2

Adopting an Analytical
Mind-Set

From Reactive to Proactive

If you want to evolve into an analytical marketing organization, first you need to focus on your mind-set, which is why my guide to change starts here. The evolution of your organization begins by shifting how you and your organization think about using data and analytics in the way you go about marketing.

The shift to adopt an analytical mind-set certainly had an impact on my career. It was initially driven by our need to do a more effective job at measuring the value of what we were doing, especially when considering our impact on the sales pipeline and revenue. And we had to think differently because we now had all kinds of new data and metrics available that would allow us not only to look at effectiveness, but also to identify potential. Measuring impact in a quantitative way was critical. Being able to use analytics in predicting what would work or responding to the customers' behaviors was an even more powerful motivator.

These changes were also forced on us as marketers because, due to the massive influx of new digital channels on the Internet and social media, our customers' expectations had changed. And they continue to change. We understood that if we wanted to respond more effectively to our customers on their decision journey, we needed to build an analytical culture capable of interacting with them in a highly personal and customized manner that was also flexible enough to change directions quickly.

The first step in shifting your organizational mind-set to that of an analytical marketer is changing how you think (and feel) about data.

Be Accountable for Your Data

Everyone on both the consumer and business sides appreciates the value of data—all types and sizes of data, as well as the evolution from Big Data, and the Internet of Things. The key for the analytical marketing organization, regardless of your industry or your customer, is to understand how you can use all that new data to effectively personalize your interactions with your customers. After all, as Emmett Cox, a business intelligence expert who has worked for large organizations like Walmart, GE, and Kmart, writes in his book, Retail Analytics, “Data without use is overhead.”¹

Historically, that kind of data expertise lived in a different part of the organization, most likely in IT. If you wanted data in the form of a report or some such, you would submit a request to IT and wait until you got what you asked for. That's no longer viable. Analytical marketers are now producers as well as consumers of data, which is changing how we have to structure partnerships within our organization (something we will dig into in the next chapter) and the kind of people and skill sets we hire (a subject we'll explore in more detail in chapter 4).

Data is the holy grail of marketing analytics, whether big or little, complex or simple, structured or unstructured. That's because all data tells a story. The most valuable data, as well as the most complex, is customer data. Once you have good, clean data to work with, you can then begin to apply analytical tools so you can make proactive decisions based on the stories that the data is telling you about your customers and prospects. If you begin to see data as the “author” that is sharing trends, helping you identify behaviors, and measuring the value of your activity along the journey, you will truly appreciate the impact it can have on all elements of marketing.

The key is that you and your whole marketing organization need to shift your mind-set so that you understand the importance of data. It's your responsibility to collect it, care for it, and establish rules for governing your data. Cox told me that, while most organizations have moved to embrace their data, they've also fallen prey at times to collecting data for data's sake, without first taking in the business value of that data. “Too often, organizations start chasing the next shiny object like social media before they've optimized everything they already have,” he said. A good rule of thumb in making sure you are collecting the right data is to ask yourself what three things about your business are keeping you up at night. If you are collecting data that will help address those concerns, then you're on the right path.

So what are some other components of a sound data strategy?

• Volume

• Source

• Complexity

• Structure

• Quality

• Relevance

• Integration

Our experience with a data strategy is that it is a continuous process. Even with all the tools at our disposal, data collection, management, and governance are challenging. As you grow and change, your needs will continue to shift direction. The way we sell, market, and support our customers has morphed through the years, affecting the type and volume of data we have. For marketing, we've continued to expand our data sources in an effort to gather more and more information on our customers, especially from preference and behavioral perspectives.

Our journey as an organization to better embrace the stories behind the data we were collecting began in 2009 when we modernized our approach. Until that time, we had relied on a disparate array of data sources, mostly dozens of spreadsheets, to track our marketing campaigns and responses. It was largely managed manually and on an ad hoc basis, which created huge hurdles every time we needed to create a contact list for a new campaign. Sometimes it took up to three months just to build an accurate list we could use as part of a promotion. In time, we began to realize how untenable that kind of system was.

That's why we shifted direction and began to implement what we call our data mart, which is quite simply a structured source that pulls together all the disparate data that used to live in spreadsheets. With our data all in one place and updated daily, we could begin to paint a full picture of our customers, because we could, for the first time, see how all of their different behaviors were connected. For example, in the past, we might have lost track of which contacts at which customers we had reached out to during certain campaigns, and how they responded. “A lot of marketers have access to web analytics and behavioral data,” Matthew Fulk, a marketing director at SAS, said. “The differentiator is learning to ask the right questions about the data and making it a priority to do so.”

A powerful aspect of the data we were now collecting was what people were doing on our website, something we call “customer experience analytics,” or CXA. We now knew how much time customers were spending on our website, what they were looking at, which white papers they downloaded, what videos or webinars they watched, and so on. That all combined with more transactional records, such as installation records on a customer's software service record or how many times a customer called tech support and why, as well as offline behaviors, such as when a customer attended a conference or event and what he or she did there.

Through our data mart, we also began to tie customer behaviors back to building sales pipelines and generating revenue. In other words, the data we now captured could tell us that $X of the pipeline eventually turned into revenue and could be tied to which outbound or inbound marketing actions we took.

Added up, that meant we now had a much more complete picture of our customers based on the breadth of the data we could collect, store, and, perhaps most importantly, analyze. We had new opportunities to apply our analytical tools to examine customers' different behaviors, depending on where they happened to be in their decision journey.

Data needs to be clean and accurate; otherwise, your analysis and, ultimately, your decisions could be flawed. Now and then, you'll need to rebalance it. Look for sources you no longer need or data that isn't providing value and get rid of it. Reinvest your resources where you're getting returns. Help the data tell the story so the analytics can learn from it. In our marketing organization, we focused an initiative on data source performance and were able to eliminate the poorest-performing investments. The data gave us the confidence to let go, which translates to cost savings and increased effectiveness.

Creating a Data Mart

A common challenge in customer data is consistency, especially with the increase in unstructured data from text sources in mobile and social channels. When we began collecting and governing our marketing data so we could gain insights into our customers' decision journeys, we created a platform, or data mart, that helped us standardize and create connections between various sources that enabled us to paint dynamic pictures of what our customers wanted.

A Hundred Data Elements from Four Key Sources

• Sales database

  – Leads

  – Pipeline

  – Invoices

• Marketing database

  – Online and offline interactions

• Customer database

  – Training

  – Publications

  – User groups

  – Installation

  – Tech support

• Purchased data

Some additional examples of the kind of data that we track to better understand customers' decision journeys and their progress are:

• Live event attendance
• Website traffic
• Technical support queries
• Customer satisfaction survey data
• Customer reference activity
• Webinar attendance
• White paper downloads

 

Having a reliable and clean source of data to work with is critical. Those using reports or analysis can lose confidence when something doesn't seem correct in the outcome. The first place they blame is the data, and when they lose confidence in the data, all sorts of problems result. When everyone in the entire organization takes full accountability for the value of the data and its care and feeding, and they truly know the data, the results are strong. The confidence level in decision making is high, the level of innovation is robust, and the impact is evident.

The Data Oath

Your analytics will only be as good as your data. The marketing organization's job is to develop a culture of respect, care, and maintenance of data.

1. Protect the data

   – Establish and enforce the governance.

   – Define a clear process and joint accountability.

   – Leverage expert methodologies and technology.

2. Respect the data

   – Make data quality everyone's job—both internally and externally (don't let bad data sources into your data).

   – Ask questions about the data; get to know it well.

   – Have only one version of the truth; kill the spreadsheets.

3. Love the data

   – Don't blame the data (or each other).

   – Listen to the story the data is telling.

   – Provide unconditional support for and nurture your data.

 

When you are open to hearing the stories it can tell you, data will likely lead you to a particular point of view. Done right, data digging will bring you to conclusions and additional questions. This provides the foundation upon which you can make solid business decisions. Unfortunately, all too often, we can fall into the trap of using data to confirm a presupposition or bias. That's why the data by itself isn't enough.

From Mad Men to Math Men and Women

This is no longer the era of Don Draper, the now infamous lead character in the TV show Mad Men. The series, set in New York City during the 1960s and 1970s, followed the exploits of an ad agency and its various clients. Draper was like a magician, a creative genius who always seemed to know what kind of campaign the clients would fawn over. But we're in the middle of a shift where data and analytics, not the creative talent in the art department, drive the success of a marketing campaign.

Marketing has traditionally been considered an “art”—a practice based on creativity, gut-based decision making, and no real expectations that it could directly affect the firm's bottom line. But the shift in the science means that marketing has come into its own and evolved to what we're calling “marketing analytics.” Marketing now is more about science or math that is driven by an influx of data, channels, mobility, and, most importantly, changing customer demands.

Rather than relying on Don Draper types, we're leaning more on the work of folks who help analyze behavioral data and the digital footprint of our customers and prospects. They produce reports that lead to campaigns more focused on where customers are in their decision journey and what they are looking for.

The term forensics has entered everyday marketing vernacular. Typically associated with the use of science and technology to establish facts in a criminal case, data forensics refers to the practice of using data discovery to establish the facts of a marketing activity, campaign, or broader initiative. But beyond the basics of data digging, data forensics incorporates intangibles—the piecing together of anecdotal and qualitative tidbits along with quantitative data to develop a rich picture of performance. The combination of qualitative and quantitative data provides the context necessary for sound decision making.

Marketing Analytics at Work

Finding Your Customers' BFFs

One of the most powerful sales tools is often something that you can't foresee or control. Even though customers read papers, visit websites, and talk with a salesperson, another factor can make all the difference: a referral from a friend or coworker.

Consider the way that sites like Google, Yelp, and others have changed the way consumers make everyday decisions, such as choosing restaurants. You can go to the restaurant nearest you or one you've visited before. Or, you can try something new by looking at your smartphone to see which dining spot has the highest ratings or the best reviews. Why? People place a premium on the personal experience of those in their networks.

For business-to-business software companies like SAS, the impact of customer advocacy is critical. These influencers can set the tone and provide a consistent positive influence throughout the customer journey. Unfortunately, this type of advocacy is tough to measure and hard to predict.

The Challenge

Although a customer may be a single record in your database, she doesn't exist in a vacuum. Each contact has a connection to others within her business or the industry. Understanding and fostering good relationships can have a huge effect on your retention and loyalty efforts.

During the effort to formalize a new customer journey, the SAS marketing team began to focus on different phases of this cycle. The customer journey contained the following phases:

• Acquisition, which includes need, research, decide, and buy
• Retention, which includes adopt, use, and recommend

On the retention side, the team knew from anecdotal evidence that some SAS customers were advocates of the technology and for the company overall. In fact, several SAS geographies and divisions had data confirming the idea that finding and rewarding high-value customers led to big returns. What was lacking was an overarching program for getting customers to advocate for SAS technology.

For a larger effort, the team assessed the customer behavior data, examining those who attended events, provided feedback on surveys, sent ideas to R&D, and generally stayed engaged with the company. From a revenue standpoint, those people were often the ones advocating for the use of new SAS technologies or the expansion of existing deployments.

What was less understood was the reach of these influencers and how their activities affected others within the account. With that information, SAS could identify more advocates and nurture that behavior.

The Approach

The SAS marketing team members started by digging into the data that they had on customers. They first identified a segment of the top accounts, which contained more than twenty thousand individual contacts. Once identified, the team began to examine the behaviors exhibited by that group:

• Live event attendance
• Website traffic
• Technical support queries
• Customer satisfaction survey data
• Customer reference activity
• Webinar attendance
• White paper downloads

This information provided a better understanding of the range of activities that customers undertake. However, simply cataloging the behaviors wasn't enough. The team applied a scoring model for different types of interactions. This allowed the team to weight certain activities, helping further identify which customers were the best advocates—“BFFs,” or “best friends forever,” as the marketing team began to call them.

The Results

SAS marketing used the information to create a model that is the foundation for customer-focused data exploration. The initial effort helped shed light on how influential advocates can shape retention and additional sales. As a result, sales and marketing worked together to highlight BFFs within key accounts in an ongoing effort to foster better relationships with those key individuals.

Initiatives to locate and encourage advocates used the model to identify the likely candidates within customer organizations. The team then designed campaigns and outreach efforts to give these advocates the tools to foster and expand their influence.

The marketing team now focuses on advocacy campaigns that target potential BFFs. The goal is to build more SAS advocacy during the recommend phase of the customer journey (see figure 2-1).

Figure 2-1

1. Title: Customer "Best Friend" Analysis.
2. What: Scores and analyzes customer behaviors and determines which behaviors are most highly correlated.
3. Value: Groups customers into high-, medium-, and low-engagement categories and identifies behavior trends that are most commonly grouped together.

   – The goal is to keep your customer as a best friend and identify opportunities to create more best friends.

   – Identifies cross-promotion opportunities.

   – Identifies opportunities to better service the customer.

 

When you add analytics and visualization tools to your marketing toolbox, you can begin to tell stories from the data that you have worked so hard to cultivate and harness. Essentially, you have created “the art of analytics” or the art of analytical storytelling—the perfect blend of art, science, marketing, and math. You can begin to see connections you had missed otherwise. You can also see how the story changes when you make changes. You can then begin asking what-if questions to see what kinds of new stories you can begin to tell. That's how you can shift from being reactive to proactive, to create the future rather than just being weighed down by the past. The better you begin to understand how to tell the stories, the more you become empowered to ask additional questions and to see if you can find even more data to expand on the themes you've uncovered.

We've also seen this lesson in companies we work with. Take, for example, the credit card company Visa. In a company that size, it's obvious that there's a sophisticated marketing machine behind its portfolio of digital advertising, television commercials, sporting event sponsorships, and credit card offers. Maintaining a position of leadership in the financial space requires complex processes, critical and creative thinking, and, according to analytics executive Ramkumar Ravichandran, a pervasive, analytical mind-set.

As director of analytics and A/B testing at Visa, Ravichandran supports executives, leaders, and decision makers in product, marketing, sales, and relationships. He explained to me that “we are the custodians of the data, so our responsibility is to enable our users to have confidence in the decisions they make using that data.”

One of the biggest changes the analytical era of marketing has brought about is that things need to happen much faster than before. “We used to have a very linear approach,” Ravichandran told me. “Now when something is going live, there's already an immediate need to respond. We need to be able to take action on the fly.” Because of those changes, marketers can no longer think about analytics as something that supports them or a function that just one person, likea chief digital officer, would perform. Rather, analytics is now an integral part of marketing's value chain.

Ravichandran said that numbers by themselves are historical. That's why, while data is needed to inform campaigns, at the end of the day, it still comes down to marketers using their gut feelings to make the best decision possible. “And we can use data and analysis to inform and guide us in the right direction,” he added.

Because data and analytics are now so intertwined with marketing strategy, expectations for leadership on the marketing side have changed. “It's no longer acceptable to say you're a marketer, but you're not a numbers person,” Ravichandran said. “Executives are demanding more data literacy as a precursor for being a good marketer.” And it's not just in the marketing space. He added, “All of our chief executives are comfortable with numbers and data-driven approaches.”

Ravichandran was quick to clarify, however, that a focus on data, numbers, and quantified measures should not replace the value of vision: “I have an enormous respect for data, but I also believe all of it has to be driven by strategy, the business case, benchmarking against the industry, all those things that provide a broader perspective. You have to understand what specific metrics you're trying to impact with your actions.” He advocates the importance of understanding your company's business model, applying and measuring the right metrics, and truly understanding your competitive position and your customers' needs.

The big mind-set shift we need to make, therefore, is recognizing how our intuition is now informed by data and analytics. When someone comes to a marketing manager or leader with a proposal to spend, say, $250,000 on a campaign, she had better come armed with data, analysis, testing plans, and expected outcomes, as well as what her gut is telling her.

Of course, marketers have always relied on a variety of metrics for measuring their campaigns. One prominent example is the response rate to a specific campaign. Those kinds of metrics give us something to react to. Metrics tell us where we have been, where we are now, and how we performed compared to a previous year. They paint both a historical perspective and a current status, but they fall short in revealing what's next. The new reality is that reacting to metrics isn't enough. Customers now have greater expectations about what they expect us, as marketers, to know and how we interact with them.

By contrast, a sustainable marketing analytics strategy needs to be equipped with advanced analytics, so we can begin answering questions such as:

• Where is the opportunity?

• Where should I invest next? Or differently?

• What needs to change?

• What is the full story that my data is telling me?

• How do I stay ahead of the customer's expectations?

Today's analytical marketers need the kind of data they can use to become proactive, predictive, and agile enough to make changes quickly and easily. We need to employ better tools that allow us to talk to our customers and prospects based on their location in the decision journey. We need to interact differently with our customers, to better personalize how we respond, when and how they find us. Advanced analytical capability and approaches enable marketers to make fact-based decisions about design, audience segmentation, channel optimization, inbound marketing, and nurturing efforts. Unlike in the past, marketers are now able to deliver valuable information about trends and the digital dialogue of customers and prospects.

For example, if a sales lead comes from a live event we host, we need to have more information on that prospect's behavior, such as:

• After the conference, what web pages did he view?

• How long was each visit to the website?

• What assets did he download?

• What other activities did he participate in-webcasts, other conferences, sales calls?

With this data, you can start to assemble a picture of how every part of the marketing spectrum affects a sale (see figure 2-2). The figure shows a snapshot of a contact at an individual company, but we could look at similar metrics for everyone in the company and beyond. The value of information expands exponentially when you start to evaluate the aggregate behaviors of hundreds, thousands, or millions of contacts. With this level of data, you can build a better picture of customers' behavior and start to assemble marketing programs designed to anticipate their demands.

But it's essential to note that the element of art in marketing isn't going away. The opposite is true. “Data is definitely not the answer 100 percent of the time,” Shawn Skillman, a senior marketing data visualization analyst at SAS, explained. “It's just a mind-set shift where we can use the numbers to help us make better decisions than when we relied simply on our gut.” Science or analytics has actually enhanced the art elements of a campaign by empowering marketers with factbased decision making and insight beyond what they ever imagined.

Another important aspect to this shift is that the more we rely on the facts that analytics helps us uncover, the less biased we become about the kinds of campaigns we run as well as who we target with them. In other words, we don't want our gut to cause us to go places we could have avoided if we had first looked to the data for clues about what might work well and what wouldn't. “We have become outcome agnostic,” said Scott Sellers, a segmentation analyst at SAS who exemplifies the kind of analytical marketer we are grooming. “From an analytical standpoint, we don't have a dog in the fight. Our job is to show someone how we got the data and how we interpreted it. My whole goal is to give logical reasons why we reached a certain conclusion. You may reach a different conclusion based on your interpretation, so let's have a conversation about that.”

Marketing has shifted from functioning purely as an outbound effort into one more receptive to a high volume of inbound opportunities. Customers and prospects are 60 percent of the way through their decision-making journey as they learn about our products and solutions through visiting websites, downloading content, joining communities, and chatting online, among other things.

Now we have a two-way interaction with customers as they proceed on their decision journey; there is much more emphasis on inbound marketing efforts than the outbound activities of the past. But inbound marketing demands a whole new set of analytics that help us understand the experience and behavior of the customer. “It's not about building lists anymore,” Fulk told me. “It's about putting the customer's behavior at the forefront and thinking strategically from there.” It's also about learning to leverage both inbound and outbound efforts to maximize our chances of connecting with our customers in a way that they're most interested in. We still need to get out our outbound messages, but by fine-tuning where we target the audience, we can then stimulate additional inbound opportunities because we've sparked someone's interest. If we write a post for a popular blog on, say, customer intelligence, that might then spur a potential customer to visit our site later on to learn more about our offering on the way to becoming a solid sales lead. “Your outbound and inbound efforts have to be symbiotic,” Jennifer Chase, a senior marketing director at SAS, said. “You won't have any visitors to your website, for example, unless you are doing the necessary outbound activities to drive them there.” And, then we have to have the relevant inbound strategies to follow up.

When you combine outbound marketing data with inbound marketing data, and then apply advanced analytics, for instance, you begin to shift to a more proactive marketing strategy that measures value. Marketers are then enabled beyond gut instinct toward fact-based decisions. They are not simply making static recommendations, but instead personalizing offers based on behavior and demographics. We no longer have to make assumptions about what customers are doing, because we have the data pieces to better understand where they are in their decision journey and what they might want to talk to us about.

In the big-picture context, marketing has shifted from a function that involved saying what you wanted to say into a mechanism that communicates what the customer actually wants to hear. Put another way, by using marketing analytics, we no longer have to rely on just telling everybody about everything we are doing. “It used to be that marketers would spend their time and energy promoting their own products and events to whomever would listen,” Julie Chalk, a marketing manager at SAS, told me. “We have now flipped this model. Using analytics we now start by analyzing the audience and seeing what the data tells us. That way, we can understand who we are trying to go after. Instead of just pushing out the message we might want to say, we can now meet people at whatever stage they are in their decision journey.”

Marketing Analytics at Work

How Analytics Empower Campaign Agility

A common practice in traditional marketing is to first choose a target market to focus on. You then align your organization's strategies and messaging to create a campaign in that target market. But what happens when it becomes clear that the campaign you created isn't working? How agile are you in terms of adjusting on the fly and adapting to the needs of your prospective customers?

The Challenge

A campaign we ran at SAS was targeted at small to medium-sized businesses, or SMBs. We felt we needed to come up with some tailor-made messaging for this group, which would be distinct from similar campaigns we were launching targeted at larger, enterpriselevel companies. To do that, we highlighted what we thought were business needs, language, and case studies that would resonate most with the SMBs.

But after the program launched and began running, the results were disappointing: we saw lower than expected results for performance metrics, including click-through rates, conversions, and conversion rate. So we tweaked the messaging, offers, and program structure in ways we thought might improve results. After crunching the numbers again, the results came in: the campaign was still floundering.

We were now forced to take a fresh look at the situation. What had we done wrong? After some reflection, we also came upon an even more telling question: Did we actually need to separate the SMBs from the larger organizations? We had begun with an underlying assumption that we needed to treat the SMB market differently. Had that been a mistake?

The Approach

To help guide us forward, we selected a roster of key performance metrics to analyze:

• E-mails sent
• Open rate
• Click-through rate
• Opt-out rate
• Conversions (those who filled out registration forms to receive the promoted asset)
• Conversion rate
• Lead-generated SSOs (internal metric that measures the number of conversions we score as leads who progress to become a sales opportunity)
• Rate of completed leads to lead-generated SSOs (internal metric that measures the percentage of conversions scored as leads who progress to become a sales opportunity)

We then looked at how the SMBs responded to the SMB-specific campaign compared to how they responded when they received the enterprise-level messaging.

The Results

Much to our surprise, the SMB targets responded more strongly to the enterprise-level campaign (see table 2-1). Our assumption had been proved wrong. So we adjusted by closing down the SMB-specific campaign and then retargeted the SMBs with our enterprise-level messaging instead.

The key takeaway for us was a reminder that we cannot afford to let our own assumptions about the market hinder our ability to adjust to the needs of our customers. In this situation, we relied on the power of analytics to provide the answers about what people wanted rather than continue on in a losing cause.

 

We can meet customers along their decision journey by relying on advanced analytics, which can increase the quality of a marketing campaign by using tools like scoring, optimization, and predictive capabilities. The standard spreadsheet-based reports that marketers used to rely on to see how their campaign performed have now shifted to interactive visualization dashboards they can use daily to track the efficacy of their campaign, while making changes on the fly when necessary to ensure a campaign is making the most of its potential. The biggest difference is that marketers now have these tools at their disposal; we no longer have to submit requests to the IT department to get this information (we will dig into the partnership between marketing and IT in chapter 3).

Embrace Marketing Agility

With analytical tools, we can now better target and personalize our outreach to customers as well. We no longer have to put ads for our products everywhere. Rather, we can target people who likely want to see ads for our products, which results in more effective campaigns and investments, and a better experience for the customer.

Consider an example in which we ran a campaign for one of our software products. In the past, when we might have run a direct mail campaign, someone like Julie Chalk did analysis on the back end to determine who responded to the campaign and who didn't. Meanwhile, the campaign team simply moved on to its next project, and everyone learned from the experience. But that's all changed. Now with digital campaigns, Chalk is able to monitor the effectiveness of a campaign in real time. If we're trying to encourage prospects to sign up for a webinar, for example, Chalk can analyze the kinds of behavior those prospects are exhibiting when they show upat our website. She might see that some 26 percent of people who land on the sign-up page decline to register for the webinar, which then gives her license to ask why that happened and what adjustments we need to make to increase that conversion rate. “I love this transition from just looking at reports the next day to being able to keep trying and adjusting,” Chalk told me. “We have so much data now available at our fingertips. And it's fun to experiment with ways we can use it to make the best decisions possible.” Customers are always on, and so are we.

As another example, it's great to know that a prospect who attended a live event converted to a sales lead and, ultimately, a sales opportunity, which is our phrase akin to a “sales pipeline.” What is more insightful, however, is learning to understand the behavior that specific contact exhibited before and after the marketing live event. This is the practice of what we call “pathing analysis,” meaning “what path did a customer take to turn into a sale?”

FIGURE 2-3

By looking at figure 2-3, which illustrates the activity a prospect took on her way to becoming a customer, which we call “the happy path,” we can answer such questions as: Did this particular contact download a white paper to do her research before investing her time and energy into attending a vendor-sponsored live event? What happened following the live event? Was the sales organization the only team that engaged with the contact? Did the contact register for or attend additional marketing events to further and fully vet the product she was interested in purchasing? If so, what was the logical trail of marketing events that made this contact comfortable enough to agree to become a sales lead and ultimately a sales opportunity? Skillman explained that “with pathing analytics, we are able to peer into the behaviors of our current and prospective customer base. Displaying a flow of data from one website or event to the next can enable us to capture and gauge interest in a product or solution. The value is to model behavior for opportunities, such as a win, loss, or sales pipeline, and superimpose that profile on top of our current customers to influence the desired outcome. We can reduce churn and better enhance content marketing and make the customer journey more efficient.”

By using tools and analytical approaches like pathing analysis, we can get answers to such questions as:

• Which marketing interactions are most useful in helping close a sales opportunity? In creating brand awareness? In generating useful leads?
• Did the contact have multiple touch points with marketing before a sale opened or closed?
• What behaviors are outliers but still useful for understanding how well marketing efforts identified, targeted, and/or nurtured a contact along his journey?

What becomes complicated, though, is that we can't get trapped into giving too much credit to any one action we have taken, something we marketers call “attribution.” In the past, whatever led to the last click or contact got all the credit for a customer conversion. But in a multichannel world, it's no longer that simple. Every companystruggles with attribution, because assigning credit or allocating marketing dollars to activities that seem to be working more than others can get tricky. But it's difficult to assign credit or give attribution to any one activity. It's often a mix of the many things you are doing, both outbound and inbound, that works in combination to turn a prospect into a customer. The answer, then, is that we need to keep pushing with our understanding of the paths customers use, and then use our analytical capabilities to change, along with how the customer is changing. “Attribution must take into consideration the cost for the marketer as well as the cost for the customer,” Skillman said. “Cost being not only hard costs, but also soft costs, such as time to travel and time spent engaged with a website, webinar, live event, and so on. Each company will have specific needs to complete a true attribution puzzle. One may start with a first-touch model or a model that gives ‘credit’ to each touch point to start. This will paint a very rudimentary picture from which each business can tweak and add in more complex business requirements.”

We relied previously on static information to launch a marketing campaign. Say, for example, we knew that a potential customer once downloaded a white paper on a piece of technology we offer. We used the information we gathered on that customer to keep her updated on that same technology. But what if that customer's needs and interests changed? What if, by sending along the same information, we were now annoying that potential customer? On the other hand, if we use our dynamic data and analytical tools to find out what that potential customer is currently interested in looking at, we can then adjust the kind of information we are sending her in a campaign that's relevant to her current interests.

Again, this is a big switch from how we used to launch campaigns, a practice you might call mass marketing. We had a databasemarketing analyst create a list of people to market to solely based on guidelines the campaign manager provided. The analyst would then build the list using those criteria, which might involve variables like someone's title, industry, or purchase history. When the list was complete, the campaign manager would then implement the campaign. But what if the criteria were outdated? How effective would that campaign be?

Today, we can use our latest data to track everything from how often someone has been on our site, what support tracks he might have open, which classes he has attended, what campaigns we have already marketed to him, and more, all of which help build his dynamic profile. Even if the list of names we build for a campaign is smaller than in the past, the results far exceed what we once got.

Making this shift can actually be difficult for some marketers who still struggle with the notion that their list isn't big enough. But that reaction is based on the kind of response rate they have become accustomed to. Let's say you wanted to get fifty people to respond to a particular campaign. How many names would you need to get that kind of response: ten thousand, fifty thousand, or a hundred thousand? Who knew? But today, using data and analytical tools like scoring, segmentation, and modeling, you might be able to get a near 100 percent response rate to a campaign.

Another example involves traditional print advertising. Running a single ad in a national magazine might cost $30,000 to $50,000. Not only is that extremely expensive, it is also difficult if not impossible to measure the effectiveness of those ads. Yes, there are ways to try and capture how someone reacts to an ad, like including a URL on it or a special phone number to call. But how many people actually respond to those things?

When you launch a similar digital campaign, you not only can reach a broader audience, but also can literally track how people respond to it. I admit that it took me some time to adjust to the possibilities this new approach offered. But the more I saw the effectiveness of analytics, the more we began to push them throughoutour organization. And the better we got at embracing analytics, the better stories we were then able to tell to our leadership team, which helped us build internal support for getting to the next level of analytical techniques.

In building an analytical culture in marketing, you are essentially creating the foundation for operationalizing all of the different types of analytics that you will need to inform, evaluate, and drive decisions. If you want to operationalize all of the various forms of analytics for the various audiences, then you must put those analytics to use. For our organization, we exercised our analytical muscle across the entire global department; analytics are on every desktop and are part of every conversation. You don't need a single analytics approach; you need several that build off each other and continually challenge your organization.

Some of the most powerful uses of analytics involve marketing optimization techniques through which we can dramatically decrease opt-out rates and increase conversion rates. Optimization and modeling are pivotal in campaign design, list segmentation, and, ultimately, campaign execution. Scoring and web analysis anchor inbound nurturing campaigns, allowing us to better engage website visitors who arrive via search marketing and other sources. If marketing can see the digital dialogue and footprint a prospect has brought along with her on her journey, the opportunities are endless.

Scoring, for instance, helps us better discern how well a lead will convert to a sales opportunity. In the past, we considered anyone who performed a predefined action on the site a lead and passed that lead along to sales. While seeming like a good idea at the time, it was really more of a volume game than one based on precision. As a result, sales had to sift through these leads to find the best ones. Now, by using our data and analytical tools, we can do a lot of that prescreening before we ever send any lead to sales. We have developed different sets of rules to help us answer the question of how good a lead someone might be, such as evaluating the prospect's job function, the size of his company, and his title when he registered, for example, to download a white paper. Based on the rules we have established, we can then assign a prospect a score that tells us the most appropriate actions we should undertake with him.

As you can see in figure 2-4, if a prospect scores highly on both her implicit actions and on her profile, we consider her ready to hand over to our sales team. If the prospect scores well on only one of the two valuations, then we move her into what we call a “nurture” program, which essentially means that we will follow up with some relevant e-mails and messaging to learn more about how she might evolve into a stronger sales prospect. If someone scores low on both counts, we'll then put that contact on a “watch” list, where we might set an automatic reminder to follow up with her in three months as a way to reevaluate her interest. We have also employed more sophisticated scoring models that align to the performance of certain channels and preferences.

Events and users groups are another area within our scope of marketing offerings that have been completely transformed by analytics. Before we dove deeply into the world of analytics, many of our marketing efforts involved hosting events as a way to drum up interest in our products and solutions. They seemed to work well: it was a great opportunity to connect with prospects and customers face-to-face and find out exactly what they might be interested in. Event marketing remains a critical channel for reaching customers and prospects; nothing can replace human interaction.

But events also have a downside. They cost a lot to organize in terms of both time and money. And while we could collect metrics on which attendees eventually bought a product, calculating the return on investment (ROI) from a live event wasn't always clear. To be sure, there were less tangible brand-building benefits we could collect from hosting events. But we began to ask ourselves if that was good enough anymore, now that we had seen the potential of our digital campaigns. “We began to wonder what we could accomplish if we shifted resources from events to more paid search,” Chase told me. “Instead of reaching a few hundred people at a time, we could be reaching thousands instead. It became a question of scale and how could we speak to more of the market.”

As an example, our field marketing team had been implementing nationwide road shows beginning in 2013 and continuing into 2015 to raise awareness and generate leads for one of our flagship products. But in May 2015, we analyzed how effective those road shows were to better understand the impact of this extended campaign and whether we could spend resources in more efficient and effective ways to generate interest in the product.

As a way to measure that, we asked questions about the data such as:

• What does it mean in this context to ask, “Are the road shows effective?”
• What were the registration and attendance numbers?
• What were the no-show percentages?
• How many leads were generated?
• How many sales opportunities (SSOs) were generated from those leads?
• Did the contacts come to the events with a preexisting SSO or did those deals open as a result of the event?
• Do certain road show locales perform better than others?
• What was the return on investment?

Once we knew what questions to ask, we ran the numbers in several ways to account for one big deal that had skewed the numbers. We ultimately found that:

• Only one road show was associated with multiple wins (two), so there were no obvious geographies where we should focus our attention.
• Only two associated wins contained related products.

The end result of that analysis—which showed a relatively low ROI from road show–style events—led our organization to make a somewhat radical shift in the number of events we host each year, cutting that figure from 190 to 80, a decision that made some members of our sales team somewhat uneasy. Fortunately, as an analytical marketing organization, we have developed a sense of trust and partnership with sales, a dynamic we will discuss in more detail in chapter 3.

Embracing analytical marketing tools and techniques not only gives you better results, but also allows you to become more flexible and responsive. The ability to make decisions more quickly, alter investment strategies, change channels, adjust volumes, and test newapproaches means that as an organization, you will be more relevant. By making the mind-set shift to embrace analytical thinking, you also gain the ability to make adjustments on the fly; you don't need to wait six months to see whether your campaign was effective or not. The data and analysis will tell you in real time what's happening, which gives you the opportunity to make changes and tweaks, also in real time, based on how customers are responding. That's powerful and empowering across the organization.

APPLYING THESE IDEAS TO YOUR ORGANIZATION

Is Your Marketing Organization Agile Enough to Support an Analytical Mind-Set?

You and your team should answer the following questions to see where your marketing organization is currently positioned along the analytical spectrum. Use this opportunity to assess how you treat your data, what tools you're employing to leverage insights from that data, and then assess how nimble you are to take advantage of those insights. While there are no right answers here, your responses will shed some light on gaps and opportunities for your organization to pursue.

✓ How long does it take to design, test, and execute a digital campaign?

✓ How many people and how many different systems are involved in delivering marketing campaigns?

✓ How quickly can you respond to a market-based hot topic?

✓ How quickly can you respond to a customer interaction or request?

✓ How long do you take to adopt a new channel? Do you have any focus on emerging channels and a testing plan?

✓ How long do you wait to find out results?

✓ Do your marketers have access to all of the data and analytics they need to make changes to their campaigns?

✓ How risk averse is your organization?

✓ Do you reward or punish failure?

✓ How quickly does your marketing organization adopt change?

✓ How consistent is your customer experience across the company?

✓ What is the level of marketing's dependency on other departments?