There’s a line where email marketing crosses-over into spamming. A recent test has me thinking a lot about it.

By cutting our send list by -50% we sent less emails, saw +10% better email engagement, beat forecast for the promotion.

Expert Best practices

I have been working with an email deliverability experts for the past 2 years and he has been suggesting constantly that sending email to unengaged users will hurt your ability to be seen in the inbox. Sending lower volumes of email to a higher-engaged will get a better.

Leadership Pushback against best practices

The mandate from leadership is to send the promotion to the entire list, following a “do not leave any money on the table” strategy. We did that for several years and got great results. However, the effectiveness begins to wane.

Lower visibility = YOLO opportunity

In this era of 2026, the AI era is coming full steam with tens-of-thousands of tech layoffs so the likelihood of anyone having the same job at the same company over the next 24 months is probably not good.

Figuring I probably am going to lose my job pretty soon anyway, I figured it would be a good time to test out the battle of “best practices” versus “no stone unturned” approach to email marketing. YOLO!

Setup:

We only sent emails to the people that we know like the product: email clicks within 1-2 years, any current or previous customers, or anyone with recent application usage regardless of payment.

We have two tiers of product. A super high-ticket price offer and a standard offer. Past paying customers see the high-ticket offer. New customers would get the standard offer.

Results

On the first day, rather than sending emails to our full list, we sent to our top 50% most-engaged. Results were impressive in spots, but poor in others.

Highlight: Our best-engaged audience was more-engaged than ever. They bought the high-ticket offer in droves, exceeding expectations and driving ARPU up considerably higher than our typical non-promotion week.

Lowlight: The lower-engaged audience had less sales than typical years because the volume was low. Purchase rate for this audience remained stable year-over-year so by sending dramatically lower volume we also got less sales.

Key takeaway: Best-practices work. It took some courage (or recklessness) to follow the guidelines but now we have the proof that it works.

A beer-pouring analogy

I have been recently thinking of deliverability as a beer keg.

I supply the keg full of customers but the bartender ESP/ISP controls the flow.

If I supply a keg of bad beer, I will get inherently bad results no matter how well it’s poured.

If I supply a keg of good beer, the ISP bartender can only fill the glasses so fast. If i try to send all of my emails at once, the ISP pours too fast. The mug fills with worthless foam, even if it’s a good beer.

If you supply a good keg of customers and give the ISP time to pour a proper drink then you will have a good result.