Do AI memes get more replies in cold email?

Short answer: no.

In our own outbound, adding an AI-generated meme to the final email didn’t move replies, and it didn’t lift meetings either.

This is part of our How AiSDR Uses AiSDR series, where we publish the wins and the misses. The full spreadsheet lives at aisdr.com/experiments/.

We’ll be completely honest with you here: we genuinely wanted the memes to win this one. For months, we had been dropping an AI-generated meme into the final email of our sequences, and it simply felt right to the whole team. It seemed more human and less robotic, which is exactly the kind of touch that earns a reply when one more polished pitch would get ignored. We were fully prepared to write the post about how a well-placed meme beats a plain-text email.

Hypothesis

The thinking behind it was reasonably simple. Inboxes are crowded and competitive places, and the final email in a sequence is usually where forgotten deals quietly go to die. A meme placed right at the very end of that sequence might be just enough to jolt a distracted reader into finally replying.

So we naturally expected the meme arm to pull in noticeably more replies, and ultimately far more meetings, than the plain-text version ever would.

Methodology

We run our own outbound on AiSDR, and before we made the public case for memes, we decided to test the idea properly.

We ran 2 campaigns aimed at the same audience type, which was people who had recently engaged with a competitor’s LinkedIn posts. Both campaigns deliberately used the exact same ICP, the same sequence structure, and the same send window. The only variable we changed was the final email, which carried a meme image in one arm and plain text in the other.

Results

In the end, the meme arm simply did not win, because the response rate came out effectively flat, at 5.39% with memes against 5.32% without, while the meetings per engaged lead stayed statistically indistinguishable, at 0.625% versus 0.718%. The replies themselves certainly did show up in the inbox, but the underlying pipeline simply refused to move at all.

The plain-text arm booked only 5 meetings in total, so its slightly higher meeting rate is really small-sample noise rather than a genuine win. Even with that imbalance in the data, the memes showed essentially zero lift on the one metric that we care about most.

Memes may well pull additional replies out of people. But those replies don’t necessarily convert themselves into booked meetings. Here is a real example that came straight out of the meme arm. The meme itself landed nicely, but the meeting never materialized, because the lead simply passed us along to another team instead of booking any time on the calendar.

Decision-makers often perceive memes in cold email as unprofessional, and whether a meme lands successfully depends heavily on the industry and the particular type of company that you’re targeting. The people who do reply warmly are often not the serious buyers you need.

So if your only goal is more replies, then memes might look like a reasonable win. But if your real goal is pipeline, then they have to prove that they lift meetings rather than only engagement. In our own controlled test, they clearly failed to do that.

Key takeaways

What’s next in the lab

The next test is already queued up. We’re comparing AI memes against a plain-text first message, this time at balanced volume, and we’re tracking positive replies per lead as the primary metric.

Want to see every experiment we run, including all the losses? The live spreadsheet lives at aisdr.com/experiments/.

FAQ

Do AI memes increase cold email reply rates?

Not according to our data. In a matched-ICP test, the meme arm replied at 5.39% and the plain-text arm replied at 5.32%, which is effectively flat.

Did memes book more meetings?

No, they did not. The meetings per engaged lead were statistically indistinguishable, at 0.625% with memes against 0.718% without, on a considerably smaller plain-text sample. The memes demonstrated no measurable lift on pipeline whatsoever.

Should I use memes in cold outreach?

It really depends quite a lot on your particular audience. Keep memes switched off by default for senior buyers in regulated industries. They may be worth a test for peer-to-peer audiences like sales managers and RevOps ICs, but only if they can prove that they lift meetings rather than only replies.

What is AiSDR testing next?

We’re pitting AI memes against a plain-text first message, this time at balanced volume, with positive replies per lead serving as the primary metric. Every experiment, wins and losses alike, gets published at aisdr.com/experiments/.