What is the best course of action for the highest ROI?

There are a number of services providers that have proprietary processes to correct approximately 20% to 50% of records that are not +4 coded or have a missing/invalid secondary (apartment) number. However, even after this hygiene processing, there will most likely still be 1% to 3% of records with deliverability issues. Now the questions become: Should all of these be mailed? Should none of these be mailed? What is the best course of action for the highest ROI?

Even though a record has deliverability issues, it doesn’t mean it won’t get delivered. Here are two examples:

Jane Jetson
124 Main St
Saint Cloud, MN  56301

Jane’s address is not +4 coded because 124 Main St is not a valid delivery point. However, 123 Main St. is valid and is, in fact, where Jane Jetson actually lives. Because of this, there is a strong likelihood this mail piece will be delivered.

Fred Flintstone
456 Oak River Run Apt C3
Saint Cloud MN  56301

Fred’s deliverability issue is an invalid apartment number.  Instead of C3 it is 3C. In this case, there is also a very strong likelihood this mail piece will be delivered.

The goal for marketers is to identify and drop records with deliverability issues performing below breakeven to ensure they only mail to the most effective set of records. Over the past year I’ve worked with a top 200, top 100 and a top 50 retailer whose breakeven deliverability analysis saved them each between $900,000 and $1.7M annually.

How would it affect your budget if you had similar results? If direct mail is a critical part of your marketing ecosystem today, read more about breakeven deliverability analysis.

There are two types of NCOA^Link processing available – Full Service Providers with 48 months of change of address (COA) data and Limited Service Providers/End User Mailers with 18 months of COA data. Unfortunately, some who use NCOA^Link regularly (monthly or bi-monthly) don’t understand the differences between the 18 and 48 month databases or the benefits they provide.

In four recent tests, each of the mailers used a Limited Service Provider for the previous year for their processing.  When these files were matched to the Full Service NCOA^Link database, the overall average match rate was 1.66%.  Of these matches, just over one half of one percent (or 34 percent of the total matches) had a Move Effective Date that was over 18 months old, which means they would not be picked up by a Limited Service Provider.

Coding additional records with ZIP+4^™ codes and correcting some of the records with missing/invalid information by using proprietary processes beyond CASS™ also means a significant number of additional valid NCOA^Link matches will be identified. In four tests against four other Full Service Providers who did not have these proprietary processes, there was an average of 7.5% additional valid NCOA^Link matches that they did not get.  One word of warning about these proprietary services – there are some companies offering these today where the quality is not good, so you need to be careful of which service provider you use to do this.

The sequence of processing that includes NCOA^Link can also provide significant benefits if done correctly. Running NCOA^Link before identifying duplicates will identify additional movers. In a recent test for a large mailer, my company moved 3.86% of the records via NCOA^Link.   Of these moves, 19.6% of them became duplicates.  This increased the number of duplicates identified by 2.42% or by .75% of the input records.

But there is also value in running NCOA^Link again just before the file goes to the printer.  This is primarily due to the move dynamics in the U.S.  Since NCOA^Link is updated weekly and 15 percent of the population moves each year, typically for every week since NCOA^Link was last run, about one quarter of one percent of the records being processed will have moved. It is cost effective to run Last Effort COA for almost all of our clients where it has been two weeks or more since NCOA^Link was last run.

Using all or some of the recommendations below when using NCOA^Link® will provide significant benefits to your company :

Use a full service provider for NCOA^Link processing. For those companies with an end-user license or those who are currently using a Limited Service Provider, at least semi-annually and possibly even quarterly, records should be processed by a Full Service Provider to pick up older moves that would otherwise be missed. 

• Use a service provider that has proprietary processes to accurately apply ZIP+4^TM codes to additional records and correct records with missing/invalid apartment numbers.  If your current provider does not have these capabilities, at least twice a year have your database cleaned by a company that does have these capabilities and that accurately makes the adjustments to the address.
• Run NCOA^Link® before deduping to identify additional valid duplicates.
• Consider running Last Effort COA if it has been two weeks or more since NCOA^Link was last run on the records

Applying these recommendations could provide a benefit of about $20 per thousand records mailed in terms of increased response and reduced wasted advertising.

In addition, companies who apply these steps will improve mailing results while having a more accurate and complete view of customers.

-Reactivate old customers – The holidays are a great time to reach out to inactive customers.  To improve the performance of your reactivation efforts, ensure that your old customers are still at the same address by using an external verification service. Our clients have seen a 40% lift in response from a verified name and address versus a non-verified one.  And during the holidays, the lift in performance may even be greater.

-Use NCOA^Link® – Run NCOA^Link on your house records as part of the processing for every holiday mail campaign vs. monthly or even less frequently.  For each week since NCOA^Link has been run, .2% (or almost 1% per month) of your customers have moved, and mailing to a person’s old address dramatically reduces response. Testing shows that if you mail to an old address the response rate is 75% less than if you mailed to the right person at the right address.

-Maximize response and performance – Use proprietary change of address tools from a service provider to get the most updated address for the targeted people on your mailing list. These proprietary Change Of Address (COA) tools can identify changes of address for some of the 22%+ of movers who do not file a change of address form with the USPS^®.

-Mail those who have responded to you in the last 12 months, even if the addresses have deliverability issues – Make sure in your zeal to identify and drop records with potential deliverability problems that you don’t drop those where ROI is above break even. Interestingly, testing has repeatedly shown that the performance of 0-to-12 month responder records with deliverability issues perform well above break-even. This is even truer during the holidays when the delivery of mail is at its best.

-Include “Or Current Resident” when your offer is more generic (catalogs, for example) – This makes sense over the holidays because despite a dramatically reduced response rate, your piece will get delivered to someone as opposed to being thrown out – and since more people are shopping you have the potential to get some response.

-Target the right people – Use tested and validated models on both your customer and prospect mailings. Incorporate the depth and breadth of external data to create stronger models.  And prospect models are much more effective when done at an individual/household level versus a ZIP™ level.