Back
All Resources

Taking Academic Email Marketing Research into the Real World

The Name Similarity Effect
Brett Friedman

The Big Idea

It's a common conversation in my family that a Friedman from any part of the world will help any other Friedman out on the sole principle of our shared name. As an Ashkenazi jew, I know the same goes for ethnicity. Jews help each other out.

But it's not only Friedmans and Jews who give preferential treatment to those like them. It's everyone.

In fact, this concept is one of seven "Principles of Persuasion" rolled out by renowned speaker Robert Cialdini in his work on social influence. Moreover, there's a number of academic papers on the topic.

The Inspiration

Recently, a new paper by NYU PhD Kurt Munz was published in the journal, Marketing Science, entitled Name Similarity Encourages Generosity; this extends that research to Email Marketing.

In short, the study found that when someone was asked to donate to a teacher with the same last name as them, they were more likely to open the email, click the email, donate, and donate more than the control group. Moreover, even matching just the first initial of the last name significantly increased those core metrics. One last finding, matching ethnicity predicted donations as well as name similarity, but not as much so for email open rates.

The Hypotheses

There are a few caveats of the research and its transfer to cold email: it was done through a charity with a list of email recipients who had previously donated, people had to be unaware of the fact that our names matching was intentional, and rarer names have a larger effect i.e. more common last names are less likely to induce generosity through similarity.

My audience was much colder and might more easily guess that their name similarity was orchestrated given how common hyper-personalization is in email marketing.

All that in mind, I set out to one core question:

Does the name similarity effect transfer to cold email?

My hypothesis was simple: yes. Yes it does.

I also wanted to take this test a little further to learn if first name matching worked equally well, as well as if the combination of ethnicity and last name first initial would compound results. I would have tested ethnicity and last name, but it's implied that sharing a last name shares ethnicity (barring marriages and adoptions which I can't control for.)

Altogether, I tested matching:

  • Ethnicity
  • First Name
  • Last Name
  • Last Name First Initial
  • Ethnicity + First Initial
  • Randomized Control Group

The Experimental Method

To implement the test, I found about 100 emails for each category. While a true test would have 100,000 emails, I'm resource limited. Not to mention the key to a good marketing test is starting small, then doubling down on what works.

To match ethnicity, I chose stereotypical Ashkenazi Jewish last names: Cohen and Rabinowitz.
To match ethnicity and first initial (F), I used another common Ashkenazi Jewish surname: Frank.
To match first initial without ethnicity, I took a typical non-Jewish last name that starts with F: Farrell.
The other groups are self-explanatory.

To ensure emails were relevant, all recipients were either senior marketers or executives at ad agencies.
To make sure recipients knew my name, I used a spelled out email address: brettfriedman@marpipe.com and signed off as Brett Friedman.
In order to stay true to the original research, I slightly modified the email used in the study reproduced below.

Obviously the format was imprecise for my needs, but I tried it out. With a hunch that this type of email would not work for cold outreach, I sent it to 20 people in each category. Around a quarter replied with "unsubscribe".

Happy I did a small test before the real test, I replenished those 20 people per group and switched to my standard drip. While we infrequently use cold email, I've had success with a particular template in the past.

To hide my secrets, I will not reproduce it here, but in a way, it's the inverse of the DonorsChoose content. I offer $5 or an ebook to recipients if they offer feedback on our upcoming SaaS platform.

Over a period of one business week, Monday through Friday, 20 emails were sent per day per group. At the end of the week, I reviewed results.

The Results

The results were great, not perfect, but good enough to say the test was a success!

Icons from left to right: Successful Sends, Open Rate, Click Through Rate, Reply Rate, Unsubscribe Rate, and Bounce Rate

You'll first notice the bounce rate was pretty high in all categories. As you might expect from the current economic conditions, there were a lot of Out-of-Office auto-replies as well as many '[name] no longer works here' responses.

Aside from that, unsubscribes were relatively low across the board. Notably, most unsubscribes were from Friedmans. Perhaps they were the ones that knew by intuition I reached out to them because we shared a name. Or more likely it's a statistical artifact because base rate opens by Friedmans were higher than any other category aligned with the initial research and hypotheses.

Friedmans also clicked links more, replied more, and signed up for Marpipe more! In fact, the only leads we got from this campaign were all Friedmans! Unsurprisingly, matching ethnicity had second most opens, but much fewer link clicks and replies than other categories.

Matching last name, first initial with and without matching ethnicity performed about the same. Notably, there were no unsubscribes from non-Jewish matching first initial recipients.

The other stand out finding is that other Bretts do not like me or Marpipe. One possible reason is that Brett is traditionally a non-Jewish name, so they're not matching ethnicity AND people may have guessed the emails were sent solely because our names matched, as it seems less likely a cold email would be sent to someone with the same first name with no ulterior motives.

All in all, the randomized control group was second to lowest across opens and clicks, then dead last for replies while having the lowest bounce rate. That makes the other findings all the more significant.

The Implications, The Future, and The Applications

The first implication is that academic research findings are transferable to the real world. The second is that I am exclusively sending emails to Friedmans for the foreseeable future - until I contact every Friedman in marketing. Going forward, I'd like to test different geographic variables if there are enough Friedmans to experiment with. Perhaps getting closer to home will improve leads generated.

If your names are also relatively common, but not too common, see if you can replicate my findings. If you liked the structure of this content, Marpipe is launching a 'business experimentation' newsletter. You can sign up here. If you have an experiment of your own to show off, email brett@marpipe.com to pitch your topic. If it fits, you can be featured in one of our first weekly issues!

Read our full creative testing guide

SIGN UP FOR THE EXPERIMENTAL METHOD NEWSLETTER
TO GET THE GUIDE

Download Now
View Sources

Heading 1

Heading 2

Heading 3

Heading 4

Heading 5
Heading 6
"This is a block quote"
  1. This
  2. is
  3. an
  4. ordered
  5. list

This text is bold

The rich text element allows you to create and format headings, paragraphs, blockquotes, images, and video all in one place instead of having to add and format them individually. Just double-click and easily create content.

A rich text element can be used with static or dynamic content. For static content, just drop it into any page and begin editing. For dynamic content, add a rich text field to any collection and then connect a rich text element to that field in the settings panel. Voila!

How to Run a Multivariate Test

The Beginner's Guide

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Tiffany Johnson Headshot

How to Run a Multivariate Test
The Beginner's Guide

Plus, Get our Weekly
Experimentation newsletter!

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.