New 42-day free trial Get it now
Smarty

How I reduced my returned mail from 27% to 1% using address autocomplete

Mail carrier at door
Updated October 29, 2025
Tags
Mail carrier at door

The following is based on a true story. Some of the names and relationships have been changed to protect the anonymity of individuals and companies. However, the numbers are 100% accurate.

In 2023, I wanted to mail some really fancy cards to 165 businesses. I collected their addresses by asking for them or finding them in their online listing and collected them all in a neat little row. Then, I went a step further and ran these addresses through Smarty's bulk address validation tool. Everything was set and perfect.

Or so I thought.

Slowly, one by one, they started getting returned. One day, six came back at once, and I can't tell you how uncomfortable it was when my boss walked in holding those expensive mailers.

"I thought you checked these," he said.

I did. I did check them.

After about eight weeks, I stopped getting the mailers returned, and I did an audit to try and figure out what the heck happened. Here's what I found.

Of the 165 mailers sent out, 44 were returned for one reason or another. That's an enormous 27% returned (and money wasted). I had to figure out why they were returned, and I figured I was in luck since MOST of them had a note from the USPS. Here's the count:

USPS’s reason for returnNumber of returned mailers
Not deliverable as addressed13
Insufficient address9
No such street4
No info4
Attempted not known4
Unable to forward3
Vacant2
No mail receptacle2
No such number1
Forward time expired1
Not deliverable as addressed -and- wrong address

1

 

When I started comparing this data with the results I got when I ran my data through Smarty's address validation tool, I noticed something that made me put my tail between my legs as I had to explain it to my boss. Yes, the validation tool said that these addresses were in the USPS database, and the USPS summarized them as "Match-Mailable." 

However, when I scrolled through all of the metadata that Smarty delivers, I found DPV footnotes and enhanced match data that would have opened my eyes to the fact that many of these addresses shouldn't have been mailed to in the first place.

If an address is in the USPS database, it doesn't always mean you should mail to it. 

As you can see with the nine "insufficient address" results, the USPS had the address for the complex these recipients were in, but without a suite number, the address was insufficient to be delivered to. That alone would have helped drastically reduce my returned mail.

What about the "not deliverable as addressed" that made up the largest group of returned mail? Well, I checked the USPS website and learned this about "UAA" (undeliverable as addressed) addresses:

Mail may be considered UAA because:

  • The individual, family, or business to which it’s addressed has moved
  • The address is incomplete, incorrect, or illegible
  • The addressee is unknown or deceased
  • The addressee refuses or fails to claim the mail

There are many reasons, and the USPS doesn't tell you which of these is correct. I know that it was sent to businesses, so I guess there's a chance that they relocated. I know that my printed labels were legible and that there were people at the businesses I was sending them to. But like I said above, it's possible that I was missing a part of the address or that the company refused my little mailer.

Ultimately, it was a super bummer.

A year went by. The trees blossomed with spring and stood tall and proud until summer waned. As the leaves fell to the ground, we decided to mail out another set of mailers to people wrapping up their year at work.

This time, I wasn't going to make the same mistakes.

Man thinking meme

I sent out an email telling people that we wanted to mail them a gift and that all I needed was a mailing address. Instead of allowing people to hit reply and type in whatever bogus mailing address they wanted to, I had them click a link that took them to a landing page with an address form on it. 

My secret weapon? 

That form had Smarty's address autocomplete on it.

It was super easy for me to set up and tie to our CRM. And because I remembered the pain of getting "Match-Mailable" from the USPS but not being able to deliver, I set up some additional parameters in the tool.

Here is a link.

First, I didn't allow addresses to pop up to be selected if they were 'non-postal,' meaning they weren't recognized as USPS deliverable addresses. Second, I made sure to look at the DPV footnotes and enhanced match data for each address that came through. Lastly, I set up a flag to notify me if the person fully typed out the address instead of clicking on an autocomplete option.

The results were astounding.

I sent out 189 mailers this time, and only 2 came back. One came back because the individual had put the incorrect suite number, and I was able to resend it to them. The other was marked "Not Deliverable as Addressed," and I was unable to contact the recipient.

My boss never walked over to me holding a stack of returned mailers.

No one called me, upset that they didn't receive their mailer.
 

Phone that is not ringing, because address autocomplete did its job

My customers had a slick user experience when entering their addresses.

If you'd like to use autocomplete on your forms, a free account will give you 1000 addresses, and you can purchase as many as you need after that. Setting it up doesn't have to be hard, either; we've put together documentation and tutorials to help you get up and running right away.

May your returned mail be reduced to dust.

Subscribe to our blog!
Learn more about RSS feeds here.
Read our recent posts
What to look for in an address data solution: Ease of implementation
Arrow Icon
You’ve maybe chosen an address data provider (or maybe you’ve just recently fallen out of love with the one you’ve got). Now comes the time to really test what matters: implementation. Even the best solution falls short if it’s hard to integrate, confusing to use, or impossible to maintain. This blog is part of our five-part series, What to look for in an address data solution. Previously, we discussed why human support teams should be top of mind when choosing an address data solution. Not just tech support, but educational materials, help getting started, and more.
What to look for in an address data solution: Quality of support
Arrow Icon
Selecting the right address data solution is only the beginning. Rolling it out successfully, resolving issues quickly, and keeping it performing at its best all depend on the quality of support behind it. This blog is part of our five-part series, What to look for in an address data solution. Previously, we discussed what it means to be accurate, what questions to ask any address data provider before entering into a contract with them, and how to avoid the accuracy “validation gap. ”Today, we’re focusing on an often-overlooked element: a supportive, accessible, and responsive partner.
What to look for in an address data solution: Data accuracy
Arrow Icon
Once you know what capabilities you need, the next challenge is choosing a provider that actually delivers—pun intended. This blog is part of our five-part series, What to look for in an address data solution. Previously, we discussed the need to find an address data solution that understands the importance of speed, scalability, and reliability, enabling you to maintain automated, efficient, and accurate workflows in any industry that deals with addresses. Today, we’ll focus on the hidden costs of inaccurate data, defining and combating the “validation gap”, and why it’s important in any industry.

Ready to get started?