I Enriched My Outbound Sales Emails with AI for 30 Days - Here's What I Learned

In this blog post, I'll walk you through my step-by-step process for using AI to create highly effective, personalized email templates that will take your outbound sales to the next level.

As major companies like Apple (‘Apple Intelligence,’ New OpenAI Partnership) implement AI and AI startups seed funding totaled $23 Billion for the last 3 months (Perplexity AI, Motion, and Anthropic receive large investments), understanding how to use AI in your role is becoming pivotal.

Many people think that using AI is as simple as providing a prompt and getting an output. However, to produce effective AI outputs, you need to give rules, background information, and your preferred structure, and employ A/B test factors to contribute to building your prompts to get relevant and valuable results.

I have worked in outbound sales for over 4 years, and I have made it a habit to document email templates, subject lines, and outbound best practices that I have observed or received from colleagues. I have saved all of this information within one Google Document that I used to power my AI prompts.

I know firsthand the challenges of crafting personalized, engaging emails at scale. Over the past month, I have been experimenting with AI to better understand effective usage in my role and improve my day-to-day tasks. However, when AI for sales is used incorrectly, your emails can come across as robotic or, even worse, be flagged as spam.

Step 1: Gather Relevant Informationi for sales

To produce effective AI outputs, you need to provide rules, background information, and preferred structure, and employ A/B testing. I started by documenting email templates, subject lines, outbound best practices, calls to action, value propositions, and persona problems I've observed or have been shared with me by colleagues. I saved all of this valuable information in a single Google Document that I shared directly with the AI.

Step 2: Provide Specific Promptsi for sales 

The next step in creating an AI-generated email is giving a specific prompt that references the product, mentions a problem, and provides an answer to the problem based on your company's value proposition through your offering. I prefer to give AI a specific example like my first email below. Use something like this as a starting point, then provide rules in your prompt, such as "keep this under 200 words, start with a problem the persona faces, and use specific numerical benefits."

Here's an example of how AI improved my email copy:

Pre-AI Email: 

Hi {{first_name}}, I wanted to reconnect as Dow Jones recently launched a new Due Diligence tool. Our new tool Integrity Check provides cost-effective and streamlined due diligence checks through the speed of AI and research by Dow Jones.

Integrity Check is an AI-powered Due Diligence platform, leveraging natural language processing (NLP) that allows users to simply and efficiently address their due diligence needs by providing a deep dive analysis on a company or individual. This tool combines the entire Dow Jones Risk & Compliance Database, Licensed Corporate Records: 140+ global jurisdictions, and Global Internet Search. 

Would you be interested in connecting briefly to learn more? If so let me know and I'd be happy to schedule a quick call to discuss further. I also attached an overview of the product. 

I'm looking forward to hearing from you! 

AI-Generated Email: 

Protecting your organization from compliance risks is crucial, but traditional due diligence can be time-consuming and miss critical information.

Dow Jones Integrity Check revolutionizes due diligence using AI and natural language processing (NLP) to efficiently identify risks associated with third parties and customers.

Key benefits:

• Quickly identify and mitigate potential compliance risks like OFAC 50% and Adverse Media

• Access concise reports summarizing key risk factors

• Streamline enhanced due diligence for third-party risk management and financial crime compliance

Our AI-driven approach combines Dow Jones Risk and Compliance Data, licensed corporate records, and open web content to provide a comprehensive view of individuals and organizations, helping your team mitigate risks in as little as 5-10 minutes.

{{first_name}}, interested in learning how Integrity Check can help you mitigate risks faster? Let's schedule a brief call to discuss your needs and demonstrate the power of our AI-powered solution.

Best regards,

Examining the differences 

This AI email comes across as easier to read and provides specific value propositions like "helping your team mitigate risks in as little as 5-10 minutes" versus "providing cost-effective and streamlined due diligence checks." Additionally, it opens with a problem statement that a prospect can identify with in their day-to-day work.

Step 3: AI-assisted Proofreadingi for sales 

For AI to effectively proofread and identify stronger writing, I create prompts that give it a scenario and ask it the strengths, weaknesses, and where it would improve my outreach. 

For example, to review my Risk and Compliance LinkedIn Request, I asked AI:

You are a chief compliance officer at a Fortune 1000 company and you receive this LinkedIn connection request: 

"Interested in how AI-powered due diligence can uncover compliance risks for 3rd parties using AI/NLP? Love to chat about how we can streamline processes with concise risk reports. Let's connect - I'll share how it can help [company]"

Is this too pushy? Would you be likely to respond? Does this sound useful? Let me know and provide steps to make it better 

The AI Response:

As a Chief Compliance Officer at a Fortune 1000 company, here are my thoughts on the connection request:

Is it too pushy? Slightly. While the message gets to the point quickly, it comes across as a bit too sales-y and presumptuous right off the bat.

Would I be likely to respond? Probably not initially. The request doesn't provide enough context or value upfront to pique my interest and make me want to connect.

Does it sound useful? The offering of AI-powered due diligence and streamlined processes does sound intriguing, but the message lacks specifics on how it would benefit my company's compliance program.

Steps to make it better:

Personalize the opening line beyond just my name and company. Reference my role, and experience or highlight a compliance challenge my industry faces.

Provide a bit more context on Dow Jones and your compliance solutions before jumping into the AI/NLP offering. Instead of asking to connect right away, offer a piece of valuable content (article, case study) that demonstrates your expertise first.

Then invite me to connect if I'd like to learn more about how the AI solution could address my company's specific compliance needs.

An improved version could be:

"As a CCO in the [industry], ensuring robust due diligence on third parties is likely a key priority. At Dow Jones Risk & Compliance, we help Fortune companies streamline this process. I thought you might find this [case study/article] on using AI for enhanced due diligence insightful. Happy to discuss how our solutions could benefit [Company] as well if you'd like to connect."

This approach establishes credibility, provides value upfront, and opens the door for further engagement if I'm interested, without being too pushy initially. [END Response]

However, this response doesn’t fit the character limits on a LinkedIn Connection Request. We need to provide exact rules within our next prompt to ensure it will fit our parameters and best practices. Here’s what that next prompt looks like: 

My Prompt: Remember LinkedIn connection requests best practices (must be under 250 characters and is common courtesy to keep it around 200 characters) now update this LinkedIn connection request based on those rules and your suggestions in the last prompt: "As a CCO in the [industry], ensuring robust due diligence on third parties is likely a key priority. At Dow Jones Risk & Compliance, we help Fortune companies streamline this process. I thought you might find this [case study/article] on using AI for enhanced due diligence insightful. Happy to discuss how our solutions could benefit [Company] as well if you'd like to connect.”

AI Response: Here's a revised LinkedIn connection request under 200 characters for you as a Chief Compliance Officer at a Fortune 1000 company:

“Interested in leveraging AI/NLP for due diligence? Our solutions help companies like yours quickly identify risks for third parties using AI/NLP-based due diligence. Happy to share more via connection and a link to a relevant case study/resource.”

Step 4: Personalization and Measuring Successi for sales 

Personalization is key to effective sales emails, and AI can help streamline this process. By recording pain points, issues from other vendors, and typical problems faced by specific job titles, you can feed this information to your AI to reference in your prompts.

For my risk and compliance customers, common pain points include false positives, a large number of alerts, and outdated or missing information. By addressing these issues in my AI-generated emails, I can create targeted calls to action that identify buying signals and drive conversions.

It's important to track and benchmark the performance of your AI-generated emails using metrics such as open rates, click-through rates, and replies (both positive and negative). Keeping track of before and after usage, new templates, and different techniques are vital KPIs. Use this data to refine your email strategy, re-engineer prompts, and improve results over time.

Step 5: Scaling with Variables and Human Oversighti for sales 

Most CRMs allow you to add variables like {{first_name}} or {{Account_name}} to make your emails more personalized at scale. Ensure these variables are used in calls to action or when asking questions about prospects' problems to make your emails more engaging.

While AI can generate effective email templates, human oversight is crucial to ensure the emails align with your company's voice, values, and overall sales strategy. The last thing you want is an email going out saying "((company name))" or "ABC Media, LLC" instead of "ABC Media." Review and edit AI-generated content to maintain authenticity and effectiveness. AI is here to assist Sales Professionals, not replace them, so it is our job to learn how to use AI effectively.

Lastly, always remember where you are uploading data and be careful sharing sensitive information. Never share pricing, competitive intelligence, or private company information with AI platforms. Unless your company has cleared you to do so or you are utilizing an internal AI solution.

Are you ready to harness the power of AI in your sales emails? Start by documenting your email templates, best practices, and customer pain points, and share this information with your AI. Experiment with different prompts, refine your approach based on AI feedback and track your results.

To help you get started, I've created a free checklist for prompt engineering with examples you can use to guide your AI-powered email creation process. Download it now and start supercharging your sales emails with AI today!

Email Prompt Engineering Checklist for Sales.pdf

Download PDF • 182KB