Use E-mail Address Appending to dramatically increase your web site traffic and internet communication with your customers or clients:

• Improve your conversion rate through multi-channel marketing
• Create faster and more effective marketing campaigns
• Build added revenues from your current database

Your "non-E-mail" customer file is matched against various databases and when matches are found, those E-mail addresses are added to the customer file. This gives you a file with both E-mail and non-E-mail addresses.

After an e-mail match is found for a corresponding physical address, a letter is e-mailed with the option for the addressee to opt out. After allowing 5-7 business days for all responses, final numbers will be calculated by excluding undeliverable and opt-out e-mails.

When data collection is done on the Internet, subscribers voluntarily disclose personal information and have access to view a "privacy policy." The typical policy states that the personal information they provide on the web may be shared with third parties that already have and existing "customer relationship" with the individual. For example, Company A has a standard “mail” relationship with Customer 1. Company A provides its mailing list to MediaSpade for E-mail Appending. MediaSpade finds Customer 1’s e-mail address in an on-line database. Thus, Customer 1’s e-mail address can be appended to Customer 1’s mailing address.

At the completion of the E-mail Appending Process, the enhanced file is returned you. The only E-mail addresses that are appended will be those that matched the mailing information and that were successfully delivered.

The append percentage will vary depending on the accuracy of the data in your original mailing list. Consumer databases frequently achieve match rates from 15-25%. To enhance your append rate, addresses in your list should be verified through NCOA, DSF and CASS prior to processing.

Market Penetration Report

Market Penetration Reports - This automated report lists the overall number and percent of the client's records that match over 200 data elements. It provides a general "snap shot" view of a your client file. Many marketers have very little knowledge about the composition of their customer file and this report identifies both the number of records on the marketer's file that match each of the data elements and the percentage of their file that number represents. For example, for the Ethnic Markets data element, the Match Rate Summary Report lists the total number and percent of the client's records that match an ethnic category, without breaking numbers and percentages out further. For more specific counts and percentages for each ethnic category, the marketer can refer to the Match Rate Detail Report.

SuperModels: Customer Predictive Analysis (CPA) (modeling your current database for acquisition of new records)

SuperModels are designed to dramatically increase the effectiveness of targeted list selections by improving your ability to identify high-potential customers. Our proprietary modeling process combines advanced data standardization techniques with powerful predictive modeling methods, including logistic regression and Z-Score.

What makes SuperModels successful? Simply put, our models deliver exceptional performance because we take the time to integrate statistical analysis with detailed information about your business objectives. In addition, SuperModels are built using an innovative, multimodeling approach that allows us to reveal the types of hidden segments single-model techniques typically fail to identify.

The result: the SuperModel we develop for you will increase your response rates and provide you with a significant quantity of quality prospects. Guaranteed.

The Customer Predictive Analysis (CPA) is a statistical model that compares your customer file to a prospect universe within an database. With a CPA, we find new prospects that mirror your best customers by identifying the most significant characteristics of the customers in your file. In fact, we rank names in the Equifax database from most likely to least likely to buy based on how closely the individuals resemble your best customers.

A Customer Predictive Analysis can:

• Improve your response rates and expand mail volume
• Help you successfully reach hidden markets
• Give you the ability to better target your marketing messages
• Reduce mailing costs by allowing you to prioritize your best customers

We use a regression model to build your report. The regression model applies a multivariate formula to all consumer characteristics, and reveals how these characteristics interact with one another.

Building Your CPA Model

To build your CPA scoring model:

1. We begin by matching your customer file to the selected database to find individuals who are common to both files using name and address. When a match is found, we temporarily append the customer information from our database to the customer record in your file.
   
2. Next we select a random sample of non-customers from the database to compare with your customers. Through analysis and with your input, we segment the selected database by applying preselect criteria to target your prospect universe even more closely (this is optional)*. If the sample size is large enough, we randomly split your customer file into a Model-Building Sample file and a Validation Sample file. Later in the process, we apply the model we built to both files to
   verify that we receive similar results.
   
3. Then we use a chi-squared test of significance to isolate the most significant characteristics that differentiate your customers from noncustomers in our database. Using stepwise linear regression, we further refine the list of significant characteristics and remove highly correlated characteristics.
   
4. Using logistic regression, we assign weights to each of the most significant characteristics. Then we calculate an overall score for each non-customer from the database by adding up all the regression weights for every significant characteristic for each individual.
   
   * If pre-select criteria are used, from this point on, a “selected database” means the database after the criteria have been applied.
   
5. We then sort and divide the non-customers into ranks based on their overall scores. Your customers are then scored and ranked using the score cut-offs for each rank. This score measures how similar names (the non-customers) are to your customers.
   
6. Lastly, we report the relative importance of each significant characteristic and whether it had a positive or negative effect in the model. We further compare the percentage of the database individuals with each characteristic to the percentage of your customers with each characteristic, and the distribution of characteristics by rank.

Please let us know if there is an opportunity to review this information together so that we may be able to answer any questions that you may have.

We look forward to earning you business.

For more information please contact us at 678-999-8511 or sales@mediaspade.com