Talk to any personalization technology vendor and chances are you'll hear a beguiling tale of true one-to-one customer experiences made possible by their product’s uncanny ability to serve up the right content to the right person through the right channel at the right time.
It’s a seductive vision: Just rev up the personalization engine and voilà, you’ve activated your express pass to conversion and traffic heaven.
Real Clients Get Personal
But is that vision too good to be true?
This is the first in a CMSWire series that will survey the current state of the personalization field. We’ve interviewed vendors and industry experts to gather their thoughts on the potential and pitfalls of personalization today.
Even more groundbreaking though, we’ve talked with actual enterprise users who have shared their real-world personalization challenges. We think their stories powerfully illustrate how personalization theory is actually playing out on the front lines of content creation today.
Marketers often have unrealistic expectations noted Jeff Cram, co-founder and chief strategy officer for Boston- and Portland, Ore.-based Connective DX, which helps marketers implement personalization projects.
“People often get paralyzed in personalization because they still harken back to the [days when] personalization was all about the right content [to the] right person,” Cram told CMSWire.
“And [unless you’re Amazon], that’s just not really [a level] of personalization [that’s realistic.] If that’s really your end goal, it can prevent you from starting anything. It’s just not reality for most businesses.”
Data, Data Everywhere
One significant barrier most enterprises encounter is the ability to manipulate massive data sets.
An Econsultancy and ResponseTap survey last year found that of almost 2,000 marketers, 73 percent believe data is of “critical importance when it comes to understanding the customer journey.” However, the survey also found that 43 percent of companies rated their ability to slice and dice actionable data as merely “okay,” while 23 percent self-reported themselves into the “poor” or “very poor” categories.
What’s more, those data sets are often scattered and siloed. The average organization has customer data residing in ten to 15 systems and sometimes as many as 20 or 30, according to Brendan Witcher, principal analyst for e-business and channel strategy at Cambridge, Mass.-based Forrester Research.
“It’s very hard to develop a single view of the customer when you have data in so many different locations,” Witcher told CMSWire. “This idea of bringing customer data together to develop a single view of the customer is really the starting block, what gets you off the line for personalization. [But] if you haven’t got that right, your personalization will only be so successful.”
The privacy implications of personalization weigh heavy on marketers as well.
Andrew Frank, research vice president and distinguished analyst at Gartner, advocates for the idea of “personification” over “personalization.”
“You need to have a way of targeting people while maintaining anonymity for privacy purposes,” he explained to CMSWire.
Some vendors, Frank reports, receive negative feedback from privacy advocates. For example, Microsoft surveyed 13,000 people and found that 83 percent wanted to know exactly what information they were giving up to companies.
“What you really want to target isn’t a person but a persona,” Frank recommended to CMSWire. “And that’s resonated very well with a lot of companies. [If] you can target a specific group that is large enough not to be associated with any individual but targeted enough where the message is really going to resonate clearly, then it’s a win for everybody.”
Passing the Test
A commitment to testing and optimization is also a key component of an effective personalization strategy according to Drew Burns, principal product marketing manager for San Jose, Calif.-based Adobe.
Show all types of variations on a web page to see how they perform and mitigate risk, Burns suggested. “We've had a couple of customers that saw engagement drop in a test,” Burns said, “and they were able to yank that pretty quickly. You don’t want to make a decision without testing.”
Personalizing Clark University’s Website
Matthew Cyr, director of content marketing at Clark University in Worcester, Mass., embarked with colleagues on Clark’s 18-month website redesign that included personalization.
“There’s a whole lot of logic you have to build in the backend to make [our] system deliver the information you want to deliver to the right person at the right time,” Cyr discovered. “It’s that sort of figuring out how does the personalization piece interact with the typical experience of being on the website. It takes some getting used to for sure.”
Implementing Clark’s personalization vision meant beginning with a detailed up-front discovery process aimed at defining and answering the project’s central question: Who do we want to talk to digitally, and what do we want to say?
From Persona Building to Content Integration
The redesign started with persona building, journey mapping, data analysis and Google Analytics, all executed through phone calls and dozens of meetings with different university departments and administrators. They worked with Connective DX.
Before all was said and done, the project would also encompass a necessary integration between Clark’s web content management system and its chosen personalization tool, building content for six audiences while specifying up to five different actions for each audience as well as creating and specifying “variation sets” for up to eight content components.
"You can see why I spent most of my weekend doing this,” Cyr said.
Clark’s Lessons Learned
Cyr’s advice to marketers?
Think about web personalization not in terms of the backend work or even front end but what experience you want for your built-in personas, he said. “It’s a lot of thinking through the entirety of the process and really understanding the personas, goals and journey mapping in a really deep way in order to understand it."
Cyr’s redesign will culminate in a personalization tool called “My Journey” (built through Acquia Lift) that aims to walk Clark’s website visitors through personalized journeys based on their personas. It will look this way:
Advice from the Experts
What could the experts have told Cyr before he began and what does his personalization project experience highlight for others?
- Expect to be strapped for resources. Fifty percent of all digital analytics teams are composed of one and four members, according to a 2015 study from EY entitled “On the Road to Online Personalization: The EY Digital Analytics Benchmarking Survey Analysis.”
- Expect to work long and hard. Personalization involves mastering and integrating many new skill sets such as persona building, journey mapping, tool selection, logic building, tagging, A/B and multivariate testing and integration work between web content management systems, personalization tools and analytics engines.
- Don’t expect to find massive tech maturity. Personalization is in its “nascent” stage, according to Jason Daigler, research director for digital commerce at Gartner. That dovetails with the 60.9 percent of respondents in CMSWire/Jahia’s 2015 Digital Customer Experience Survey who reported being at an “early stage” in their organizations’ digital transformation.
- Expect to focus on customer privacy. Personalization comes with its share of privacy concerns, as University of Michigan researchers found in their 2015 report, “Privacy Concern, Trust, and Desire for Content Personalization.”
- Don’t expect automated help — yet. Machine-learning and automation could be the next big wave for personalization but right now it’s not the leading use case, Gartner’s Daigler noted. In fact, the most common use cases today are email personalization and product recommendations.
(Editor’s note: Coming later this week we’ll be sharing more stories from marketers like Matthew Cyr on the opportunities and challenges they have found in implementing their personalization initiatives.)
Title image by Anthony Delanoix