6 A/B Testing Strategies for Direct Mail | Lob

A/B testing helps you figure out which elements of your direct mail campaigns actually drive responses — and which ones you're spending money on for nothing. It's standard practice in digital marketing, and it should be just as essential in your direct mail strategy.

Without testing, you're essentially guessing. And guessing means wasted budget on creative, printing, and postage that might not be driving responses. Worse, you risk reinforcing tactics that underperform because you never had a control to compare against.

A/B testing your direct mail works the same way it does in digital. Design two versions of a mailpiece that are identical except for a single element — format, size, CTA, imagery, or offer mechanism. Send each version to a comparable segment, measure the results, and let the data tell you what's working. The key rule: change only one variable at a time so you can isolate what actually made the difference.

Here are 6 A/B tests you can run on your next direct mail campaign:

1. Custom image vs. generic image

Most teams default to generic product shots or stock images. That's a missed opportunity. Personalizing images for each recipient — based on prior purchases, browsing behavior, or product category — can lift conversion rates by double digits. But personalization isn't the only angle worth testing. You can also compare lifestyle imagery against product-focused visuals, or test whether images of people outperform images of the product itself.

Start simple: send one segment a generic image and another a personalized one, then compare conversion rates. Once you have a winner, test secondary variables like image placement or background color.

Wayfair uses product-specific imagery tailored to browsing history in its customer acquisition campaigns. Notice how they match product categories to recipient interests in the example below.

2. Copy and headlines

Your headline is the first thing a recipient notices — and it's where most copy tests should start. Question vs. statement is one angle ("Ready to save 20%?" vs. "Save 20% today"), but it's not the only one. You can also test benefit-led headlines against urgency-driven ones, or short punchy copy against longer explanatory text.

Don't stop at headlines, though. Test the body copy, too. Does leading with a customer pain point outperform leading with a product benefit? Does a conversational tone drive more responses than a formal one? Each of these is a testable hypothesis. Pick one variable per send, and let the results sharpen your messaging over time.

3. Size and format

Size matters — but so does format. Most teams debate whether bigger postcards outperform smaller ones (4" x 6" vs. 6" x 9"), and that's a valid test. Larger formats tend to grab more attention in the mailbox. You can even test an 11x9 bifold self-mailer to see how it stacks up.

But don't overlook format entirely. A postcard vs. a letter. A self-mailer vs. a tri-fold brochure. A letter signals formality and importance; a postcard signals urgency and simplicity. Match the format to your message.

4. One CTA vs. multiple CTAs

CTAs deserve more testing than most teams give them. One key question: does a single, focused CTA outperform giving recipients multiple options? For example, compare a straightforward "Buy now" against a piece that offers both "Buy now" and "Start a free trial." In most cases, a single clear CTA outperforms multiple options — decision fatigue is real.

Beyond quantity, test the CTA language itself. A question-based CTA ("Ready to save 20% on your next order?") creates a different response dynamic than a direct command ("Order now and save 20%"). Same offer, different framing — and we've seen response rates shift by 15% or more.

5. QR code vs. promo code

QR codes bridge print and digital — scan to land on a personalized offer page, app download, or checkout flow. The QR code could simply display a promo code when scanned, or it could take users to an app or website with the special offer. Or test whether including both outperforms using just one.

Discover places QR codes prominently in its acquisition letters — notice how the code leads to a pre-filled application.

6. Personalized URL vs. generic URL

Most direct mail includes a generic URL — something like yourcompany.com/offer. That works, but you can't track which recipients responded. Personalized URLs (PURLs) flip that. A PURL like yourcompany.com/janedoe lets you track exactly who responded and what they did next.

The test here is straightforward: does a personalized URL drive higher response rates than a generic one? Some audiences respond better to PURLs — the personalization signals that the offer was made specifically for them. Others might find a shorter, simpler URL easier to type in.

Where to go from here

These six tests will sharpen your campaigns — if your testing process is solid. Keep your experiments clean:

• Test one variable at a time. Isolate your variables.
• Set clear objectives before you send. Ask yourself: what are we trying to learn? Define your primary metric — response rate, conversion rate, cost per acquisition — so your analysis stays focused.
• Set realistic, measurable goals. Have a benchmark to compare against so you can determine whether a result is a genuine improvement or just noise.
• Give each test enough time. Direct mail has a longer response window than email. Don't call a winner too early. It may take multiple cycles to optimize.
• Document everything. Track which variable you tested, the sample sizes, the results, and what you learned. This becomes your playbook for future campaigns.

Beyond what's covered here, there are plenty of other elements worth experimenting with — colors, offers, mailing lists, send timing, even envelope design. And once testing becomes routine, each campaign builds on what you learned from the last.