I talk with business owners almost every day, and they share similar experiences, that I’m sure you’d recognize in your own calendar as well. You started your business to do the work you’re brilliant at, the client strategy and the creative thinking that no one else can do. But your day is so often consumed by the other things, the constant hum of administrative tasks that are must to run a business, yet so painful to get done sometimes. The endless inbox, the meeting notes that need summarizing, and the follow-up emails that need writing, the calls to prepare for, the proposals to prepare, the invoices to send out, you name it. This is a common for each one of us small business owners, and it’s where using AI for business can help you get some of your time back.
The goal is to give you a capable assistant for the tasks that drain your and your team’s energy without anyone feeling that they have been replaced. Think of it as a way to get the first draft of your repetitive work done, so you can apply your expertise to the final, human touch. It’s a way to delegate the routine parts of your day so you can focus on the parts only you can do. We’re going to walk through five specific areas where you can start this week, and you can pick one to start with.
1. Let AI manage your inbox
Your inbox is probably a perfect example of a place where a little help could go a long way. So much of what arrives doesn’t need your immediate, strategic attention. You can use an AI tool to act as a first-pass filter, sorting the day’s emails by priority or identifying which ones contain direct questions that need a reply. This will help you clearing the noise so you can see what matters and what needs your attention
- What to hand over: Ask your AI tool to read through a batch of emails and categorize them. You can ask it to identify anything that seems urgent, find all the client questions you need to answer, or even write a first draft of a common response, like a reply to a scheduling request. This is the kind of “mundane” work that the U.S. Small Business Administration suggests is a good starting point for AI, because it’s repeatable and doesn’t require deep strategic thought.
- What to keep human: The final edit of any reply must be yours, so it sounds like you and contains your specific knowledge. You also keep the final decision on what is truly urgent. The AI can suggest priorities, but you have the context to know that a short email from a key client is more important than a long one from a new vendor.
2. Never write meeting summaries yourself again
How much time do you spend after a great client call translating your messy handwirtten notes into a clean summary with clear action items? I used to do that myself. Funny, but even though I am a tech person, I still very much enjoy writing notes with my hands. There is just a different level of retention and gives me so much joy to take notes with my hands. The problem was, that for those notes to get into my computer, I had to dedicate specific time…this meant going through them one-by-one and typing them up. You can imagine how many of my notes ended up in the system…almost none…Then I ended up flipping back in my notebook, trying to find what I was suppose to send to the person. So stressful and not efficient. Good old days…
For me, using AI for a task like this has been such a help! This is one of the easiest and most satisfying tasks to hand over. Many tools can transcribe your recorded calls, and from there, you can give the transcript to an AI and ask for exactly what you need.
- What to hand over: Once you have a transcript of your meeting (always get permission before recording), you can ask an AI to create a concise summary of the conversation, pull out the key decisions that were made, and list all the action items with the names of who is responsible for them. This can turn 30 minutes of post-call cleanup into two minutes of review. A 2023 article in the Harvard Business Review on generative AI and productivity points to this kind of work, stating that AI is particularly good at synthesizing information and creating first drafts of documents, which is exactly what a meeting summary is. And AI got way better at this since 2023!
- What to keep human: The AI will give you a literal summary, but you were in the room. Your job is to read the AI’s draft and add any important nuance the machine might have missed. You might add a note about the client’s tone when they agreed to a deadline, or add a personal touch to the email before you send the summary over. The final document should combine the AI’s efficiency with your human insight.
- Pro Tip: I still take notes during the call, just in Notion. Then I have AI save the summary in the same page I used during the call. This way, everything is in one place and a lot easier for me to find what I am looking for.
3. Let AI draft your next proposal
Building a proposal is strategic work, but much of the document is often assembled from standard parts. You likely have sections you use over and over, like your company background, your process, or descriptions of your core services, testimonials, etc. Instead of copying and pasting these yourself, you can teach an AI to do the assembly for you.
- What to hand over: You can give an AI tool your standard proposal sections and then feed it your notes from a discovery call with a new client. Then, ask it to assemble a new draft of a proposal based on that client’s specific needs. You can even give it a description of one of your services from your website and ask it to generate a list of frequently asked questions a client might have about it. This gets the basic structure on paper, so you aren’t starting from a blank page.
- What to keep human: The most important parts, the strategy, the pricing, and the custom details that show the client you were truly listening. The AI can build the container, but you have to fill it with your expertise. Your final review is essential to make sure the proposal reflects the value you provide and is tailored perfectly to the person who will be reading it. If you want some help thinking through how to define those strategic parts, a free strategy call can be a great place to start.
- Pro Tip: Proposals used to be a dread for me. Even if I got really excited about working with a potential client, the fact that I had to pull everything together into a proposal seemed like a dread. This is not the vibe you want to go with. Ask AI to create a template for you that you can reuse for future proposals. This way, the core features will always be done already, it will all look consistent and professional. You only need to create this once, and then only need to spend time adding the specific parts for that particular project you’re pitching for.
4. Ask AI to write your follow-up emails
Following up after you’ve sent a proposal or had a networking meeting is necessary, but it can feel awkward to find new ways to say, “Just checking in.” This is a small, contained writing task that AI is quite good at handling. It can save you the mental energy of figuring out how to be both polite and persistent.
- What to hand over: Give the AI the context. For example, “I sent a proposal to a potential client a week ago and haven’t heard back. Please write three short, friendly follow-up emails I could send.” You’ll get different options, so you can choose the one that feels right. This use of AI for drafting client communication is already happening at a large scale. A 2023 Boston Consulting Group report noted that a company called Octopus Energy used generative AI to help its agents handle customer emails, resulting in faster and more consistent responses.
- What to keep human: The timing and the personal touch. You know your relationship with the client and can decide when it’s the right time to send a follow-up. Before you send one of the AI’s drafts, you should always add a small personal note that a machine could never know, maybe referencing a small detail from your last conversation.
- Pro Tip: LLMs are now able to access your emails directly and you can set up an automated task that will draft up follow ups for you. If you feel comfortable giving access to your emails, this tool can save even more time for you by regularly checking your emails and drafting emails to send.
5. Turn one conversation into a week of content
Many founders I work with are full of great ideas but struggle to find the time to turn those ideas into articles, social media posts, or newsletters. If you can talk about an idea, you can use AI to handle the first pass of turning it into written content.
- What to hand over: Record yourself talking about a topic you know well for 10 or 15 minutes. Get it transcribed, then give that text to an AI. You can ask it to turn your thoughts into a blog post outline, a series of short paragraphs for a newsletter, or five social media posts based on the key points. This process separates the act of thinking from the act of writing, which can make content creation feel much more manageable. You can find more ideas for this in our Resources library.
- What to keep human: The AI is not the expert, you are. Your original idea, your unique perspective, and your voice are the whole point of the content. The AI is just helping you with the structure and the typing. Your job is to take the draft it gives you and rewrite it so it sounds exactly like you and delivers the value only you can provide.
- Pro Tip: If you truly want to make your content expertise-based and personal, start taking notes of stories that come to you that can help people understand you content better. Then use them when asking AI to write for you.
One important warning about your client data
As you begin to use these tools, it’s incredibly important to remember one thing: do not paste sensitive client information into public AI tools. Most of the common AI models use the data you give them to train their systems, which means your private notes could become part of their knowledge base. Never input confidential details from client contracts, private financial data, or anything else you wouldn’t want to become public.
Instead, generalize the information. If you want help with a proposal, give the AI a generic client description, not a real company’s name and internal struggles. The principle is to give the machine the pattern of the task without giving it the private details.
Pick one to try
The point of all this is to find one small point of relief in your week, not to become an AI expert overnight. Looking at this list of five tasks, my advice is to simply pick the one that made you feel the most immediate sense of, “I would love to not have to do that anymore.”
Try handing just that one task over to an AI assistant for a week. See how it feels to get a first draft done for you, and notice where it frees up your time and attention for the work that really matters. If you find it helpful and want to go deeper on how to apply these ideas with your team, our Training sessions are designed to do just that.
