Hitmetrix - User behavior analytics & recording

The Right Data Can Drive Customer Action

Predictive analytics gives marketers a view to the possible. Demographics provide a basic understanding of customers. Behavior data suggests customer intent. But what marketers really need to know, with certainty, is how to change or drive customer behavior.  And, for that, not just any data will do.

Is one type of data best suited to initiate customer action, or do marketers have multiple options to choose from Twelve marketing insiders divulge which data they assert can spur customers to action, and explain why. 

Scroll to read their insights or click on the image above to download the eBook.

Liz Buderus, VP, Product Management, Epsilon

As human beings we’re drawn to personal relationships. We come into these relationships because in spending time together, we develop a connection. Over time, we learn more about these people, relate to them, and develop deeper connections.

Today, big data allows marketers to create connections with customers the same way: by listening, responding, and growing together. To do this effectively, marketers need to leverage data appropriately dependent on the stage of the relationship with that customer. Delivering personalized experiences is something that should only be developed based on mutual trust between a brand and the customer it’s trying to reach.

At the beginning of every relationship, you need to spend more time asking questions and listening, than responding. During this stage in marketing, leverage general demographics about your customers: where they’re located, age, gender, etc. Match all of this data to what they’ve purchased from you to help round out your understanding of them.

This approach is best done for newly acquired customers that you know the least about.

Once you’ve listened, it’s time to learn more about your customer and understand their needs. Compare different groups of customers based on their customer profiles (e.g., dress buyers versus new product announcement buyers) using demographics, general personas, interests, financial information, and competitor spending to paint a comprehensive image of each group to truly understand their differences.

It’s during this stage you can entice action by using data to deliver relevant, one-to-one messages.

Now, it’s time to make the relationship more meaningful and show your customers you understand them better than they understand themselves. Entice action by using data to deliver marketing messages around specific needs you know each customer has (or will have). As a guide, look at companies like healthy living website Lifescript, which saw a 30% lift in conversions after implementing targeted messaging.

The most effective way to use data is to create customer personas that take your first-party data (purchase, browsing, opening, clicking, device, etc.), demographic data, and third-party spending data into account.  These personas will help explain the different types of customer relationships you have.

Once done, focus on driving home the imagery, the words, and the products and services for each group so they will say “You get me, and I get you.” 

Daniel Ziv, vice president, customer analytics, Verint

Structured data is convenient for tracking metrics and KPIs across the enterprise, but it sometimes falls short in explaining customer behaviors and driving action. Instead, taking a look at unstructured data can offer a different view and provide a better platform for influencing customers.

Mining unstructured customer interactions (such as phone, email, and chat) that are linked to specific outcomes holds the key to influencing customer action. Customers tend to be more detailed and open when speaking with another person, so these interactions produce a richer source of data and more actionable insights.

For example, an international bank wanted to better understand what drives customer purchase behaviors. Analysis of transaction data indicated that about 15.1% of upsell attempts ended in a successful purchase, but the factors driving customers to accept or reject the offers wasn’t clear. Using speech analytics, the bank analyzed the actual words and phrases used in the calls that ended with a sale, comparing them with the phrases mentioned in non-successful calls. It discovered that agents asking for “another moment of your time,” before the upsell offer was actually driving a negative customer response. Only 6.3% of calls that included this phrase ended in the successful sale.

This analysis also surfaced phrases that were driving positive customer response, such as, “You could be earning X dollars in interest on the balance in your account; would you like me to set that up for you?” Calls using this alternate phrasing earlier in the call resulted in 57.6% sales success rate, which almost quadrupled the average customer acceptance rate.

Unstructured data is not only richer, but also more readily available. It’s estimated that more than 90% of the digital universe is in unstructured formats (audio, text, images, and video). The analysis tools for mining unstructured data are significantly more accurate and effective than they were just a few years ago, opening up a new world of possibilities and actionable intelligence.

Eric Duerr, CMO, Rocket Fuel

The Holy Grail for digital marketers is reaching the right customer on the right device with the right message. Many advertising companies claim they are able to deliver this. But the reality is more complex than the promise: It takes sophisticated technology, leveraging both artificial intelligence and big data, to reliably and accurately predict drivers of performance and consumer action. An effective predictive data set can include more than 11 million demographic, behavioral, and contextual combinations of attributes.

Real marketing value resides in using all this data to pinpoint moments of influence — the exact point where the person, the device, the context, and the intent come together with the highest probability of lift or conversion. This is driving the next wave of data-driven marketing.  

To identify the right moment, marketers need to activate the wealth of first-party data they own, and combine it with the attributes that are driving — and are predicted to drive — results. A unified consumer profile combining first-, second-, and third-party data provides a complete view of the customer to identify the precise moment to deliver ads, minimizing the wasteful delivery of ads to the wrong devices and people at the wrong time. Marketers can then tap into data generated in every channel simultaneously, which surfaces up to 200% more data than a standard profile. By focusing on moments, brands can improve direct response performance by more than 30% on average, and overall brand campaign reach by more than 50 percent.

Moment-driven marketing affords brands a much higher level of precision than targeting customers across devices, using data to identify the precise moment to deliver ads that will influence customer behavior. Using technology to score every moment, marketers can learn the predictors that make one ad more appropriate than another in a particular moment. The technology then learns over time what attributes work to achieve a marketer’s goal. Microsoft, for example, recently used a data-driven campaign to drive qualified job applications to it careers website. Using cross-device technology to make better decisions about who to target and when, Microsoft was able to beat its CPA goals. One key insight discovered during the campaign was that reaching job seekers at the right moment on their mobile delivered much better conversion. With a cross-device and moment-driven approach, execution, and measurement, brands are in a much better position to influence customer behavior and action.

Susan Bryant, CMO, DialogTech

One of the most important types of customer interactions driven from digital marketing is often ignored: phone calls. In today’s mobile-first world, digital marketing drives phone calls, and those calls drive revenue. Because of smartphones and click-to-call, this idea, which may have seemed crazy to marketers as recently as two years ago, is well understood today. 

Consumers today live their lives on their smartphones, and marketers are allocating more of their digital advertising budgets to target them. According to eMarketer, 62.6% of digital ad spending in the U.S. this year will target smartphones and mobile devices. Thanks to click-to-call, these ads will drive more than 108 billion customer calls to businesses, and those calls will influence more than $1 trillion in consumer spending. That’s why call attribution — knowing how effectively digital marketing programs generate calls, sales pipeline, and revenue — is the next critical step to optimizing digital marketing performance and spend. 

Marketers are getting much better at analyzing attribution values for their digital channels. But they’re still missing an important part of the customer journey. Marketers need to know the marketing channels driving their calls (such as search, social, display), what search keywords the person used (if they came from search), the ad they clicked on, and the caller’s path through a website before and after calling. Marketers also need data on the customer and the call — who they are, their geographic location, time and day of the call, how long the call lasted, and whether it converted to a sale. This data is important in optimizing campaigns and spend. 

For example, HotelsCorp, uses customer call attribution software with its paid search campaigns to track callers back to the search engine, keywords, ad, and landing page that drove them to call. The vacation provider also analyzes callers’ geographic location and the times and days that generate the most calls. Using call attribution technology, HotelsCorp can optimize bids for the keywords, locations, and times of day that drive the most — and best — calls. By using call attribution data, HotelsCorp generated 83% more calls and 71% more bookings, while decreasing cost-per-conversion by 10%. 

Getting consumers to call is one of the most powerful ways marketers can impact customer acquisition. Call attribution data provides the insight marketers need to do it.

Arthur Hall, multichannel consultant, Quad/Graphics

In today’s complex marketing environment, establishing links between demographic, social, psychological, and cultural characteristics is more critical than ever before. As individuals become increasingly unpredictable and customers continue to diversify, strategic segmentation can provide the insights needed to improve reach, sales, and return on marketing investment. This qualitative data approach is based on individuals’ needs and expectations and on the reasons underlying their behavior, rather than specifically on what they buy or what they do. Marketers that uncover an in-depth understanding of consumers while identifying the factors driving behavioral patterns are equipped to expertly influence creative design and copywriting, which increases relevancy and improves campaign performance.

Consider a retailer that analyzed customer data and built affinity groups based on product purchases. In one group, a head of the household (woman A) buys work clothes for men on occasion and frequently purchases women’s nursing apparel; another head of household in that group (woman B) looks similar based on past purchases. Using data that is too basic may depict false identical characteristics between these two consumers, or inaccurately exploit certain shared qualities. For example, woman A does not want to spend time researching products; she wants the best deal and she wants to make a quick purchase. However, woman B is more interested in learning about products, perhaps knowing how it was produced or its environmental impact.

With an attitudinal and behavioral marketing view, the retailer was able to identify the types of messages and information that would drive consumer response. That view also provided insight into the most effective direct marketing format for personalized, timely engagement. Woman A, who was interested in pricing and general product information, benefitted from an email approach that was less expensive than producing and delivering a mail piece. Woman B, who desired to consume information before making a purchase, received an innovative print format that provided the necessary space for product features and benefits.

As a result of understanding what imagery, copy, and mechanism would influence each woman’s buying decisions, the retailer generated more than $9 for every $1 spent on the marketing campaign; nearly a third greater than the average high performing sell-over-marketing investment.

Loretta Jones, VP of marketing, Insightly 

Companies that use predictive analytics can easily anticipate customer needs, allowing those businesses to take a proactive approach to customer service, rather than reacting to problems as they arise. In addition to existing internal customer data, predictive analytics vendors are bringing in external data, such as social profiles, funding press releases, and data on sites likeTechCrunch, to give users even more accurate buying signals from prospects and churn signals from existing customers. Having this data easily accessible can help businesses analyze it to identify customer behavior patterns that will help them make recommendations for services, products, or upgrades. At Insightly, we use a customer success application to determine the health score of our customers, their use of the product, and their likelihood to either purchase more licenses or upgrade to a higher plan. We’re also experimenting with a predictive analytics application to determine the likelihood of customers on either free accounts or on 14-day trials becoming paying customers. These applications allow us to proactively engage with those customers to expand their use of Insightly or proactively prevent the loss of that customer.

Joe Pino, director of client insights and strategy, Clutch

It’s no longer a best practice to personalize based on behavioral data; it’s a customer expectation. Today, nearly 80% of customers expect their experiences and engagements to be personalized based on their previous behaviors. Businesses that are not delivering a data-driven brand experience are already falling behind, and those that do are realizing tremendous benefits in the form of higher response rate, purchase frequency, positive brand perception, and more.

Understanding consumer behavior across every touchpoint is paramount in effectively motivating customer engagement and sales. From point-of-sale systems and e-commerce platforms to mobile applications and social accounts, customers are continuously interacting with brands and leaving behind digital footprints that can provide insights on their preferences and tendencies. Leveraging that data means having the ability to more accurately serve up engagements to the right the audiences, in the right channel, at the right time.

Behavioral data empowers businesses to understand and align customer preferences with brand experiences. While synthesizing cross-channel information can be challenging, data-driven insights provide opportunities for increased responses and returns. Marketers who successfully analyze behavioral data can understand individual preferences and tendencies like never before through segmentation and scoring. Marketers can use this intelligence to deliver offers and communications that mirror the interests and behaviors of each customer, resulting in higher levels of engagement and brand loyalty.

Putting the power of behavioral data to the test was Crabtree & Evelyn, which in November 2015 created a cross-channel campaign spanning email and direct mail. As part of its campaign, luxury personal care products company customized offers based on preference dimensions that included product type and specific product purchases. Dynamic email campaigns that featured personalized content based on established preferences were then tested against a generic one-size-fits-all campaign.

The personalized campaign elicited positive responses, generating double the response rate of the generic offer with an average order value increase of upwards of 15%. Additionally, the personalized direct mail campaign resulted in an increased response of 20% over the generic offer.

Victoria Godfrey, CMO, Avention

Combining various data sources to create a holistic, single view of customers and prospects can help B2B marketers and salespeople better identify those who are more predisposed to make a purchase and then focus efforts on them. By first gathering a variety of data sets and then applying predictive models, marketers can identify the characteristics of customers that have converted in the past, and then use that data to target similar profiles of customers likely to buy in the future. These signals lead marketers to understand what business behaviors are linked to the prospects most likely to buy, allowing them to target companies exhibiting these behaviors with messaging and content tailored to these attributes. For many B2B marketers, this starts with aggregating basic firmographic variables — such as industry, revenue, and number of employees — to provide sales with a better view of prospects and an indication of their likelihood to purchase. Breaking this data down further by looking at triggers such as whether a company has received recent funding, launched a new product, or is hiring, can provide even deeper insights into the right timing and message to close a sale.

For example, the Employee Benefits Association (EBA), a health and life insurance agency, used to take five days to manually research potential contacts. In one case it created a list of 170 contacts, only to reach out to 40 prospects and make no progress even after following up. Using outsourced business insights services, EBA was able to be more targeted in identifying the most qualified prospects that are also most likely to convert, based on predictive indicators built on business signals and ideal customer profiles. As a result, EBA was able to focus its efforts on those high-potential prospects, narrowing its target list to 46 companies, which, in the span of less than a year, yielded a 26% response rate and seven in-person meetings, including one that converted into a new client.

Kylee Hall, senior director Leadspace

Every B2B vendor is trying to deliver the right message to the right person at the right time. B2B marketers can use predictive analytics to help. Predictive analytics enables marketers to analyze behavioral, demographic, and firmographic data to find, score and segment relevant, time-specific buying signals. This is a huge step forward for B2B demand generation compared to list buying and batch-and-blast marketing.

But arguably more important is the quality, accuracy, and completeness of the data marketers use for their predictive analyses. It doesn’t do any good to identify someone ready to buy your product if you can’t find them. Nor is it good to dump leads into your CRM system if they’re out-of-date, inaccurate, or missing key attributes needed, for instance, to match a lead to an existing account already in your database.

Consequently, the best type of data to drive customer behavior is real-time data, continuously updated, segmented, and scored with superior predictive analytics to match your ideal customer, and put into action in your CRM or marketing automation systems. It’s not a simple answer, but it’s not a simple problem. If there is a weak link in the chain, everything falls apart.

One good example of how it all works together is ObservePoint, a B2B tech company offering QA tools for the web. Its SDR team used a predictive analytics platform to discover net-new leads that best match the company’s ideal customer profile, based on firmographic and demographic data aggregated from multiple sources, combined with signals gleaned from the open and social Web. The immediate result was a significant increase in the flow of qualified leads, and a boost in conversions.

ObservePoint saw dramatic, measurable improvements in their demand-gen process:

·  30% increase in SDR productivity

·  40% decrease in contact acquisition research time

·  23% increase in connection rate

·  15% increase in net-new pipe

·  8% increase in deals closed

·  5% increase in revenue

Based on those results, Doug Jensen, VP of Sales at ObservePoint, calculated the ROI of predictive analytics at 640%.

Josh Reynolds, head of marketing, Quantifind

Marketers today need to understand and improve their impact on revenue. Data that reveals why certain marketing strategies do or don’t work, as well as how to adjust them to improve their outcomes, is essential to marketers.

A well-known QSR brand wanted to understand how to improve its morning sales, particularly among teens. The brand’s breakfast menu included several options aimed at teens, and data indicated that young people loved the brand’s content marketing. Yet breakfast sales weren’t improving. 

The QSR brand aimed to solve this mystery by combining online consumer data with financial data from in-store receipts. Neither traditional social listening nor predictive analytics tools had been able to leverage this data to reveal why breakfast results continued to disappoint, however.

Using an on-demand insights platform, the QSR brand filtered out all the spam and other social noise that didn’t represent real people, then filtered again to isolate only the consumer conversations and topics that correlated with the brand’s sales. This process revealed a strongly negative correlation between the QSR brand’s breakfast sales and moms’ online conversations about its coffee. Through further investigation, the data began to tell a story: Teens wanted the brand’s breakfast but relied on moms for rides to the drive-through — but because moms didn’t like its coffee, the teens had to choose somewhere else for breakfast. So, to unlock the untapped breakfast sales potential among teens, the brand would have to offer moms better coffee. 

This was the answer the brand needed. It accelerated the launch of a new coffee product, and both coffee sales and breakfast sales grew significantly, with coffee sales doubling. 

Matt Riley, CEO, Swiftype

One of the most important stats marketers fail to utilize is the type and number of queries that users are typing into your site search box. Defined as “intent data,” users are explicitly telling companies what they’re looking for. As the number of meaningful queries increase, marketers can more accurately forecast conversions and revenue. For example, BulbAmerica, which is the largest e-commerce store for light bulbs and fixtures in North America, uses site search analytics to predict what people are going to buy. They found that users are 4.6 times more likely to convert after beginning with a site search experience.

After seeing the impact that ssearch data has had on conversions and purchase values, BulbAmerica, who has featured the search box more prominently, can now better predict how much inventory is needed in stock at any given time. Because the Bulb America team can more accurately predict what people are searching for, it has also implemented a few optimization techniques that have positively impacted growth.

First, the BulbAmerica is team able to customize certain results to make sure that top selling or higher margin products are pushed to the top of the results page. The team also deletes results that don’t convert as well.  Finally, it creates custom result sets for queries that return no results, proactively guarding against blind spots where users might hit dead ends. These proactive steps have bore substantial fruit; users now spend 12.3% more per order on the BulbAmerica site.

If you’re marketing team is trying to identify low-hanging fruit for driving customer action, then look no further than site search analytics. They’re a goldmine of information and can help you predict and forecast results while giving you an opportunity to optimize your site in real time.

Andrew Dennis, CEO, NorthPage

Using traditional analytics to understand “what happened” with a campaign is no longer enough for modern marketers. They now need actionable insights from a new generation of digital intelligence sources to drive growing volumes of successful customer actions.

Building on the measure of what happened, marketers who leverage actionable digital intelligence can determine:

• WHY: Pinpoint the reasons why something happened or continues to happen (good and bad) in live digital marketing programs

• HOW: Prescribe guidance detailing how to fix or improve the currently deployed programs to drive performance and results

• WHO: identify who has leading digital marketing program capabilities across competitors, near neighbors, and the universe of brands to gain critical market and strategy perspective

This advanced level of intelligence enables marketers to focus on developing digital areas of opportunity rather than finding them.

For example, Sears began using digital intelligence sources to grow online revenue in its appliances business unit. Though Sears’ marketers were well informed on baseline activity through its integrated digital analytics program, they wanted objective insight and an action plan to drive increased KPI performance.  Diagnostic analysis surfaced average order value as a major area of opportunity — providing best-in-class examples for guidance on how to drive lift. Sears’ marketers acted on the digital intelligence indicating where and how to optimize their efforts, implemented specific changes, and immediately recognized a 6.7% lift in average order value.

Related Posts
E-Book Popup

Unlock the Secrets of Digital Marketing in 2024!

Subscribe to our newsletter and get your FREE copy of “The Ultimate Guide to Digital Marketing Trends in 2024"