Can I A/B test images in emails?

You can absolutely A/B test images in emails. The short answer is yes. But there's a catch most guides skip over: images don't affect open rates. Opens are tracked before images even load. So if you're testing a hero photo against an illustration and watching open rates for a verdict, you're measuring the wrong thing.

The metric that actually tells you something is clicks. Did people click through the version with the lifestyle photo more than the product-on-white-background version? Did the version with no image at all outperform both? That's where your data lives.

What's worth testing

The most common image variables that move the needle are image subject (lifestyle vs product shot), style (photography vs illustration), placement (hero at the top vs inline next to text vs none at all), size, and whether you use a single focused image or a set of smaller ones. Those aren't just aesthetic choices. Each one changes how the eye moves through the email, and that affects clicks.

How to set it up

Most major ESPs support A/B testing natively. In Mailchimp, Klaviyo, or Brevo, you create two versions of the same email, swap only the image element you're testing, and send each version to a random split of your list. Keep everything else identical. Subject line, copy, CTA, send time. If you change more than one thing, you won't know what drove the difference.

Still one practical thing to keep in mind: image tests need bigger sample sizes than subject line tests. The effect is often subtler. If your list is under a few thousand active subscribers, you might not reach statistical significance before you'd naturally send again anyway. In that case, run the test across a few sends and look at patterns over time rather than expecting a clean winner after one campaign.

The image-blocking problem

A chunk of your audience will never see your images. Outlook blocks images by default for many recipients. Corporate email environments often do the same. That means your A/B test results reflect the subscribers who did see the images, not your whole list. It also means you should always test with images blocked before you send, to make sure the email still makes sense without them. If your entire message relies on a visual that half your readers never load, that's a design problem worth solving separately.

On that note: don't rely only on the image to carry the click. Your CTA button or linked text should still do the work. The image is supporting cast, not the whole show.

If you want to go deeper on what kinds of images actually move click-through rates, there's a lot to unpack there. And if you're curious how ALT text plays into your test results, that's worth a look too.