My learnings on cold email – in case helpful to any other founders

cats

New member
DISCLAIMER: You must have a product that someone wants. Otherwise none of this will work.

Anyhow...

We are early stage and have been building an outbound motion. To my surprise, having never sent or read a cold email before, it has been the most reliable and predictable sales channel we have. Would love to share my lessons if they can give anyone a head start.

We got triple digit demos last month from cold emails alone. The right combination of targeting, tools, messaging and follow-up worked for us. Some of this was obvious, a lot less so.

This has been working for us despite of Google's changes.

Warm-Up

Before you even get started with anything else, do yourself a favor and start getting emails warmed up. Deliverability is the most important part.

I'd recommend Smartlead but cannot say that I have tried anything else. It was recommended where we did research and has worked for us. People talk about Insantly too, it might be worth checking out.

Put your emails on warmup for ideally two / three weeks before you start sending anything.

Identify your ICP

This one is obvious though takes some thought and experimentation. What kind of company? What role at that company? The more granular this is, the better. If you don't know this, create a hypothesis and learn from the data.

Tools & Data

You need good data and a lot of it. Doing this by hand is not scalable, unless you want to spend all day going to websites and trying to find email addresses.

We boiled it down to the following formula and found the right tools to execute:

A. Export large set of filtered data from data source (e.g. Pitchbook, Crunchbase)

B. Identify key decision makers via LinkedIn Sales Navigator Advanced

C. Gather email addresses and related data (i.e. name, title) via Apollo (use their LinkedIn extension)

D. Cleanse email data with ZeroBounce

Getting to clean data will benefit tremendously from some basic Python scripts for efficiency vs. XLSX or Google Sheets And now has never been a better time even without any chops/experience with GPT and VS Code, if you are a non technical founder.

Messaging

I don't think I've ever responded to a cold email. Neither had anyone on our team. So our starting place was in the mindset of "how can we create something actually compelling?"

Personalization is good - in theory - but this is also where people get things wrong. "Hey {{first name}}! I know {{ company }} has XYZ service are you interested?" is not awesome IMO.

We personalized by building a tool that would use DocsHound to generate screenshots of a user guide for a company using only their URL. It would run on autopilot in the background without any manual effort. And that would be inserted into every email. Meaning every single prospect received a customized screenshot, and we could generate hundreds a day without laying a finger.

Deliverability

Avoid URLs in the email. We do not even put our company URL in. Run your email through Mail Tester. Remove spammy language. Mail Tester evaluates for spammy language but there are other more rigorous tools for that purpose.

Run your emails through GMass Inbox tester until you see a sea of green. Make sure you have follow ups. Make sure you have an unsubscribe link. And make sure NOT to send another email to anyone who unsubscribes.

Also – use spintax. Have as many variations as possible. If you do not what that is, use Google. Simple in theory, coming up with many variations across sentence structures is painful.

RECOMMENDATION: DO NOT USE AI FOR FINAL COPY. You can use it to help you craft the copy – i.e. give feedback. But lazily slapping it in will put you straight into spam. It's too easy for the algorithms to detect. "It is crucial" "Unlocking the power of" etc – instantly in spam folder.

Patience

Don't send too many. It's better to have cold running in the background on an ongoing basis vs. trying to do a big blast and then having to warm up domains and run through the whole process again. I blew this by using one tool that had no spacing between messages to send 300 in a minute, which blew our entire cold email for an entire month. Don't make that mistake!
 
@cats
"Hey {{first name}}! I know {{ company }} has XYZ service are you interested?" is not awesome IMO.

We personalized by building a tool that would use DocsHound to generate screenshots of a user guide for a company using only their URL. It would run on autopilot in the background without any manual effort. And that would be inserted into every email."

show us an example of your email if you do not mind.
 
@cats This is so cool. Would you be willing to share how you went about building the tool that generates those screenshots and then inserted them into the email?
 
@jeanada Our product does it itself!

i.e. on DocsHound during the onboarding all you do is provide a URL and then it drafts a user guide and places you into a Notion-esque editor

–– one note, we have a couple extra steps if you sign up and go through the onboarding in the product to get the table of contents nailed down. Providing a screenshot of the product and another task mapping exercise.

It's still solid without those steps of user input, so that's what we insert in the cold email.
 
@cats Thank you for taking the time to share!! Very cool stuff. So, is DocsHound something you used to help, or it’s the actual product you are selling?
 
@cats Tried to DM you a few times but Reddit is being weird so I’m not clear if it went through. I’m actually evaluating tools for onboarding, training, and support documentation now. DM me if you want to chat.
 
@cats Agree that reply rate doesn’t matter, but open rate is an indicator of deliverability. Essentially the tip of the spear in your sales funnel. Small improvements here could have an outsized impact to demo rate.
 
@poppy25 iOS 15 had an impact. Overall sentiment among email marketers is that the noise:signal ratio is high.

Even if there is some signal, it's not worth it for us as any and all content that is not banal workaday plaintext can hurt you.

You just have to decide where to place your bets. We'd rather optimize for deliverability vs. open rates.
 
Back
Top