AIDA is an acronym that serves as a guide to creating copy for a direct mail campaign. The letters stand for Get Attention, Arouse Interest, Stimulate Desire, and Ask for Action.

Appended Data can be added to any name and address file to learn age, income, home value, home ownership, presence of children, length of residence, and dozens of other valuable pieces of information about any household. This information can be used to create customer segments, and guide strategy designed to create more effective customer communications. Similar information can be appended to business to business files: SIC code, number of employees, and annual sales.

Average Customer Value (ACV) is a Key Performance Indicator that tracks differences in customer value between segments. It is sometimes also referred to as the Per Capita Spending Rate. It is the Gross Sales over a given period divided by the count of those marketed to during the same period. ACV for acquired customers is normally lower than for retained ones. Therefore, if ACV declines year to year, it is an indication that your company is pursuing Acquisition at the expense of Retention. Increasing ACV, even marginally, often has a significant, positive effect on Profit, and can be accomplished by increasing Frequency and/or AOV.

Average Order Value (AOV) is a customer's Monetary Value divided by their Frequency. It is also their Gross Sales for a given period divided by the number of orders over the same period. Increasing AOV, even marginally, often has a significant, positive effect on Profit, and can be accomplished by emphasizing cross-selling and up-selling.

Attrition is a Key Performance Indicator that measures the value of lost business. It tracks the sales activity of lost customers who bought last year but not this year. The goal is lose less customer value than what is either retained or acquired.

When Attrition is more than either Retention or Acquisition, it is a sign of trouble that can be corrected either by reducing Attrition or increasing Acquisition. Note that reducing Attrition is estimated to be ten times less costly than increasing Acquisition. One of the best ways to reduce Customer Attrition is to monitor their Expectancy and to offer them incentives to buy again when then enter into negative Expectancy territory. Monitoring the Expectancy of very high value Customers and interacting with them before then enter negative Expectancy can be especially effective in reducing their Attrition.

The Attrition Rate is the sales activity of this year's lost customers, i.e those who bought last year but not this year, divided by the sales activity of all of last year's customers.

A Bounce Back is an offer that is shipped with a package sent to fulfil a customer order. Every fulfilled order should include a bounce back, since doing so increases customer Frequency and Lifetime Value.

Breakeven Analysis measures the minimum number of sales a marketing campaign must generate in order to cover the costs of the campaign. Breakeven varies considerably by customer segment. Breakeven Analysis can be done retroactively, though it is most often done as a forecast. One of the major purposes of Customer Segmentation is to discover which segments did not, or will not, breakeven. Marketing to segments that fail to breakeven should be limited or even eliminated all together.

Breakeven Point is the point in a segmented list below which you will not cover or have not covered the costs of a marketing campaign. Breakeven varies considerably by customer segment. One of the major purposes of Customer Segmentation is to discover which segments did not, or will not, breakeven. Marketing to segments that fail to breakeven should be limited or even eliminated all together.

A Compiled List is mailing list that is used for prospecting or lead generation. It has been compiled from a variety of different sources including newspapers, directories, and public records that identify groups of people with common traits. Compiled Lists are less expensive and usually less responsive than Response Lists.

The Current Year is not the current calendar year. Rather, it is the most recent twelve month period from the Upload Date. Throughout this site, the Current Year is sometimes referred to as "this year".

Customers are persons or organizations who have placed an order during a specific period. Throughout this site, repeat customers are defined as the those customers who have bought both last year and this year. New customers are those who have bought this year but not last year. And lost customers are those who have bought last year but not this year.

Customer Segmentation is grouping customers according to their likely behavior and profit potential. Accurate, verifiable Customer Segmentation gives marketers the information they need to evaluate and execute strategies for improving profitability and marketing effectiveness. Success comes from creating useful customer segments, and developing specific marketing strategies for each. Life Stages and N-tiles are two examples of Customer Segmentation.

Database Marketing is the process of interpreting database information in order to improve marketing efficiency or to gain additional information on customers or prospective customers. Marketing based on Customer Segmentation is a common form of Database Marketing.

Demographics are characteristics defining a particular group of people - for example, age, sex, income, education, type of residence, type of business, job title, location, etc. Though Demographics have some predictive quality, they are not nearly as predictive as Transactional Data, such as Recency, Frequency, and Monetary Value. Therefore, Demographics should be used in conjunction with Transactional Data, or when Transactional Data is unavailable, such as with Prospects.

Direct Marketing is the sending of a promotional message directly to consumers, rather than via a mass medium. It is a planned system of contacts seeking to produce a Lead or an Order. Direct marketing requires the use of a database and can be measured in costs and results. It includes methods such as Direct Mail and Telemarketing.

Expectancy is the number of days until the Customer can be expected to order again. Expectancy is calculated by subtracting the Customer's Recency from the average Latency of all other customers who re-ordered while they were at that same order level. For instance, if a customer has placed two orders, his Expectancy would be determined by the Latency of all customers who had placed a third order.

Positive Expectancy means that their Expect Date is in the future. Negative Expectancy means that their Expect Date is in the past. By definition, customers in Life Stage One and Two have positive Expectancy, while those in Life Stage Three and Four have negative Expectancy. Expectancy is a secondary predictor of customer behavior.

Expect Date is the date that a customer can be expected to order again. The Expect Date is calculated by adding their Latency to their Recent Date. If a customer's Expect Date is in the future, the likelihood of his re-ordering is increasing. If it is in the past, the likelihood is decreasing.

Frequency is the number of times a customer has placed an order during a specific period. It is the second most important predictor of future customer buying behavior following Recency. Frequency can vary considerably among customer segments. Segments of customers with high Frequency will buy more in the future than those with low Frequency. By definition, customers in Life Stage One and Four have a Frequency of one (i.e. first-time buyers), while those in Life Stage Two and Three have a Frequency of two or more (i.e. multi-buyers). Frequency is a primary predictor of customer behavior.

Geographics is any method of subdividing a list with geographic subdivisions. Common subdivisions include country, state, metro area, county, city, 3-digit ZIP, 5-digit ZIP, carier route, or census tract.

Gross Sales is the amount of money spent during a specific period, either actual or projected. Gross Sales usually does not include sales tax.

Growth is a Key Performance Indicator that measures the value of increased business. It tracks changes in sales activity from the Previous to the Current Year. The goal is to show positive growth in Customers, Orders, and Gross Sales as well as in Average Order Value and Average Customer Value. Negative growth in one or more areas is a sign of possible business weakness.

The Growth Rate is the increase in sales activity from last year to this year divided by all of last year's sales activity.

A House List is a mailing list made up by a company based on current or former customers or inquires for the company's product or service. A company's House List is its most valuable asset. In many ways, the value of the House List is the value of the company.

An Inquiry is someone who has asked for literature or other information about a product or service. Unless otherwise stated, it is assumed no payment is required for the information. A catalog request is a specific type of Inquiry. An Inquiry is also referred to as Lead.

1. "Did this year's customers produce more value than last year's?"
2. "How do this year's KPIs compare to last year's?"
3. "Did repeat customers produce more value this year than they did last year, and did new customers produce more value than the lost customers they replaced?"

Latency is the time between orders. It is calculated as the average number of days that customers took to place their next order. It is calculated separately at each order level. As a customer's Frequency increases, their Latency decreases. For instance, the Latency between second and third orders is typically shorter than the Latency between first and second orders. Latency is a secondary predictor of customer behavior.

Using these data points, Leads can be analyzed and ranked by the Recency, Frequency, and Monetary Value of their inquiries, which is highly predictive. Note that if no Inquiry Amounts have been captured, Leads can be assigned an Inquiry Amount equal to the AOV. Though the Response Rate of Leads will be lower than that of existing customers, it will usually be significantly higher than that of Prospects. Therefore, lead generation can often be a more effective means of customer acquisition than prospecting.

Life Stages are the distinct phases that all Customers, to a greater or lesser degree, move through during their lifecycle. As Customers migrate through these Life Stages, they are either gaining potential value or losing it. Segmenting Customers by Life Stage is essential for developing an effective Marketing Strategy to reach them.

Customers are segmented into one of four Life Stages.

List Hygiene is the ongoing process of keeping a mailing list clean and up-to-date by adding, editing, and deleting data. For direct marketing companies, the most important aspect of list hygiene is address correction. It is estimated that 20% of most mailings are undeliverable as addressed, and another 5% are duplications. Poor list hygiene is a significant drain on profitability, since it both increases costs and decreases sales revenue.

Lifetime Value (LTV) is a measurement of a customer's long-term profitability. It is one of the most important calculations your company can make. It is the total profit that your company will receive from transactions with a given customer during the time that you continue to market to them. There are many ways to calculate Lifetime Value, but most are so complicated as to be unusable.

As an alternative, we calculate the estimated Lifetime Value to date. The formula we use is…

Knowing the LTV of customers allows companies to decide how much they should spend on their Acquisition and Retention programs. LTV can also be used to identify the most valuable prospect lists, lead generation programs, promotional offers, product offerings, and markets. LTV is closely associated with Retention. A small boost in Retention can result in a large increase in LTV.

On this web site, the LTV is limited by the Analysis Period, which is determined by the dates of the sales transactions included in the Upload File. The Analysis Period has a direct bearing on LTV, in that it limits the "lifetime" to the period covered by the Upload File.

Loyalty Programs are part of the mix of Retention building services that Database Marketing has made possible. Most customers are delighted to participate in well designed loyalty programs. Airlines have been outstandingly successful in these programs. Their use has spread to supermarkets, hotels, retail stores, and a variety of industries including Direct Marketing.

Note that Retention is inherently more profitable than Acquisition, since it is ten times less costly per customer. Retention also reduces Attrition, because repeat customers attrite less often than new customers. Reducing Attrition also makes Acquisition automatically easier to accomplish, since fewer customers need to be replaced. Nevertheless, companies seldom achieve their growth goals through Retention alone. Therefore, the best Marketing Strategies find a balance between Retention and Acquisition.

Monetary Value is the amount of money a customer has spent during the analysis period. It is the third most important predictor of future customer buying behavior following Recency and Frequency. Monetary Value can vary considerably among customer segments. Segments of customers with high Monetary Value will buy more in the future than segments of customers with low Monetary Value. Monetary Value is a primary predictor of customer behavior.

Net Acquired Value (NAV) is a Key Performance Indicator that tracks the change in value of new customers compared to the lost customers they replaced. In other words, it is the difference between Acquisition and Attrition. It is difficult for a company to grow unless their Net Acquired Value is positive. The goal is to acquire new customers that produce more value than the lost customers they replaced.

Net Retained Value (NRV) is a Key Performance Indicator that tracks the change in value of repeat customers from last year to this year. In other words, it is the difference between Retention in the Current Year (Retention CY) and Retention in the Previous Year (Retention PY). It is difficult for a company to grow unless their Net Retained Value is positive. The goal is to retain repeat customers that produce more value this year than they did last year.

N-tiles are groupings of equal sized segments of customers that have been ranked and grouped into 100 Centiles, 20 Vigintiles, 10 Deciles, 5 Quintiles, 4 Quartiles, or 2 Semitiles. Grouping by N-tiles makes it easier to compare customer segments. When properly done, a disproportionate share of the Gross Sales, LTV, and Profit will come from customers in the top N-tiles, leaving the money-losing customers at the bottom.

Orders is the count of Orders placed during a specific period, either actual or projected.

The Previous Year is not the previous calendar year. Rather, it is the twelve month period immediately preceding the Current Year. Throughout this site, the Previous Year is sometimes referred to as "last year".

Predictive Models append demographic data to transactional data to create models that predict which customers are most likely to defect, and which customers are most likely to respond to new initiatives. Modeling can be quite expensive, but it is very powerful technique that can increase Response Rates and reduce Attrition. Some high-end Predictive Models include regression analysis, neural networks, genetic algorithms, Chi-Squared Automated Interaction Detectors (CHAID), and Classification and Regression Trees (CART).

Predictivity is the power to predict. Recency, Frequency, Monetary Value, and Lifetime Value have high Predictivity because they have the power to predict future customer buying behavior. These are referred to as primary predictors. Other predictors, such as Latency, Expectancy, Life Stages, and KPIs, are less predictive, but are essential for developing an effective marketing strategy. They are referred to as secondary predictors.

Product Category is an important consideration when segmenting customers. Usually, customers should be analyzed, ranked, and segmented regardless of the category of products they have purchased. But if a company has two or more distinct product categories, customers should also be analyzed, ranked, and segmented by product category. This is especially true for companies who have customers that fall into two main groups, those who've purchased high ticket items and those who haven't. In such cases, each group should be analyzed, ranked, and segmented separately as well as together.

Profit is the residual value gained from operating your business. Specifically, it is your sales revenues minus the expenses associated with producing those revenues. The formula we use for calculating operating profit from a direct marketing business is…

Profit = Gross Sales - (Fulfillment + Marketing + Returns + COGS + Sales & Service + Overhead).

A unique feature of the FreeRFM.com™ website is that it allows you to calculate Profit using your own values for each of the cost variables.

Profit Predictors have the ability to predict future customer buying behavior, which if properly applied will make your company more profitable. Recency, Frequency, and Monetary Value, especially when factored together with Tenure, are primary predictors, meaning that the are highly predictive of customer behavior. Latency, Expectancy, Life Stages, and KPIs are secondary predictors, meaning that they are less predictive, but nevertheless important for developing effective marketing strategies.

Prospects are persons or organizations who have neither made an inquiry or purchase, but do share certain demographic characteristics with those who have. Assuming that the Prospect has a need for your product, the most predictive demographic characteristic will usually be their income. Prospects who can afford to buy your product are much more likely to make a purchase than those who can't.

Whenever possible, the selection of a Prospect list should be modeled using the LTV of existing customers. Prospects demographically similar to customers with positive LTV should be included, but those similar to customers with negative LTV should be excluded. Since the Response Rate of Prospects will usually be significantly lower than that of Leads, prospecting is often a less effective means of customer acquisition than lead generation. In other words, it is sometimes better to prospect for Leads rather than for Customers.

Psychographics are characteristics or qualities used to denote the lifestyles or attitudes of prospects and customers. Though Psychographics have some predictive quality, they are not nearly as predictive as transactional data, such as Recency, Frequency, and Monetary Value.

Rank is the numerical order of customers according to a sorting sequence or mathematical formula.

Recency is the number of days since a customer's last order was placed. It is the most important predictor of future customer buying behavior. Recency can vary considerably among customer segments. Segments of customers with high Recency will buy more in the future than segments of customers with low Recency. Recency is a primary predictor of customer behavior.

A Response List is mailing list that is used for prospecting or lead generation. It is made up of persons who have responded to another company's ad, catalog, direct mail package, TV ad, or other offer. In the past, companies kept their customer lists strictly private. Today, most lists are shared, exchanged, or rented. As a result, there are tens of thousands of Response Lists on the market. Response Lists are more expensive and usually more responsive than Compiled Lists.

The Response Rate is the level of replies or orders to a particular marketing offer given as a percentage. It is calculated as the count of orders received during a given period divided by the count of customers who received the offer. Response rates vary dramatically among customer segments. The average Response Rate for a marketing campaign can be misleading, since the top segments will have Response Rates much higher than the average, while the bottom segments will have rates much lower. As a rule of thumb, Response Rates for Customers will be five times that of Leads, and Leads will be two times that of Prospects.

Retention is a Key Performance Indicator that measures the value of repeat business. It tracks the sales activity of repeat customers who bought both last year and this year. Since they are the same customers in both years, Retention must be calculated separately for each year. Retention PY is their sales activity for the Previous Year, and Retention CY is their sales activity for the Current Year. The goal is to retain customers that produce more value this year than they did last year.

The Retention Rate is the sales activity of either RetentionPY or RetentionCY divided by all of last year's sales activity.

RFM is an acronym that stands for the three primary factors of predictive customer analysis: Recency, Frequency, and Monetary Value. However, it is an inexact term that can mean the factors themselves, the analysis of the factors, or even the result of the analysis. Recency, Frequency, and Monetary Value are the foundation of all predictive customer analysis, including high-end statistics-based predictive models, such as regression analysis, neural networks, genetic algorithms, Chi-Squared Automated Interaction Detectors (CHAID), or Classification and Regression Trees (CART). Because of the expense associated with these high-end models, RFM Analysis is usually a more cost effective option for small- and medium-sized companies.

RFM Analysis is a marketing technique that uses the Recency, Frequency, and Monetary Value of customers to predict whether or not they are likely to buy again. Customers who have purchased more recently, ordered more frequently, and spent more monetarily are much more likely to buy again when compared to customers who have purchased less recently, ordered less frequently, and spent less monetarily. RFM Analysis can be done in a variety of ways, some are sorting sequences and others are mathematical formulas. Mathematical formulas, when properly used, are usually more predictive.

R,F,M (read as RFM Sorted) is a sorting sequence for ranking customers using Recency, Frequency, and Monetary Value. Some experts recommend sorting a customer database 31 times. Sort it first by Recency, and divide it into five equal groups numbered 1 to 5. Then sort each of the five Recency groups separately by Frequency, and divide each of those groups into five equal groups numbered 1 to 5. Finally, sort each of the 25 Recency/Frequency groups by Monetary Value, and divide each of those groups into five equal groups numbered 1 to 5. The result will be 125 groups of equal size numbered from 111 to 555. Because it is a sorting sequence instead of a algorithm, R,F,M can be difficult to do correctly, and it is generally less predictive than factoring RFM as in our algorithm, R·F·M.

R·F·M (read as RFM Factored) is an algorithm for ranking customers using Recency, Frequency, Monetary Value, and Tenure. The algorithm we use for R·F·M is…

R·F·M = ((Frequency / Tenure) x (Monetary Value / Frequency)) / abs(Latency - Recency)

Note that Frequency divided by Tenure equates to Orders per Year. Monetary Value divided by Frequency equates to Average Order Value. And Latency minus Recency equates to Expectancy. Furthermore, ((Frequency / Tenure) x (Monetary Value / Frequency)) is mathematically the same as (Monetary Value / Tenure) which equates to Sales Revenue per Year. Therefore, the algorithm is simply calculating the customer's Sales Revenue per Year divided by the time until they can be expected to order again. The actual algorithm rounds both Tenure and Expectancy to the nearest six week period. This equalizes the time factor in them and provides a workable forecast of sales activity for the upcoming six week period.

Tenure is the number of years since a customer's first order was placed rounded to two decimal places. Tenure is often overlooked, since by itself, it is not a predictor of customer behavior. However, Tenure significantly enhances the Predictivity of Recency, Frequency, and Monetary Value when factored with them. Tenure divided by Recency is a strong indicator of a customer's loyalty. Tenure divided into Frequency is a strong indicator of a customer's value, and equates to Orders per Year. Tenure is also an indispensable factor in calculating Lifetime Value. And finally, Tenure is key ingredient in our customer ranking algorithm, R•F•M.

Transactional Data can be used to predict future customer buying behavior, and it is much more predictive than demographic data. The most basic form of transactional data is a sales transaction that includes the Order ID, the Order Date, and the Order Amount, along with the Customer ID and other identifiers, such as name, address, and phone number. Other kinds of transactional data, such as sales inquiries and product categories, are also predictive.

The Upload Date is the date that the file of transactions are uploaded to the FreeRFM.com™ website. It is assumed that the upload file includes the most recent sales transactions available, even including those transacted on the Upload Date. Only use analysis of transactional data that has been recently uploaded, since the reliability of the analysis decays quickly.