Strategies for marketing the newspaper have evolved over the years as the concept of cost per unit of circulation, with its inherent incorporation of cost per order and retention, has taken hold. More recently, acceptance of tiered subscription pricing strategies and the valuation of subscribers based upon advertising revenue has added another level of complexity and opportunity.
Subscriber acquisition and retention budgets have to be optimized. It is a necessary element in a newspaper’s strategy to improve overall operational effectiveness. Today’s level of sophistication now requires enhancements to ensure that the best possible prospects, those most likely to subscribe and be retained, must be targeted. While demographics are a useful tool in marketing to consumers, many newspapers are discovering the benefits of targeting customers through their behavior patterns.
In its most simplistic terms, behavior modeling works in the following way:
As an example, a newspaper may develop a behavior profile on prospects that visit the newspaper’s website for sports information and then develops an acquisition piece based on this behavior. In certain instances, the timing in which a customer routinely pays on their renewal notices may lead to specific touch points or messaging for this type of behavior.
People are creatures of habit and by studying past behavior, one is able to predict future behavior – in regards to marketing offers, price sensitive customers (churners), cross-selling and up-selling – this presents opportunities for newspapers and other industries that employ these types of analytics. This permits newspapers to be more efficient with their marketing efforts.
The following are insights on how two newspaper companies have successfully applied the principles of behavior modeling and pursued the technology necessary to implement these practices as the cornerstone of their subscriber acquisition and retention strategy.
Behavior Modeling – The Beginning
The Lexington Herald-Leader dared to ask the question “Can you tell us what we don’t know?” As Julie Achauer, Senior Circulation Manager of Audience Development, The Lexington Herald-Leader explains, “The question was loaded since we knew that we had a pretty decent retention program going on and we were coming off two successful years of collections and retention efforts that finally put a tiny dent in our subscriber stops in 2007. Still, we could feel the tide turning and we needed to get ahead of the curve.”
She further explained that they had done everything we could with smart guesses, wild guesses, and pure luck as to why, based on the traditional analysis of stop codes and payment history, their customers were leaving. In the following excerpt from an article she authored she describes what came next:
"In the words of my grandmother …and probably many other notables, 'be careful what you wish for!' After a year of searching for the right provider for this information we began our journey in January 2008 and it has been one filled with a lot of 'what we didn’t know.'
It all sounded so easy; get the last five years of customer database information, retention programs, stop and start codes, payment cycles, advertising zips and sub zip zoning, ABC reports, preprints and a multitude of other database programs and codes along with demographic and psychographic data. Please do not be afraid, just be prepared.
Behavior modeling can provide a multitude of insights into a company’s past and current subscriber base. We specifically were targeting subscriber retention with acquisition programs as secondary. We have gained an understanding for subscriber interactions, lifecycle values to the company and the probabilities of future customer interactions. This is critical to the development of future resource allocations for retaining current subscribers and acquiring new ones.
“This tool allows the newspaper to find the best opportunities to grow or maintain circulation and the most optimal resource allocation for each…” |
Behavior modeling specifically targets our customers at the household level. It clearly defines our customers past and current behaviors. We’ve gained a retention analysis specific to our strategies based on our subscribers’ values. We will be able to better understand the variables between stop reasons, effective sales channels, geographic locations and how these influence subscriber buying decisions. We will be using this information to maximize acquisition efforts from the sales channels most effective in driving new subscribers to core subscribers. We now have a breakdown of our sales channels and their effectiveness on producing core subscribers. This should improve the way we do historical trending and sales pressure analysis for upcoming budgeting.
The list is long on what you can do with all of the information and analysis. It’s easy to get caught up in vast amounts of information. While it’s exciting to find out what you didn’t know, it can also be painful. There’s no room for egos or the faint of heart. We’ve learned to build pivot tables and models in house. We’ve already changed timelines, the look of our bills, retention campaigns, but again we did ask for it! We also know more; we know how to set a true value to our core subscribers; these are subscribers that pay four quarters consecutively without any lapse in payment. We have identified which customers pay always early, always late, always on time, mostly early, mostly late, etc, all of which improves our targeted retention calling and timelines. For example; we know that there are 9000 core subscribers that always pay late. Yes, that is correct they always pay late. We’ve built a targeted list that we will begin testing in July to determine how to get them to pay on time. We feel that the changes made in our billing will encourage earlier response time, however we will still need to “undo” the behaviors that have been built over the last five years or longer. We will call these 9000 subscribers and “disrupt” what they are use to with special offers that make it worthwhile to pay us on time. We are looking to save grace expense dollars on the average of four dollars per subscriber. This is just one of many changes we will be making. With each test you need a control group. Patience is required but we have to be flexible and be able to evaluate the results as we proceed. Our customer database is not always so accommodating.
We are building our loyalty programs to foster one-to-one relationships based on the subscriber values. We are developing new compensation programs for our contractors to better meet our retention strategies. We are preparing to test every single one of our existing retention calling campaigns with new programs." |
Of the subscriber retention solutions offered, The Lexington Herald-Leader focused on three main target areas for analysis for optimal resource allocation:
-
Revenue and Time Based Analysis: Helps to determine the revenue dollars per subscriber over the customer’s lifetime. This allows the newspaper to focus their retention efforts on those customers with the greatest ROI for the company.
-
Behavior Scoring: Creates a grouping of similar behavior customers so those customers can be marketed to in a like manner that has shown the most effective results. It identifies and segments the customers as Core, Churn and New.
-
Payment Analysis: Describes payment behaviors relative to the customers’ expiration dates and then classifies them as Early, On-time, Late and Always Late. This analysis guides the Lexington Herald-Leader in the administration of a cost-effective solution for their collection and renewal efforts based upon the behavior.
All of this is incorporated into a descriptive and predictive modeling tool which allows the newspaper to find the best opportunities to maintain and grow circulation.
Qbase was also used to create tables with drill-down capabilities and list formation of the subscriber groups that include: New, Reactivated, Retained, and Lapsed subscribers. Variables within these groups are cross-referenced against one another and compare and contrast subscriber attributes. This process created three subscriber-centric predictive models:
- ‘Core’ subscriber model to predict those most likely to lapse, as well as those likely to restart and move back to ‘Core’ status
- ‘Churn’ subscriber model to predict those most likely to go to ‘Core’, as well as those most likely to lapse and remain lapsed
- ‘New’ subscriber model to predict those most likely to go to ‘Core’ and those most likely to go to ‘Churn’
Julie Achauer listed the tangibles as “What this done for us:”
- Effective Retention Calling Campaigns:
- Concentrating on customers’ behavior helps to determine which customers are truly about to lapse versus those that are only paying late. The retention calling effort is spent on the customers that are at risk instead of those that are just late payers and permits development of tactics to address the various stop reasons.
- Strategic calling campaigns to move customers who always pay late to paying early or on time.
- Change the billing timeline to encourage payments before incurring a grace expense.
- Enhance the subscriber bills to encourage online payments prior to expiration date.
She summarizes; “For years, circulation just looked at the numbers and made assumptions to the best of their abilities. Specific market and subscriber behavior analysis is critical to the way we transform our business!”
A solution that is both comprehensive and designed for precise, personalized targeting
In 2005, the New York Times Regional Media Group embarked upon a three year strategic plan to construct a centralized consumer marketing database to:
- Manage all customer intelligence, including behavioral, transactional, and product usage histories.
- Build and administer all consumer marketing campaigns.
- Conduct performance analysis across various levels.
- Measure return on investment of all activities.
- Perform a broad range of reporting and analytical needs
The challenge facing NYTRMG ( this division of the New York Times Company, is comprised of 15 daily newspapers with a cumulative circulation of 550,000. The Regional Media Group also publishes 16 city/regional magazines, Web sites for all its papers and national sports Web sites)
management was that they needed a solution that was both comprehensive and designed for precise, personalized targeting unique to their needs across their markets. After research and determining their needs, they chose ASTECH InterMedia, a division of smartFOCUS, with their MAAX™ Solution Suite (Marketing Automation & AnalytiX).
NYTRMG deployed the systems’ capabilities to create and manage a comprehensive demographic segmentation of each newspaper market – including customers, former customers and prospects. All multichannel marketing campaigns – direct mail, telemarketing and email – are driven by NYTRMG’s segmentation model. MAAX facilitates personalizing and customizing communications based on the unique characteristics and potential of each household target. As an example, direct mail executions employ digital print technology to tailor graphics and content based on the newspaper market, transaction history and the unique PRIZM characteristics of the recipient. Using digital printers allows NYTRMG to gain substantial economies of scale by combining all affiliated newspapers into a single print run.
At The Ledger newspaper (Lakeland, FL) they implemented a readership growth strategy aimed at maximizing home delivery penetration in key geographic zones within The Ledger’s primary market. This integrated marketing concept has been used similarly in previous markets and is greatly enhanced by applying their Super Segmentation to micro-target their most attractive subscription offers.
The targeted growth model was formulated by a cross-department management team, consisting of the Publisher, and all members of the newspaper Operating Committee. The goal was to maximize core print product penetration in areas most valued by the newspaper’s advertisers, thereby achieving not only higher circulation growth in focus areas, but also driving incremental preprint advertising gains, and strengthening overall market share.
Within the “Super-Segment” framework, they established three distinct target subscriber groups: Younger (Light/Occasional print readers); Family (Busy/Weekend readers); and Mature (Price Conscious daily readers).
"Over the course of two years, 52-week subscriber retention rose from 22 percent to 32 percent - a 46 percent improvement." |
The campaign involved a carefully coordinated and integrated series of customer contacts utilizing outbound telemarketing, direct mail, email marketing, and product sampling. They also featured some of their most aggressive pricing. For example, certain segments were targeted with Sunday Only annual subscription offers, while others were targeted with quarterly Weekend subscription offers.
In Lakeland, since late June 2008 the multi-channel campaign has yielded a 5.2% response, which has netted 1,744 paid in advance subscription orders, with 63% enrolling in automatic renewal. Through twenty-six weeks of the program, they have increased the core product daily circulation average by 488 units of circulation, or +1.5%, within the Targeted Growth Model zones.
Across the NYTRMG markets, over the course of two years of implementing the MAAX system, 52-week subscriber retention rose from 22 percent to 32 percent - a 46 percent improvement. This improvement directly led to annualized source savings in excess of $1 million. In addition, the company has reduced its discount investment by 30 percent while increasing revenue per subscriber by 8.4 percent.
By combining PRIZM address coding at the zip+4 level across their entire address universe, and by layering other household demographic data along with transactional and product usage histories, NYTRMG have created a unique segmentation model. The model consists of seven “super-segments” within three “life stage groups.”
These 7 super-segments form the basis of their audience development strategy for growing reach across print and digital product lines, while also improving product development and advertising sales programs. The approach has enabled them to more precisely target relevant marketing messages, using the most appropriate communication vehicles or sales channels, and presenting the right offer that is likely to generate the best overall response.
In summary, it has long been accepted that database marketing provides marketers with a better chance for success. The three main components of any offer continue to be The List, The Offer and The Creative. Testing has proven that The List is the most important component that leads to the success or failure of a campaign. While demographic information and segmentation data from Claritas and Prizm have led to many marketing successes, overlaying behavior trend analysis will provide greater response rates.
For more information contact:
Julie Achauer, Senior Circulation Manager of Audience Development, The Lexington Herald-Leader. jachauer@herald-leader.com
Daniel L. Williams, Audience Development Manager, New York Times Regional Media Group at 863.802.7251 or email Daniel.Williams@nytrng.com