I shared a process automation earlier this week on Linkedin. It’s how I’m handling prospect calls and generating proposals. The response on Linkedin was pretty strong and that’s exciting—everyone loves automation, right? The thing that a lot of people don’t realize is that it took weeks for me to get to a point that I could automate this process.

It’s not that it’s super complex or anything but there was a point where I didn’t even know that I needed a process, let alone what it was, or how to build it.

How do I know I need a process?

The surest sign you need a process is if you find yourself doing the same thing over and over again. If you’re sitting there and a project or Slack or email hits your inbox and your first thought is “Again?!” that’s when you know you need to figure something else out.

I’m going to use the proposal process above as an example. As my business has grown over the past year+ I’ve realized that there will be certain tasks I need to do all the time. Sending proposals is one of them.

When I was getting started I didn’t have to send send that many so I would open up a Notion Doc type down a few thoughts and send it out. After sending a few of those out I realized I needed something a bit more polished—I’m helping with brand building after all! That led me to building a Canva template for proposals.

The Canva template was great. I was able to add my branding into the mix and build it in a way where I could adapt the template for coaching, consulting, and fractional work. The problem was it took a while to build. Partially because I’m using a design template instead of a doc (Canva, we can do better with Docs) and partially because of the actual content. Turns out the content was way more important than the design and much harder to figure out. This leads to the first lesson.

Don’t assume you know what your process is going to be

I can say honestly now that when I first started writing the proposal content it was a bit loose. I’d hop on a call or Zoom, take notes in my notebook, and then rewrite those notes to create the proposal. That worked for a while until I realized that scope creep would creep into the work.

It’s not that clients were overreaching but that we’d talk about things that maybe I forgot about or didn’t write down which we’d agree to verbally but not on paper. It forced me to rethink my approach to notetaking. I tested Zoom’s meeting companion—wasn’t for me. I then ran through a bunch of AI notetakers to see what worked for me and Fathom blew me away.

Now I had an AI partner that would provide notes and takeaways. I wasn’t ready to let go of my notebooks (still haven’t tbh) so I was taking my notes, comparing them against the AI, then rewatching video clips to get to where I needed to get to. The proposal notes became better but it. took. forever. to put together a proposal.

It was around here that I changed up my entire proposal structure to focus on problems solved and outcomes rather than hours or broad statements like “Update tech stack.” Now I thought about it like this:

*Key Issues:*
• Issue 1
• Issue 2
• Issue 3

*Proposed Solutions*
Issue 1
• Solution 1
• Solution 2
... and so on

This changed up everything. If I had built automations at this point I’d have spent a fair amount of time getting to this point only to more or less blow it up and have to start over.

The Automation Trifecta

I hit what I call “The Automation Trifecta” of needing to automate:

Repetition

My biggest beef with running to automation solutions is that in a lot of cases you don’t actually know what the work is. You start creating shortcuts and you don’t know where the road actually leads so when something goes wrong or the process inevitably evolves you’re at a complete loss.

Standardization

Once I had a proven formula that I was repeating consistently I knew exactly what I needed to do. There was less exploration and more copy/paste.

Complexity

Reviewing my notebook + AI notes + call recordings was exhausting and took forever. There had to be a better way to do this.

Time to automate

We hit the trifecta so let’s go to work. Before jumping into new tools you need to explore what you have.

My notes clearly weren’t cutting it anymore so it was time to get rid of them since I needed to manually type them up anyway. Fathom was really helpful though. I wanted to explore that even more. There’s this “Ask Fathom” feature with every call and it let you chat with an AI agent that would answer questions based on your calls. I came up with two questions:

1. What are [name]'s key issues?

2. What were ideas, solutions, or recommendations I offered for these issues?

These became the primary drivers of the proposal and Ask Fathom made getting the answers 10x faster. What Ask Fathom didn’t solve was getting things into the doc so it was time to think beyond what was available to me directly.

I wanted to automate the proposal creation but I also wanted it to happen automatically rather than when I had the time so I built this automation in Zapier.

You can use this Zap too

1. The trigger is a new transcript in Fathom. It’s generated automatically after a call and gets fed into the Zap.

2. I record all of my meetings but not all of my meetings require a proposal. This looks at the meeting title for the word “Intro” and if it meets the criteria it keeps going.

3. I can’t use Ask Fathom in Zapier so I had to rethink my approach and brought in ChatGPT instead. I feed it the transcript and ask it for the key issues.

4. Using context from step 3, I then ask for solutions I brought up.

5. I create a Google Doc with the content structured in a way that’s usable for me.

Since I’m still using Canva for my proposals and I can’t hook it up to the Zap, there’s a manual step at the end, but overall this automation now saves me a ton of time.

Think about what comes next

That manual Canva step is a good reminder that a good process is always evolving so as I build them I think about how I may improve them in the future. The two most obvious ways for me are

1. Add Canva - if there was a Canva action that allowed me to import this data directly I could skip the Google Doc which would save even more time.

2. Build a Proposal GPT - one of the things on my to do list is to build a GPT with all of my services and solutions. Since most companies I work with tend to have similar problems, the solutions are pretty similar too. There’s no reason to manually check everything (aside from due diligence). It should also be able to price things for me.

3. Auto email with pricing - if I have a foolproof GPT putting the proposal together then I could probably email it automatically. I’d probably start with a review step for myself but how cool would it be to just hang up and have a proposal ready to go in moments?

4. Enroll prospects in a workflow or sequence - automations on top of automations, right? You could add prospects to lists and follow up automatically too.

The thing about these is that I haven’t explored them yet and if I haven’t explored them then I don’t know that they’ll work or work in the way I want them to so I’m taking my time.

This is how I think about automations and it’s how I recommend you do it too. Don’t just start automating. Take the time to understand the process, where you stand to benefit, and only then start building. As you’re iterating in the future you’ll be grateful you took the time to do it.

I love building automations. It’s honestly one of the most fun parts of this job for me. In the past few months alone I’ve automated solutions for:

• B2B SaaS

• B2C Tech

• Event companies

• Agencies and Consultancies

If you’re struggling with any aspect of your process (or you need one), book some time with me. I'd bet we could come up with some cool solutions to whatever you’re facing.