Five tips for getting the most out of your merge/purge

1.Segment as much as you can while maintaining measurable cell sizes.
Short of applying statistics for valid and reliable segment sizes, you can apply the Rule of data_cleaning_img02100s to setting cell sizes. This rule essentially suggests that to get a meaningful and repeatable result from a segment, you would like to be able to get 100 orders for that segment. This means that if you expect a 2 percent response from a segment, you would need to mail at least 5,000 names in that segment.

In the merge/purge, this is important because mailers often find themselves in need of splitting or nth-ing segments at the end of a merge to meet mail quantity specifications. If you find yourself in this position, and you have applied as extensive a segmentation as possible, allow yourself greater flexibility by selecting inclusion records on the back end. Instead of having to nth from a set of 30,000 36-month buyers, leave yourself the option of selecting for a subset of 12,000 36-month multibuyers.

2.Use different merge rules for business and consumer segments. Merge rules are a set of instructions that dictate whether duplicates are to be matched at the individual level (e.g., same individual at the same address), at the household level (e.g., same surname at the same address) or at the address level (e.g., same address regardless of name). Each set of rules defines what is or is not considered a duplicate.

Most service providers will allow you to assign different rule sets to consumer records and business records, as well as to enhance those rules with special instructions. An example would be to allow the merge to mail multiple records to a business site but only one per family household within a single processing run. By assigning different rules, you cut down on processing costs and allow the right mail strategy to align with the right segments in your file.

3. Bring in more names than you need.
This tip applies to both housefile segments and prospect lists. As mentioned above, there often are mail quantity specifications that are set prior to going into the merge/purge. Printer contracts have been signed that guarantee a specific mail volume. In addition, you may be testing new lists or new segments within proven lists in search of new opportunities to grow your file. List order minimums often are 5,000 to 7,500 names and frequently, these are the test order quantities that mailers opt for. However, you should consider the impact on quantities in the merge/purge.

Just how clean is your data? Identify where your data requires attention, allowing you to choose which areas to improve.

Get Free Email Append Test from AverickMedia

Every mailer that is a member of the Direct Marketing Association (DMA) should run all outside lists against the DMA’s National Do Not Mail file. When you do that, you may be surprised how many names from your prospect lists will match that file and, as a result, not be included in the mailing. Also, consider the Rule of 100s mentioned earlier, and couple that with the fact that many mailers realize response rates from prospect lists in the range of 0.8 percent to 1.5 percent. Given these points, if you want to get 100 responses from a prospect list at a 0.8 percent conservative response rate, you need to mail 12,500 names. Now consider the effects of the merge, where you may lose 10 percent of a list to duplicates and Do Not Mail matches. You would need to bring about 13,850 names into the merge/purge to get the 12,500 names you want to mail. Said another way, if you order 5,000 names (the typical order minimum) you may only get 4,000 in the mail which, on a good day, might produce 48 orders—not much to learn from.

download (1)4. Go random in the merge with prospect lists.
There are varying opinions about how to treat prospect list hierarchy in the merge/purge. Some mailers take the position that all exchange lists should be ranked higher than cooperative database lists, which would rank above any pure rental files. This approach is better for managing costs, particularly when net-name arrangements are put in place for rented lists.

Another approach, however, is to allow all prospect lists to rank the same in the merge/purge, a concept known as going “random in the merge.” Allowing all outside lists to go random in the merge gives each list the opportunity to “keep” the better multilist names. In this way, the post analysis can provide an apples-to-apples comparison of list performance. Unless you are a large mailer, you’re probably not yet in the position of finding purely incremental names. For your business, finding new list opportunities still is critical to growth, and allowing lists to go random in the merge gives the greatest opportunity for finding new lists that work for your offer.

5. Make multis a part of your mail plan.
Multis, or multilist prospects or multibuyers, are names that show up on more than one outside list. If, for example, John Doe’s records showed up on four outside lists that you rented for a given mailing and he wasn’t already on your customer file, he would be referred to as a “four-time multi.” These multibuyers generally are strong-responding, highly-qualified prospects—in other words, the best possible prospects you can get.

Many mailers deal with these records as an afterthought, if they deal with them at all. But to get the most of your prospect lists, make multis a part of the mail plan by evaluating past merge/purge results and estimating how many multis you will have available after processing. Include these names in earlier and later drops than your main prospecting drop. The key is that when you plan to include them, they can make a significant impact on sales and new customer acquisition, and may even outperform older house segments.

By Steve Trollinger

Article From: http://





Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s