Supercharged AI outreach
Connecting people and collaborating with others are key business and life skills. The ability to make introductions and link together the various parts of your network can be hugely valuable from a sales and relationship building perspective.
So, for the next few weeks, Three Things is going to practice what it preaches. We’re going to be joined by a number of guest contributor’s, all of whom we either know well and bring a new and exciting perspective to the world of sales and business development within professional services.
Who better to kick things off than the newest edition to the LINAR family, our AI lead, Sam Stamp. Over the next few minutes, Sam’s going to explore how Generative AI (GenAI) can supercharge the drafting of cold outreach and follow up. Through the intelligent and strategic use of Large Language Models (LLMs) it’s possible to drastically speed up the rate at which you can draft effective copy. Here’s his top tips. Enjoy.
#1. Choose the right LLM
There are a lot of different LLMs available for you to use, and they all have slightly different strengths and weaknesses. Understanding your specific needs is the first step in selecting the most suitable LLM for your tasks. It is also important to understand if your organisation has any particular guidelines, although this does not appear to be the norm at the moment.
- Research: If you need to gather and synthesise information, models like OpenAI’s GPT-4 are generally considered market-leading at the moment.
- Drafting Messages: For crafting personalised emails or messages, tools like Jasper or Copy.ai can be more intuitive, and are designed specifically for this purpose.
- Budget Considerations: Decide whether you need a premium version for advanced features or if a free, generic model suffices. Familiarise yourself with the nuances of your chosen LLM to fully leverage its capabilities, before investing in any paid versions.
#2. Use the right prompts
Effective prompting is the bedrock of any effective engagement with GenAI tools. For most LLMs (e.g. Chat-GPT, Claude, Google Gemini), they are general purpose – ready to create the ‘most likely’ answer to any question given. In order to get something more specific to your needs, you must be more specific in your request.
- Craft Specific Prompts: Instead of generic queries, provide detailed instructions to guide the AI. For example, “You are the Sales Director of a leading legal consulting firm. Please draft a sales email targeting mid-sized law firms highlighting our new AI integration service. Focus on the core benefits of immediate productivity and efficiency gains.” (more another time on why it’s still good to say ‘please’ to an LLM!)
- Prompt it again, Sam: There’s no reason you can’t ask for something to be re-written again and again and again. Each time you can add more detail or ask for further clarifications and changes. There is likely a diminishing return here though, and at some point you can take what’s there and finalise yourself.
- Data Privacy: Always ensure you’re not sharing sensitive or personal data in your prompts. Avoid using client-specific information that isn’t publicly available.
#3. Check the output
LLMs generate responses based on patterns, not direct searches, so you always need to double check what you’re sharing. This isn’t only for risk of ‘hallucinations’ (where an AI has literally just made something up simply because it is producing something that is ‘most likely to be true’) but also for tone, style and relevance.
- Verify Information: Cross-check the generated content for accuracy and relevance.
- Understand Context Windows: Be aware of how much prior conversation the LLM considers when generating responses, and ensure it includes all necessary context. For example, you may have requested certain specific language was used, but after an hour or so of crafting messages, that language may no longer being referenced.
- Stay Human: Remember that you are still the arbiter of what goes out with your name attached. Just because an LLM produced something at lightening pace does not mean it’s right. If you want to change some or all of it, do – sometimes it’s just the fact of not starting with a blank page that is the biggest benefit.
In the near future, as AI technologies advance, writing the ‘perfect’ outreach email will become standard practice. This shift means building genuine relationships and in-person interactions will become even more crucial. Allow AI to handle the routine, repetitive tasks, freeing you to focus on what truly matters—building meaningful connections.
For more insights and hands-on experience, join our upcoming workshops designed to help you integrate GenAI into your business processes seamlessly. Register your details here.
