Skip to main content
All CollectionsSequencing
Advanced Sequencing Strategies
Advanced Sequencing Strategies

Guide to advanced Autobound email sequences using AI-assisted content generation, with focus on template structure and fallback instructions.

Joyce Frias avatar
Written by Joyce Frias
Updated over 2 months ago

Overview

This guide provides advanced tips and strategies for maximizing the effectiveness of your Autobound sequences. By leveraging AI-assisted content generation and following best practices for email structuring, you can significantly improve your outreach efforts and boost response rates.

Watch this video and read below to learn more.

Structuring Your Emails: The Prompt/Template Hybrid Approach

To get the most out of Autobound's AI-generated content, we recommend using a prompt/template hybrid approach. This method allows you to maintain control over key messages while leveraging AI for personalization and creativity.

Key Tips for Effective AI-Assisted Emails:

  1. Use square brackets [ ] for instructions and guidance to the AI.

  2. Leave non-bracketed content relatively intact.

  3. Add instructions at the top of the email to set the overall tone and purpose.

  4. Use inline brackets throughout the email to guide specific sections or transitions.

  5. Try typing prompt-like instructions for more control over the output.

  6. Provide Fallback Instructions: This is critical. Always include fallback instructions for the AI when requesting specific data usage. This prevents the AI from making up information when the requested data is not available.

Email Structure Components

When creating your email templates, include the following components:

  1. Subject Line: Include instructions for AI personalization.

  2. Overall Instructions: Place these at the top of the email body.

  3. Pre-written Content: This will remain mostly unchanged.

  4. Bracketed Sections: Include specific instructions for AI-generated content.

Example Templates

Example 1: Cold Outreach

Subject: [clever subject line, 5 words or less, most words are lower case, super unique, personalized, and eye catching. i.e Autobound Co-Founder reaching out]

[instructions: write an INTENSELY, hyper-personalized email focused on {{first_name}}'s SPECIFIC pain points. 100 words or less. Lots of paragraph breaks so it's easy to read].

Hi {{first_name}},

[super personalized opening sentence that calls out something relevant based on the prospect, ideally linked to the necessity of email personalization, how hard it is, the need to grow pipeline faster, or low SDR productivity. if nothing can be found that's actually relevant to their pain points, then instead ask a question that's super focused on their pain points given their job title. write this in CLEAR-AS-DAY manner].

[explain the unique selling proposition of Autobound, but do NOT lead with features AT ALL. Instead, focus on the unique, HIGHLY specific ways that Autobound solves problems for companies similar to theirs. Be hyper-specific in listing a few insights we track that are ultra-salient to what they might care about].

Open to hearing how our AI-writer is helping similar teams 1.5 - 2x their reply rate?

Best,

{{sender.first_name}}

Example 2: Free User Upsell

Subject: [multiple sellers at the prospect's company have signed up for Autobound, Reference that in the subject line, keep it under 6 words]} {{account.name}} reps using Autobound + Quick chat?

[instructions: write an INTENSELY, hyper-personalized email focused on {{first_name}}'s SPECIFIC pain points. 100 words or less. Lot's of paragraph breaks so it's easy to read. The purpose of this email is to notify the prospect that multiple reps on their team have signed up for Autobound to help improve response rates and save time. Ultimately, there is a lot we can do to improve their team's accounts to help them book more meetings so the email should encourage the prospect to take a meeting to discuss this further].

Hi {{first_name}},

{{sender.first_name}} here, I lead biz dev at Autobound. Great to e-meet you.

{{account.custom7}} reps on your team are using Autobound to improve response rates and save time crafting personalized content. Is this an initiative you're focused on?

[super personalized opening sentence that calls out something relevant based on the prospect, ideally linked to the necessity of email personalization, how hard it is, the need to grow pipeline faster, or low SDR productivity. if nothing can be found that's actually relevant to their pain points, then instead ask a question that's super focused on their pain points given their job title. write this in CLEAR-AS-DAY manner].

For your understanding, here are a few things Autobound found on your that can be used by the AI to write personalized messaging:

[The purpose of this section is to showcase info we found on the prospect. List 4-5 tidbits of information found on the prospect and their company to demonstrate how Autobound can surface relevant information. Only do this if you find relevant insights. If there is no information on the prospect, just give some examples of insights we could potentially find, such as YouTube references, podcast mentions, LinkedIn posts, but make sure the prospect knows those are just examples. When referencing info on the prospect, call out where you found the info, i.e (from your LinkedIn posts or profile, in the news article on prnews.com, etc]

As a sales leader, I thought you'd be interested to learn how we can collaborate to help your team's messaging improve even further.

Do you have 30 minutes to explore how we can help your team generate more [revenue, opportunities, or pipeline, whichever is most relevant based on their seniority in their job title]?

Thanks a ton,

{{sender.first_name}}

Critical Note on AI Instructions and Data Availability

It is crucial to understand that the AI will attempt to follow your instructions regardless of whether the specific data you've requested is available. This can lead to the AI potentially making up information if not properly guided. To prevent this, always include fallback instructions in your prompts.

Example of Proper Instruction with Fallback:

[Personalize the email using the prospect's recent LinkedIn posts. If there's no information on the prospect's LinkedIn posts available to personalize off of, then don't use it. Instead, leverage other prospect info available. If nothing's available, just make the email hyper-relevant to the pain points {{userCompanyName}} solves for the prospect. Never make up any information]

This approach ensures that:

  1. The AI uses LinkedIn post information when available.

  2. If LinkedIn data isn't available, it falls back to other available prospect information.

  3. If no specific prospect information is available, it focuses on general pain points your company solves.

  4. The AI is explicitly instructed never to fabricate information.

By following this practice, you maintain the integrity and accuracy of your personalized outreach while still leveraging the AI's capabilities to create relevant content.

Best Practices for Fallback Instructions:

  1. Always Provide Alternatives: Give the AI multiple options to choose from based on data availability. Consider just sayings “leverage the available prospect information”

  2. Be Explicit About No Fabrication: Clearly state that the AI should not make up information if none is available.

  3. Default to General Relevance: Provide a fallback that focuses on your company's value proposition or general industry pain points when specific data is unavailable.

  4. Use Conditional Statements: Structure your instructions with "if-then" statements to guide the AI's decision-making process.

  5. Review and Test: Regularly review your templates and test them with various prospect profiles to ensure they produce appropriate content in all scenarios.

Did this answer your question?