We use cold email because it works. Cold email, like any other skill, takes time to develop. But you can achieve significant improvements quickly.
Over the course of a week, we were able to increase our email open rate from a pedestrian 17.4% to a ridiculous 47.8%. There is good news. This process is repeatable.
I’m going to share with you the exact step by step process we used in this article, breaking down our thought process behind each step.
Our General Process
The first step we took was establishing a general process we would use to make our emails better. We are big proponents of the experimental method.
One excellent (and much more detailed) guide on the experimental method can be found here, but the general steps are below.
- Identify what you know and what you think you know. Write out your assumptions and the hypotheses you would like to test.
- Select one hypothesis to test and create an A/B test to determine if it’s correct.
- Perform the test.
- Analyze results – what happened?
- Draw conclusions – what do these results tell us? Are the results significant enough to validate our hypotheses or are further experiments needed?
Choosing What to Experiment on First – Our Purpose
The first question we asked ourselves was, “What is the most effective way to improve our cold email efforts?”
There are many things we could have focused on: the email body, the call to action, the subject line, length of the email, time the email was sent.
Our assumption was that the subject line would be the best place to begin improving our email campaigns.
Why We Chose to Test the Subject Line First
Joseph Sugarman, one of the most legendary copywriters of all time, puts it succinctly:
“The sole purpose of the first sentence in an advertisement is to get you to read the second sentence.”
In the world of email, that first sentence is the subject line.
To us this makes the subject line the most important part of the email. If people do not open your email, it does not matter how spectacular the body of your message is or how compelling your call to action.
Constructing a Hypothesis to Test
To set up an experiment we first need to construct a hypothesis around subject lines. What do we believe would improve our subject lines? Why do we think that? How can we tell if we were right?
Applying What We Know
We need to take into account what we already know or think we know, to make an informed guess on what will happen.
We all receive copious amounts of email. If the metric we want to increase is open rate, it is easy to look at our own personal experiences and see which headlines we ignore and which ones we don’t.
Unexpected “sales” emails were universally ignored. For illustration, look at these headlines I received in my inbox in the past few days.
- “Re: OEM Vape band and mouthpiece”
- “Pokemon’s everywhere! Time to Start Playing”
- “15% off your order”
Based on this, our first assumption was made. Overly business headlines will not be as effective as more personal emails.
Everyone in our office opens personal emails. So we looked at some of the subject lines from personal headlines.
These are some examples.
- “As it is”
- “”Grrr”
- “A little something”
- “Duck”
Looking at these headlines we decided to add another two assumptions for subject lines that get opened. Firstly, shorter headlines work. All of these headlines are super short, the longest being three words.
Secondly, these headlines pique curiosity. They aren’t overly descriptive. They entice you to click through to read the email.
All of this is straightforward. But most of the emails I receive fail.
Determining a Metric
Next, we needed to identify a metric to determine if our experiment was successful or not.
We considered two metrics.
The first was the open rate or the percentage of recipients who click and open the email.
We like this metric because it is obvious the subject line is the most important variable. There are only two visible variables, the subject line and the first sentence of the body, and of the two the subject line is much more prevalent.
We also considered the reply rate (or click through rate depending on the type of email sent).
We liked this metric because ultimately, this is what matters most – what percentage of customers show interest and respond to our offer.
However, the purpose of this experiment is to optimize the subject line. Because there are so many other confounding variables in the reply rate (body copy, call to action, tone, length etc), we chose open rate as our metric.
The Final Hypotheses
These assumptions led to our hypotheses: Open rates will increase by 30% because short, personal, subject lines that arouse curiosity will outperform longer, descriptive, business subject lines.
Setting Up the Experiment
In order to test our subject lines we set up a simple A/B test. The plan was to send out two emails with different subject lines and measure the open rate to determine the winner.
Both emails were identical in every aspect, including sending address, salutation, body, call to action, and signature. The only difference was the subject.
We controlled the audience by pulling emails from the same source and randomizing who received which message.
Our First Subject Lines
Our First Iteration
To test our hypotheses we crafted two subjects.
Our personal subject header was “Quick Question”.
Notice that this subject line echos all three of our assumptions. It is personal, short and attempts to solicit curiosity in the reader.
Our business subject was “NEW Camera App for Contractors.”
This is a standard business styled email subject we are using as a control. We then sent out a ton of emails. Here are the results.
Quick Question: 26.7% open rate
NEW Camera App for Contractors: 17.38% open rate
That is a massive difference. The more personal subject received 50% more click throughs than the business oriented subject line.
Our Second Iteration
However, one single test doesn’t fully validate a hypothesis. Our subject could just be bad in general. We repeated the experiment with two different subject lines.
For the personal subject line we tried “Reaching Out”.
For the more formal subject line we dropped the all caps and instead went with “Camera App Built for Contractors”.
Again, notice that the assumptions that we laid out in our hypothesis are well expressed. “Reaching Out” is short, personable, and leaves an implicit question – “Why are you reaching out?”.
The results were surprising. First, the new business line.
Camera App Built for Contractors: 21.81%
Eliminating one word, “NEW”, increased click through rate by over 25%! Despite this increase it still lagged behind the previous personal email by a large margin.
Next we looked at how the new personal subject line did.
Reaching Out: 47.8% Open Rate
This new subject almost doubled our previous high mark. It also reinforced the previous results. The personal subject line outperformed the business control for the second time.
At 47.8% open rate we felt were ready to move onto working on the email body and call to action.
Subject Line Learnings
I believe this experiment produced two solid learnings for us and our particular market.
Small changes can create huge differences. Removing a single word “NEW” increased open rates from 17.38% to 21.81%. It proves that when testing copy you can’t just write off a whole approach. You have to spend time with it and get better. For our market, personal outreach leads to better open rates. The results help validate our initial hypotheses as to why.
I do not think that two experiments are enough to fully validate this finding. However, we are more than happy with our current open rate.
Since the initial experiment we have used the subject line a number of times, achieving 33.8%, 37.6%, and 32.8% open rates. I think this is an important note. There is significant variance even with the same subject line.
This deviation could be caused by changing audiences (we’re sending to different lists), and highlights the importance of trying to control for audience as much as possible.
Despite these lower numbers, the “Reaching Out” subject line has still performed well beyond all other iterations.
Conclusion + Next Steps
Going forward we want to apply these learnings to our followup emails. We will begin with various personal, short subject lines crafted to pique interest.
Further, we want to begin testing the actual content of our email.
Open rate is not the end goal. We want the recipients to engage with us. We can now begin testing the actual body of the email – length, tone, word choice, formatting etc.
Cold email has become an important source of new leads for us. As we continue to improve the other base components of the reply rate, cold email will become more and more profitable for our company.
Do you use cold email to grow your business? What subject lines have you found effective?
Image: New email message in the inbox concept
