question answer

Optimal_Database_Marketing_Strategy_Development_an..._----_12_-_Strategic_Reporting_and_Analysis.pdf

Home >Business & Finance homework help >Marketing homework help >question answer

Strategic Reporting and Analysis

235

12 Due to the cutting edge analysis techniques brought forth by Keri Lee at Inside Source over the past 2 years, their subscriber base has grown by more than 25%. As a result, Keri was promoted to corporate vice president.

In her new role, no longer having an opportunity to oversee the day-to-day operations of the marketing team, Keri decided to request some key strategic reports to help her keep an eye on the health of the business. In particular, she wanted to see monthly counts of active subscribers and their key demographics. She also asked that LTV analyses for the various sources of new-to-file customers be performed on a semiregular basis. In addition, she was also eager to assess the long-term value to the corpor- ation of implementing a new retention program tested 2 years ago.

Realizing the workload implications of such requests, she created a new division whose sole responsibility was to create such key strategic reporting and analyses. She realized that this would not be a small undertaking and would require the dedicated efforts of a team. Because no database is 100% clean, inconsistencies in monthly counts would undoubtedly result and investigation would be required. In addition, she also realized that to properly calculate customer LTV, time would be required to obtain and then link each customer to actual cost and revenue figures associated with the various promotions.

Key to the establishment of proper strategy for any database marketer isthe tracking of customer counts, activity, demographics, and value over time. Without such analyses, a database marketer will be unable to make the best long-term decisions regarding name sourcing, offers, or treatments. Unfortunately, most direct marketers do not know how to properly assess customer information over time or determine customer value. This is mostly

Drozdenko12 2/26/02 6:17 PM Page 235

Drozdenko, Ronald G., and Perry D. Drake. Optimal Database Marketing : Strategy, Development, and Data Mining, SAGE Publications, 2002. ProQuest Ebook Central, http://ebookcentral.proquest.com/lib/nyulibrary-ebooks/detail.action?docID=996727. Created from nyulibrary-ebooks on 2020-07-20 13:41:10.

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

236 OPTIMAL DATABASE MARKETING

due to a lack of understanding by the direct marketer of the issues involved in the proper implementation of such strategic reporting and analysis. In fact, according to the 2001 Direct magazine annual database survey, only 27% of b-to-b direct marketers and 56% of consumer direct marketers calculate LTV of their customers. With respect to the catalog industry (both b-to-b and consumer), only 23% calculate the LTV of their customers, based on the 1999 DMA State of the Catalog Industry Report.

Undertaking the development of strategic reporting and analysis is by no means a small task. Dedicated and properly trained resources must be put in place to ensure success. This is especially true for direct marketers with complex business models. For example, a direct marketer offering only a single magazine title such as Aficionado will have an easier time of properly tracking customers over time than companies like Time-Life, Rodale Press, or The Reader’s Digest Association, in which each has multiple product lines and divisions offering an array of products. In fact, the Aficionado database may be leveraged more for advertisers’ communications to its readers than the offers from the publisher. Strategic reports would reflect subscriber counts by demographic groups.

In this chapter we discuss how to set up and establish some key strategic reports and analyses for your business. In particular, we discuss

♦ How to set up key “active customer” counts to help you monitor your customer base

♦ How to calculate various statistics to assist you in monitoring the health of your business

♦ How to establish LTV profiles to help you better understand the worth of certain marketing programs or sources of names

♦ How to calculate customer LTV

In addition, we also discuss

♦ How to set up impact studies so you can properly assess the impact of new product lines or profit centers on already established business units

♦ How to monitor promotional intensity within your company ♦ The importance of key coding names to ensure analyses discussed in

this chapter can be properly conducted

Key Active Customer Counts _____________________________ We begin our discussion of strategic reporting with the creation of a key active customer counts report. Such a report allows you to properly monitor your core customer base over time. These types of reports show the number of active customers by each major product line, magazine title, service offered,

Drozdenko12 2/26/02 6:17 PM Page 236

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

key demographic group, or source of customer. They prove valuable for monitoring your core customer base, allowing you to observe trends over time.

In the case of a magazine publisher of two niche magazine titles, the key customer counts they want to create and track over time are the

♦ Number of customers receiving own copy on Title A ♦ Number of customers receiving own copy on Title B ♦ Number of active donors of Title A (those giving the magazine as a gift) ♦ Number of active donors of Title B (those giving the magazine as a gift) ♦ Number of active recipients of Title A (those receiving the magazine

from a donor) ♦ Number of active recipients of Title B (those receiving the magazine

from a donor) ♦ Number of customers receiving own and active donor of Title A ♦ Number of customers receiving own and active donor of Title B

Each category can then be broken down further by key demographics to assist in monitoring, for example, readership age.

For a direct marketer such as Rodale Press, which offers one-shot products, they need to decide what constitutes an active customer. Defining active customers is not as straightforward for these direct marketers as it is for a magazine publisher.

Consider a direct marketer offering one-shot books and videos. Such a direct marketer may decide to define an active customer as anyone who has made any payment (greater than or equal to $0.01) within a certain amount of time for the product line of concern (e.g., within 24 months). For this example, the key segments to track include

♦ Active within the past 24 months on the books product line ♦ Active within the past 24 months on the video product line

The direct marketer may also want to assess the counts from a corporate point of view. For example, active from a corporate point of view may mean any payment on any product line within the past 24 months. Or they may decide it to mean activity of any kind (payment, order, address change, etc.) initiated by the customer. Either way, this is a decision the direct marketer must make.

For continuity or club direct marketers, active customers may be defined as “still in club” or “paid shipment made within a certain amount of time.”

Key active customer counts should be generated as often as needed. They can be generated weekly, monthly, or quarterly, depending on the promo- tional activity of the company. Companies that promote customers weekly want more frequent count reports than those that promote customers monthly or quarterly.

Strategic Reporting and Analysis 237

Drozdenko12 2/26/02 6:17 PM Page 237

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

238 OPTIMAL DATABASE MARKETING

In addition, companies with very large databases in the millions may find it beneficial to run such counts on a sample of the customer file. For example, a company may elect to pull a 10% sample of the customer database each month to generate the monthly active customer counts and any other counts requests (ad hoc or standard). The savings in computer processing time may be substantial, with differences ranging in the hours.

List Vitality Customer Statistics ____________________________ In addition to simply monitoring counts of key active customers, important statistics can be created for these groups, allowing you to monitor the vitality of your customer base. The key statistics include

♦ Percentage new-to-file ♦ Conversion rates to other titles, product lines, or services ♦ Retention or renewal rates ♦ Reactivation rates of nonactive customers

These statistics will allow you to more easily determine the success, at a corporate level, of various marketing strategies. This is done by breaking down the number of active customers (however they are defined) as shown in Exhibit 12.1.

View these counts weekly, monthly, or quarterly, depending on the promotional activity of your company.

Number of Active Customers as of Today

New-to-file Reactivated from

within the product line, title, or service

Converted from another product line, title, or

service

Retained as an active customer within the product

line, title, or service

Exhibit 12.1 Key Active Customer Statistics

Key List Segment Counts and Statistics _____________________ New-to-file, conversion, retention, and reactivation statistics can also be generated for each product line’s core customer segments. Suppose the book product line segmentation scheme for a direct marketer of books and videos is as shown in Exhibit 12.2.

Drozdenko12 2/26/02 6:17 PM Page 238

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

To better monitor the health of the book product line, the book product line manager views new-to-file, conversion, retention, and reactivation statistics by each major segment as follows:

♦ Number of all promotable names on the database ♦ Number of previous book buyers

♦ Number of customers with book activity within the past 24 months ♦ Percentage new-to-file ♦ Percentage retained ♦ Percentage reactivated ♦ Percentage converted

♦ Number of customers with book activity greater than 24 months ♦ Number of nonbook buyers

The video product line manager creates the same set of statistics but from his product line’s perspective/segmentation scheme.

Again, you should view these statistics weekly, monthly, or quarterly, depending on the promotional activity of your company.

Strategic Reporting and Analysis 239

Promotable Names

Previous Book Buyers Nonbook Buyers

Exhibit 12.2 Book Product Line Segmentation Scheme

_________________________________________ Calculating LTV Calculating LTV of a customer or a group of customers can be a very powerful tool for any direct marketing division. Doing so will allow a direct marketer to

♦ Assess the long-term results of various marketing programs, name sources, treatments, or strategies

♦ Effectively conduct what-if analyses to assist in determining the likely long-term benefits to be derived prior to test or rollout of a new program, source, or treatment

♦ Segment the customer file

To be successful, the information must be given in a meaningful way with appropriate directions about its use. When properly presented, this informa- tion will not only allow product managers to better manage their individual product line but also take into account the corporate impact of their decisions.

Drozdenko12 2/26/02 6:17 PM Page 239

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

240 OPTIMAL DATABASE MARKETING

The value of a group of names has been observed to reverse in compar- ison to other groups when the contribution of those names to the corporation was examined over a longer period of time. LTV calculations allow you to look at exactly this—the effectiveness of a campaign, treatment, or strategy not only with respect to the initial results but also its impact across all product lines over time, thus allowing better decision making.

Examples of how LTV calculations can assist you include

♦ Determination of which outside lists generate the most cost-effective customers over the long run by division or for the company as a whole

♦ Evaluation of a particular resuscitation or customer reactivation effort’s payout over time

♦ Understanding of various name acquisition programs (co-op mailings, card decks, banner ad placements, etc.) and how they compare in terms of the customers’ value over time

♦ In subscription rate base management, identification of the most cost- effective rate base from an overall company perspective via what-if analysis

♦ Assistance in prioritizing opportunities for strategic alliances with other companies through coendorsements and so on

♦ Evaluation of which product lines or products are more beneficial to the corporation as the front door to the company

LTV Methodologies

There are three basic ways to calculate the LTV of a customer or group of customers. Some are less complicated to implement than others. It depends on the ease of obtaining and then linking accurate and complete cost and revenue information associated with each promotion and purchase for the time period under consideration. For example, when we compare various outside list sources of names or categories of names for a given fiscal year, there are three possible ways to view the customer information to access their long-term value to the corporation:

1. Display key information regarding purchase and promotion activity (also called LTV profiles)

2. LTV calculations based on average cost and revenue figures and aggregate promotion and purchase data

3. LTV calculations based on actual cost and revenue figures and actual promotion and purchase data

The details of each method and examples follow.

Drozdenko12 2/26/02 6:17 PM Page 240

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

LTV Profiles

LTV profiles allow you to gain valuable information about long-term behavioral differences for various groups of names or the treatments thereof. LTV profiles give you, for example, a very clear picture of why one source of names may be more profitable than another. For a magazine pub- lisher, such reports can reveal which source of names become better donors, which group of names renew their subscriptions the next year at a higher rate, and which group of names are more likely to purchase cross-sell products. In this respect, LTV profiles are preferred over actual LTV calculations. They tell you which group is the most profitable in the long term and why.

For direct marketers with a complex business model, this may be the only or preferred method of assessing customer value over time. It may not be feasible to obtain accurate cost and revenue figures across the multiple promotions, products, and services.

Exhibit 12.3 is a sample LTV profile as previously described for a new source of names for a cataloger of home decorating products. The fields were chosen such that simple estimates of LTVs could be calculated by the product manager if desired. Additional information could be added to this report such as age and income information. What you display will depend on your needs.

By developing these reports across various sources, comparisons can be made to determine which are the best from a corporate point of view. Such reports will help highlight sources in which, although the initial order rate was quite high, there is a lower long-term value when compared to other sources.

Exhibit 12.4 is a 2-year LTV profile comparison report for each major source of new names for a b-to-b direct marketer of custom-imprinted products, business forms, and motivational prints. Product managers in

Strategic Reporting and Analysis 241

Exhibit 12.3 1-Year LTV Profile

1-Year LTV Profile as of 12/1/00

Source: New Outside List 1 Catalog Mail Date: 12/1/99 Promotion Plan Specifics: PLAN 12345

Corporate Statistics Value

Average total $ spent $67.34 Average number of paid products 4.28 Average number of returned products 0.29 Average number of catalogs sent 2.80 Average number of repeat purchases 1.76

Drozdenko12 2/26/02 6:17 PM Page 241

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

242 OPTIMAL DATABASE MARKETING

charge of the custom-imprinted products catalog need to assess the value of various sources of new names not just for their product line but corpor- atewide. This is done by developing an LTV profile that includes key statistics on the other product lines.

Again, quick calculations could be made for each major source to assist the product managers in determining where to focus their attention. Additional variables could also be displayed in this report depending on the needs.

To properly assess the 2-year value of new-to-file customers by source as displayed in Exhibit 12.4, you must not purge names from the database that became inactive during the evaluation period. Otherwise, the statistics will be inflated, because they will only be based on “actives.” If purging of inac- tive customers is a regular maintenance routine at your company, you can avoid the problems this will cause by the creation of a “perpetual” file that contains and tracks a large sample of all names residing on the database. This allows you to go back in time to conduct such strategic analyses.

Exhibit 12.5 shows the results of an expensive retention offer made by a cellular telecommunications company to their customers. In this example, the cellular telecommunications company gave free voice mail to their cur- rent customers for the life of their contract in hopes of increasing billable air-time usage and retention rates. The value of this promotion will be determined by comparing, after 1 year, phone usage for those customers given the free voice mail versus a holdout sample of customers not given free voice mail. An LTV profile was developed for assessing the value of this offer. Statistics were generated for key variables.

On examination of this report, we notice that the free voice mail has certainly increased air-time usage and revenue. But is it enough to offset the yearly cost of managing voice mail for each customer? To determine

Exhibit 12.4 2-Year LTV Profile Comparison Report

2-Year LTV Profile Comparison Report By Source Category

Profile of new-to-file customers coming on file via a custom imprinted products catalog offer between 1/1/98 and 6/1/98

Outside Business E-mail List Program Co-op Mailers Solicitations

Custom Imprinted Products: Percentage active 35 23 18 Total number of orders 4.3 3.2 1.6

Business Forms: Percentage converted 14 19 11 Total number of orders 1.4 1.3 0.9

Motivational Prints: Percentage converted 20 21 14 Total number of orders 2.5 2.6 2.0

Drozdenko12 2/26/02 6:17 PM Page 242

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

Strategic Reporting and Analysis 243

this, the product manager can perform some simple calculations by weigh- ing all costs against revenues.

The product manager may also want to display in Exhibit 12.5 such statistics as the conversion rate to other add-on options (e.g., call waiting, caller ID) or the percentage of customers ordering new phones for their spouse or children.

Actual and Aggregate LTV Calculations

You have two options when you calculate LTV values associated with customers or groups of customers: you can base the calculations on actual or on average cost and revenue figures.

Actual Cost and Revenue Figures. Although this method provides the most relevant analysis for precision decision making, it is often a tedious and difficult method for organizations not currently set up to directly link the revenues and costs to each promotion. In addition, the more complex the business model, the more difficult this type of calculation is.

Average Cost and Revenue Figures. This method is easier to apply than using actual cost and revenue figures because aggregate cost and revenue figures can be used. This is a preferred method by direct marketers when they have complex business models or when they cannot link revenues and costs to each promotion.

As previously mentioned, obtaining and linking actual cost and revenue figures to each promotion in order to calculate an actual LTV can be quite tedious and difficult. This is especially true for direct marketers who have a complex business model with multiple product lines and divisions. In these instances, calculating and applying average cost and revenue figures to actual or aggregate promotional and purchase data by product line or division will be much easier to implement. For example, when comparing

Exhibit 12.5 1-Year LTV Profile Comparison Report

One-Year LTV Profile Comparisons Report By Offer

Profile of customers receiving and not receiving the free voice mail option one year later

One-Year Statistics Received Free Voice Mail Did Not Receive Free Voice Mail

Total Air Time in Minutes 1,740 1,140 Percentage renewed contract

at end of 1-year period 89 82 Total dollars paid $474.72 $414.60 Total number of calls placed 468 312

Drozdenko12 2/26/02 6:17 PM Page 243

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

the 2-year value of one source of new names versus another source during fiscal 1999, direct marketers can use average cost and revenue figures by product line or division for the combined fiscal 1999–2000 periods in the LTV calculations.

The basic components of the LTV calculation, whether using actual or aggregate data, include

♦ Total revenues ♦ Total costs ♦ Adjustment for the net present value (NPV) of future profits

LTV is simply the sum of all revenues derived by the customer, less costs expended, for a set time period when adjusted for the NPV of future profits.

To determine how to adjust future profits for the present, you must first calculate the discount rate.

Calculating the Discount Rate and NPV

Future profits will not be worth as much in today’s money. For example, $1.00 in the future will buy less than $1.00 today. Because of this fact, when you determine the future value of customers, you must adjust the profit value calculated to reflect its true value in today’s money. To deter- mine the value of $1.00 next year as of today, you must apply what is called a discount rate. Not doing so can result in making incorrect decisions.

The formula for calculating the discount rate is

Discount Rate � [1 � (m � market lending rate)]y

Where y is an exponent that represents the number of years from the present m is a multiplier representing the risk associated with the business venture (most companies double the current market lending rate)

Once determined, you will calculate the discounted future profit, also known as NPV, as follows:

NPV � Profits/discount rate

LTV calculations are often represented in a table that provides a modi- fied income statement. Exhibit 12.6 shows the LTV calculations for a book club direct marketer who sent a special offer to a group of 10,000 currently active members drawn randomly from the database. The market lending rate was 10% and the risk multiplier was 2.

244 OPTIMAL DATABASE MARKETING

Drozdenko12 2/26/02 6:17 PM Page 244

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

The discount rate and NPV for Year 1 were determined as follows:

Discount rate in Year 1 � [1 � (2 � 0.10)]0

� (1.20)0

� 1.00(any figure raised to the zero power is 1)

NPV for Year 1 � $75,000/1.00 � $75,000

The discount rate and NPV for Year 2 were determined as follows:

Discount rate in Year 2 � [1 � (2 � 0.10)]1

� (1.20)1

� 1.20

NPV for Year 2 � $30,000/1.20 � $25,000

The discount rate and NPV for Year 3 were determined as follows:

Discount rate in Year 3 � [1 � (2 � 0.10)]2

� (1.20)2

� 1.44

NPV for Year 3 � $13,500/1.44 � $9,375

The discount rate and NPV for year 4 were similarly derived. Using the information provided in Exhibit 12.6, you can also derive

average LTV per customer. To determine the average LTV of a customer after 4 years, you divide the cumulative NPV by the number of customers at the beginning of the evaluation period:

Average 4-year LTV � $113,125/10,000 � $11.31

Strategic Reporting and Analysis 245

Exhibit 12.6 LTV

Year 1 Year 2 Year 3 Year 4

Customers 10,000 5,000 3,000 2,250 Retention rate (%) 50 60 75 80 Revenue $150,000 $60,000 $27,000 $13,500 Costs $75,000 $30,000 $13,500 $6,750 Profit $75,000 $30,000 $13,500 $6,750 Discount rate 1.0 1.2 1.44 1.77 NPV profit $75,000 $25,000 $9,375 $3.750 Cumulative NPV Profit/LTV $75,000 $100,000 $109,375 $113,125

Drozdenko12 2/26/02 6:17 PM Page 245

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

Consider a small cataloger of imported children’s toys from around the world. In 1998, they decide to track all new customers for 3 years. They do not purge any customers coming on file during this time period, regardless of inactivity. At the end of this 3-year evaluation period, they will calculate each customer’s average LTV. To do so, they will need to

♦ Calculate total dollars spent and the average number of catalogs sent per year for these customers; they will derive these fields from the database

♦ Estimate revenue as a percentage of sales ♦ Determine the average catalog costs per year

The LTV calculations are shown in Exhibit 12.7 with all necessary for- mulas displayed. We see that a new-to-file customer after 3 years is worth $42.46 on average.

In an effort to increase the retention rates, this cataloger is considering sending a free surprise toy to all customers with their first order and again every year afterward on their 12-month anniversary. On the basis of a similar test last year, the cataloger determined, via key coding, that retention increased by 3% and sales increased by 4%. Assuming this same increase in retention and sales holds true for future years, this cataloger will perform a what-if LTV analysis. Doing so will allow them to assess

246 OPTIMAL DATABASE MARKETING

Exhibit 12.7 LTV Report for Toy Cataloger

Year 1-1998 Year 2-1999 Year 3-2000

Customer Figures: Customers (A) 20,000 8,000 5,000 Retention Rate (B) 40% 62.50% 75%

Revenue Figures: Total $ Spent (C) $1,600,000 $720,000 $525,000 Average $ Spent (D � C�A) $80 $90 $105 Revenue as % of Sales (E) 40% 40% 40% Total Net Revenue (F � C � E) $640,000 $288,000 $210,000

Mailing Cost Figures: Average # Catalogs Sent (G) 4.6 4.7 4.5 Cost Per Catalog (H) $1.25 $1.29 $1.36 Total Mailing Costs (I � A � G � H) $115,000 $48,504 $30,600

Profit Figures: Net Profit ( J � F – I) $525,000 $239,496 $179,400 Discount Rate (K) 1 1.2 1.44 NPV Profit (L � J�K) $525,000 $199,580 $124,583 Cumulative NPV Profit (M) $525,000 $724,580 $849,163 Average Customer LTV (N � M�20,000) $26.25 $36.23 $42.46

Drozdenko12 2/26/02 6:17 PM Page 246

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

Strategic Reporting and Analysis 247

the long-term value of implementing such a program, ensuring that they make the correct decision. The LTV table for this what-if analysis is shown in Exhibit 12.8.

By comparing the 1-year average LTV of names receiving the special treatment to the 1-year average LTV from Exhibit 12.7, it is obvious that this new strategy does not pay out. However, when it is assessed over time, we notice that breakeven occurs at Year 3. As a result, the cataloger has a decision to make: Is a 3-year breakeven a viable option?

Sample Types Used in LTV Calculations

Numerical calculations in LTV reports will differ, depending on the type of customer data/sample you are examining. There are two types of customer samples typically analyzed:

1. New-to-file names

2. Current customers (e.g., when analyzing the effect of a resuscita- tion effort on a group of current customers)

Exhibit 12.8 What-If LTV Report for Toy Cataloger

Year 1 Year 2 Year 3

Customer Figures: Customers (A) 20,000 8,240 5,307 Retention Rate (B) 41.20% 64.40% 77.30%

Revenue Figures: Total $ Spent (C) $1,664,000 $771,264 $579,524 Average $ Spent (D � C�A) $83.20 $93.60 $109.20 Revenue as % of Sales (E) 40% 40% 40% Total Net Revenue (F � C � E) $665,600 $308,506 $231,810

Mailing Cost Figures: Average # Catalogues Sent (G) 4.6 4.7 4.5 Cost Per Catalogue (H) $1.25 $1.29 $1.36 Total Mailing Costs (I � A � G � H) $115,000 $49,959 $32,479

Gift Cost Figures: Cost of Free Toy ( J) $1.75 $1.75 $1.75 Total Gift Cost (K � J � A) $35,000 $14,420 $9,287

Profit Figures: Net Profit (L � F � I � K) $515,600 $244,126 $190,044 Discount Rate (M) 1 1.2 1.44 NPV Profit (N � L �M) $515,600 $203,439 $131,975 Cumulative NPV Profit (O) $515,600 $719,039 $851,013 Average Customer LTV (P � O/20,000) $25.78 $35.95 $42.55

Drozdenko12 2/26/02 6:17 PM Page 247

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

248 OPTIMAL DATABASE MARKETING

For example, to calculate total motivational print orders (see Exhibit 12.4) after a 2-year period for new-to-file customers, the reports will simply pick up the value of that summary field on the marketing database. However, this is not the case for LTV profiles involving current customers. In the case of our telecommunication example (Exhibit 12.5) and determining “Total Air Time in Minutes” after 1 year, the report will subtract the value of the summary field when the customer was first given the free voice mail option from the same data field 1 year later. For this example, a frozen file is required, reflecting what the customer looked like at the time of the test promotion. (Chapter 6 contains additional information on frozen files.)

Forecasting LTV

In addition to calculating the historical LTV of customers or groups of customers, you can also forecast LTV. This is done via regression modeling. For example, you can build a model to predict the value of customers after 2 years on file, based on the customers’ actions during their first few months on file. The model can additionally incorporate any demographic, psychographic, and census-level data that might be available.

With these models, a direct marketer can rank customers shortly after they come on file from those most likely to become good customers to those less likely to become good customers. On the basis of this knowledge, the direct marketer is in a strong position to implement appropriate strategies to deal with likely strong and weak customers.

Impact Studies __________________________________________ If you are a direct marketer considering the addition of a new multiple prod- uct line, you must be prepared to properly gauge the impact of this addition on your current product lines. You do this by establishing an impact study, which allows you to compare the value of the new product line with the costs of the potential cannibalization of customers from your current prod- uct lines. In all likelihood, the new product line will have a negative impact on the rest of the business. The question is, How much? It depends on your customers and how much competition there will be between the new prod- uct line and your current product lines. For example, direct marketers cur- rently selling one-shot do-it-yourself books will in all likelihood see a large impact to this business unit if they decide to launch a new do-it-yourself book club. You can gauge how much by establishing an impact study dur- ing the test phase. However, if these same direct marketers are looking to launch a new catalog of collectable plates and they do not currently sell plates or collectable items of any type, the impact may be minimal.

Drozdenko12 2/26/02 6:17 PM Page 248

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

Strategic Reporting and Analysis 249

The steps involved in establishing an impact study to determine the effect on the one-shot business of starting a do-it-yourself book club are as follows:

1. Select a large sample of names

2. Split the sample 50/50—half of which will be key coded to receive the new do-it-yourself book club offer and half will be key coded not to receive the new do-it-yourself book club offer

3. Promote only those names key coded as such with the new do- it-yourself book club offer

4. At a later time, promote both groups with a one-shot do-it-yourself book offer

After a certain amount of time has elapsed, you can determine the impact of the new do-it-yourself book club on the one-shot do-it-yourself book business by examining the one-shot response rates for each group. Is the one-shot response rate lower for those names promoted with the club offer, or is it the same? To assess the long-term implication of rolling out with such a club, you can even prepare a 1- or 2-year LTV profile or calculation.

_________________________ Monitoring Promotional Intensity

If direct marketers begin to experience significant reductions in response rates, they may want to monitor promotional intensity. In other words, the direct marketers may be mailing too many promotions, which results in list fatigue. To establish a test of this hypothesis, execute the following steps:

1. Select a random sample of active customers for the product line or division of concern

2. Split the sample into two equal samples

3. Key code one group to receive promotions as usual

4. Key code the other group to receive fewer promotions

You need to decide on the definition of “fewer promotions.” You may want to test a few variations of fewer promotions in an attempt to gauge intensity more accurately.

After a set period of time, for example 1 year, examine responses to pro- motions for each group and determine if those receiving fewer promotions eventually yield higher response rates to promotions they did receive. Did the rebound in response rates cover the lost revenue from fewer promo- tions? Did you receive fewer customer complaints for this group? On the basis of the answers to these questions, a marketing plan may be built to combat these issues.

Drozdenko12 2/26/02 6:17 PM Page 249

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .

250 OPTIMAL DATABASE MARKETING

Many types of reports are key to the establishment of proper marketing strategy. In this chapter, we discuss key active customer counts, which monitor the size of important universes of customers, and list vitality customer statistics, which allow a marketer to monitor the vitality of the customer base. Calculating customer LTV is also a very important market- ing tool but also one of the most complex to implement. Several LTV methodologies are discussed, some of which are easier to implement than others. In addition, we discuss ways to assess the impact of new product lines on existing product lines and how to monitor promotional intensity.

Chapter Summary _______________________________________

Review Questions _______________________________________ 1. What measures are used to monitor the vitality of the customer

base?

2. What is the advantage of producing LTV profiles over actual LTV calculations?

3. Why would a direct marketer calculate LTV based on average cost and revenue figures as opposed to actual cost and revenue figures?

4. What is NPV and why is it important when calculating LTV?

5. Discuss the steps involved in establishing an impact study.

6. Discuss the steps involved in successfully monitoring promotional intensity.

Through proper strategic reporting and analysis, the health of any direct marketing firm can be monitored and proactively managed.

Drozdenko12 2/26/02 6:17 PM Page 250

C o p yr

ig h t ©

2 0 0 2 . S

A G

E P

u b lic

a tio

n s.

A ll

ri g h ts

r e se

rv e d .