How to Sell an AI Service With Cold Email

TL;DR

Sell an AI service by testing multiple combinations of niches, offers, lead sources, and email copy, then removing weak variants and improving the winners. The demonstrated outreach produced 78 replies, including 34 opportunities, 27 positive replies, and 19 meeting requests, showing how market feedback can guide gradual improvements instead of relying on a single untested campaign.

Transcript

In this video, I'm going to start and sell an AI service completely from scratch. I'm going to do it live on video in front of you right now. I have no idea what the service is. I have no idea who the heck I'm even selling it to, and I don't even know what degree of results I'm generating. You're seeing all this stuff ahead of time, but I'm going t... Read More

Key Insights

• The campaign generated 78 total replies, including positive reactions, negative responses, and follow-up questions about how the proposed service worked. These varied responses provided direct market evidence about whether the offer and messaging attracted attention.
• An opportunity is defined as a recipient who asks positive questions about the service or expressly indicates a desire to proceed. Using that definition, the outreach created 34 opportunities and recorded 27 positive replies.
• The offer generated 19 meeting requests from people who had no prior relationship with the sender. Interested recipients shared phone numbers, proposed times, or asked for a short conversation about moving forward with the service.
• The outreach framework consists of four variables: target niches, niche-specific offers, lead-scraping methods, and email copy. Each variable is developed separately and then combined with the others to create campaigns that can be compared through real responses.
• Two alternatives across each of four campaign variables create 16 possible variants because 2 multiplied by 2 multiplied by 2 multiplied by 2 equals 16. This structure tests more market surface area than relying on one niche, offer, lead source, and message.
• The market is treated as the most reliable source of feedback because campaign advice begins as speculation until prospects respond. Testing multiple combinations reveals what works for the chosen service and audience without requiring blind trust in another person's preferred formula.
• Reply-rate improvement comes from dropping weak campaign variants, continuing strong ones, and iterating on the copy. The example progression moves from an initial 2 percent average to 3 percent, then 3.5 percent, with 5 percent or more presented as a desired later result.
• A suitable B2B niche for this method is digital and high ticket, with purchases above $5,000. The stated rationale is that digital businesses have fewer physical cost constraints and that one additional customer can create enough value to justify a substantial service fee.

Questions & Answers

Q: How do you sell a new AI service with cold email?

Start by selecting several plausible digital, high-ticket niches from first principles. Develop different offers for those niches, identify multiple ways to scrape relevant leads, and write alternative email copy using the variables collected during lead research. Send the resulting campaign variants, compare replies and meeting requests, remove weak combinations, and continue refining the strongest offers and messages based on market feedback.

Q: What results did the AI service outreach campaign produce?

The campaign received 78 total replies. Those responses included positive interest, explicit rejection, and questions requesting more information about the service. The results were classified into 34 opportunities, 27 positive replies, and 19 meeting requests. Some interested recipients supplied phone numbers, asked to proceed, or proposed a specific time for a short conversation despite having no previous relationship with the sender.

Q: What are the four variables in the cold email testing process?

The four variables are target niche, offer, lead-scraping method, and email copy. Niches determine which kinds of businesses receive the outreach, while offers are adjusted to suit each niche. Lead-scraping methods determine how prospects and useful variables are collected. Copy then presents each offer, potentially using those collected variables to provide more options for tailoring the message.

Q: Why should cold email campaigns test multiple variants?

Multiple variants let the sender compare many combinations and discover which one produces the best return for the least input. Choosing two niches, two offers, two lead-scraping approaches, and two copy versions creates 16 combinations. This approach reduces dependence on speculation because recipients' replies reveal which combinations deserve continued investment and which weak performers should be dropped.

Q: How can cold email reply rates improve over time?

Reply rates can improve by treating the first campaign as an initial test rather than a final answer. The process removes nonperforming variants, keeps the stronger combinations, and revises the copy repeatedly. The example describes a possible progression from 2 percent to 3 percent, then 3.5 percent, followed by further copy iteration toward a desired rate of 5 percent or more.

Q: How should a niche be chosen for digital outreach?

A target niche should preferably operate digitally because digital businesses avoid many constraints associated with physical operations, such as rent and inventory costs. The process also favors businesses that can direct a larger share of their budget toward growth. Agencies and e-commerce businesses are given as examples of digital categories, while brick-and-mortar companies are described as more complicated targets for this outreach method.

Q: Why does the process favor high-ticket B2B businesses?

The process favors businesses selling purchases above $5,000 because the service provider can justify fees through a relatively small number of additional customers. If an AI service helps a high-ticket client gain one more customer each month, that customer can represent another $5,000 in monthly revenue. A low-ticket product would require far more customer acquisitions to create comparable value for the client.

Q: What counts as an opportunity in this outreach campaign?

An opportunity is a reply from someone who asks positive questions about the proposed service or expressly indicates that they want to proceed. It is narrower than the total reply count because total replies also include negative reactions and other responses. Under this definition, 34 of the campaign's 78 replies qualified as opportunities, while 27 replies were reported as positive.

Summary & Key Takeaways

• The process begins by choosing target niches from first principles rather than automatically accepting existing assumptions. Suitable niches for digital outreach should operate digitally and sell high-ticket products or services. Digital businesses can allocate more of their budgets to growth, while offers above $5,000 make the economics of a B2B service easier to justify.

• After selecting niches, the process creates niche-specific offers, multiple lead-scraping methods, and alternative versions of email copy. Testing two choices in each of these four categories produces 16 combinations. This broad testing surface lets actual market responses reveal which combinations generate the greatest return for the least input instead of depending on speculation.

• The initial outreach generated 78 total replies, with responses ranging from rejection to follow-up questions and direct interest. Among them were 34 opportunities, 27 positive replies, and 19 meeting requests. The proposed optimization cycle removes poor performers, retains successful variants, and repeatedly revises copy so average reply rates can improve over time.