Donât get AI to write emails. Get it to fill in the blanks.
 â â â [Demand Curve]( [Read on demandcurve.com]( The Growth Newsletter #170 Donât get AI to write emails. Get it to fill in the blanks. This one is for all the people out there who are tired of bad cold emails.  Let's just dive in  â Neal Brought to you by [Aurora](âa new way to run cold outbound at scale without being spammyâand it finally just works.  Matt and the team have sent millions of emails, generated nearly $1B in pipeline, and are neurotically obsessed with getting the little details right.  [Learn more about them here](, or read on to learn how to replicate their process. Want to be featured in front of 97,640 founders and marketers? [Learn more here](. Donât get AI to write emails. Get it to fill in the blanks. Insight supplied by [Aurora](.  AI still doesnât write usable copy.  Itâs verbose and either overly formal or casual. Peopleâs internal AI-detection algorithms are already well-tuned to sniff it out.  This is particularly true with cold emails. People are already wary of you.  But AI is good at writing tightly constrained variables. You can use these variables inside templates written by humans to make them feel fully personalized to get responses like this: Hereâs how to use AI to make cold emails that work  When I talked with Matt from [Aurora](, their process really impressed me:  #1. Pick a specific dream customer.  Draft a specific email for them. This is a forcing function to write a great email youâll use as a template.  #2. Break down the dream email.  Identify the static parts (your name, company, etc) versus the parts that should be customized for each person.  Donât just constrain yourself to boring variables like company name. Instead, think: - Common pain points your customers experience
- âJobs to be doneâ of your product or their product
- Summary of LinkedIn posts or company/personal bios
- Their current goals as an organization (fundraising, scaling a channel, etc) Â For example: #3. Use the 3 C's to build prompts to fill in the blanks. - Context: Give ChatGPT the context it needs to know what to do.
- Creative Constraints: Tell it exactly what you want the output to look like (x words or less, casual, lowercase, no quotes, etc.) The data you feed into the prompts can come in two forms: - Explicit data:
- Data from LinkedIn, like a company's description or a post.
- Scraped data from company pages, job boards, fundraising databases, etc. - Inferred data: Input for a prompt can be the output from another earlier prompt. For example, for the first email in Auroraâs outbound campaign (example above)
- Use the LinkedIn company bio + scraped homepage as inputs to build a âdream ICPâ output from ChatGPT.
- The ICP output (and the company bio + scraped homepage context) can all be used as inputs to another prompt to output a "pain moment" output.
- Both are used as inputs to determine what companies should look for in their prospects' job posts and whether they need their service. What the above template looks like when you get AI to fill in the blanks (read [Mattâs post]( for example prompts used): ^ Notice this doesnât have a CTA. Heâs just sharing an idea for free. Heâs not trying to close a sale immediately. It warrants a response since it doesn't feel like a pitch.  If Mutiny or LexCheck want to run with the idea themselves, cool. If they want to learn more and maybe work with Mattâs agency, even better.  Setting this up is a bunch of work  Itâs far less work than manually writing thousands of emails, but it still takes time and expertise to get it all set up, running, and converting. And way more effective than "spraying and praying."  If this is interesting, we recommend working with the folks at [Aurora]( who supplied this tactic. Theyâre experts in all of this.  Yes, they sponsored this post, but I love this tactic because most cold emails are terrible, and AI is making it worse.  [Learn more about Aurora â](  Resources to help implement this yourself - Read [Mattâs blog post]( that dives deeper into all of this. It includes ChatGPT prompts and examples.
- He also has a [few rough notes]( with more example prompts.
- Read our past tactics on [Sales](, [Copywriting](, and [Email](.
- Learn how to write [really good hooks]( for your emails.
- Follow [Rob Lennon](. He's got some pretty good posts and courses on using AI. What did you think of today's newsletter? ð Loved it: Forward to a friend or send a replyâa ð will do! If really helps. ð¤·ââï¸ Meh: You can unsubscribe [here](), or manage your subscription [here](. ð¤ I'm new here: You can join the party [here](. Something fun  From [thenintendoworld](. The world is a strange place. â How we can help you grow - Read our free [playbooks](, [blog](, [growth guide](, and [teardowns](âwe break down the strategies & tactics used by fast-growing startups.
- Enroll in the [Growth Program](, our marketing course that has helped 1,000+ founders get traction and scale revenue.
- Become [UNIGNORABLE](. The April 22nd cohort of our audience building course is coming. The pre-sale sold out in 10 minutes. [Join the waitlist](.
- Need help running ads? Weâve built [the]([ ads agency]( for startups.
- Get in front of startup founders by [sponsoring]( this newsletter. Thanks, everyone!  I'm currently really excited about the next cohort of [UNIGNORABLE](. It's basically where all my attention is going at the moment. It's going to be fun! [Neal]( [Neal O'Grady]( [Grace]( [Justin Setzer]( â © 2024 Demand Curve, Inc. All rights reserved. 4460 Redwood Hwy, Suite 16-535, San Rafael, California, United States
[Unsubscribe]() from all emails, including the newsletter, or [manage]( subscription preferences.