the brain is constantly getting smarter. Machine learning within the Signal Hub constantly performs test and learn scenarios. Closed loop feedback on your data “signals,” like a car blinker, that message X might be good for customer Y because of action Z. For example, if I buy vitamins at Walgreens, I might be susceptible to an offer for protein powder. The system tests it out, sees if it works and optimizes the message the next time it sees someone like me buy vitamins.
Contrast that to current analytics approaches…
In today’s world, the treatment of data is typically done on a one-off basis. In a typical hypothesis-driven approach to optimizing your marketing, you have to rewrite code for every regression or cluster analysis. The way the code is written in this Signal Hub platform allows us to reuse common algorithms. It provides a library of existing signals.
For example, as a retailer, you may often want to analyze factors like store visits, geographic proximity to store, coupon use, etc. These are built into the Signal Hub. They’re like macros in excel—common formulas are pre-coded into the system and let us get things done more quickly. This saves our analysts a LOT of time; the system handles about 90% of what you’d want to do as a data scientist. Ultimately, this allows us to move more quickly and agilely as marketers.
The best part of using a Signal Hub?
Everything is integrated into one system and one interface. This article in CIO explains that most enterprise marketers have upwards of 20+ technology solutions, and getting the most out of them requires combining all of their data points. The typical fix is less than ideal:
“This is why many of today’s marketers rely on impractical solutions such as cramming as much data into Microsoft Excel spreadsheets as possible. You can only imagine the pain and suffering of manually gathering data points into complex Excel documents.”
With Signal Hub, we can more easily use ALL of our data to inform predictive and prescriptive analytics in one interface. This improves ease of use while also improving the traceability of our data across systems for a single view of the customer.
The future of marketing
This is the future of marketing…using technology to analyze massive amounts of data to find patterns that humans would never be able to find. It’s a machine-driven approach that uses self-learning systems to explore large amounts of data more quickly than we can imagine.
And at the end of the day, it helps us to be more human by better responding to each customer with the right message in the moment that she needs it.
You may also enjoy CMOs: This is the Year to Focus on Artificial Intelligence.
About the Author
Jennifer brings 13+ years of marketing and product management/innovation experience to Harte Hanks. As Director of Innovation and Growth, Jennifer leads the positioning of Harte Hanks' marketing technology solutions—which includes the modernization of our customer data, data management, and advanced analytics platforms.More Content by Jennifer Miles-Losapio