
EPISODE SUMMARY In this episode of The ChatGPT Experiment, host Cary Weston brings back Greg Howe, founder of Gimme Info, for a deep dive into how small businesses and solopreneurs can use AI and automation to reclaim their time. Greg specializes in...
Loading summary
Amazon Music Ad
Alright, quick break. Knock knock. Who's there? Amazon Music. Amazon Music who? Amazon Music, where prime members can listen to top comedy podcasts. Ad free. No awkward interruptions, just non stop laughs from your favorite shows. It's free, it's funny, and hey, it's better than my knock knock jokes. Download the Amazon music app and let the good times roll. Or go to Amazon.com adfreecomedy that's Amazon.com adfreecomediDy to catch up on the latest episodes or without the ads. Finding great candidates to hire can be like, well, trying to find a needle in a haystack. Sure, you can post your job to some job board, but then all you can do is hope the right person comes along. Which is why you should try ZipRecruiter for free at ZipRecruiter.com Zip ZipRecruiter doesn't depend on candidates finding you, it finds them for you. Its powerful technology identifies people with the right experience and actively invites them to apply to your job. You get qualified candidates fast. So while other companies might deliver a lot of hay, ZipRecruiter finds you what you're looking for. The needle in the Haystack.
Kerry Weston
See why 4 out of 5 employers who post a job on ZipRecruiter get a quality candidate within the first day. The smartest way to hire. And right now you can try ZipRecruiter for free.
Amazon Music Ad
That's right, free and at ZipRecruiter.com zip. That's ZipRecruiter.com zip ZipRecruiter.com zip.
Greg Howe
Hey there.
Kerry Weston
And welcome to the ChatGPT experiment. This is the podcast designed to help you better understand what ChatGPT is and what it's capable of and how you can find a nugget or a way to make value of it in your personal or professional life. My name is Kerry Weston. I'm your host. Glad you're here. Got a good one for you today. Straight from the listeners survey. You told me that one of the key things that you want to spend more time learning about or hearing about is automation. How do I automate workflows? How do I connect things with and outside of ChatGPT so that my life can be a little bit more easy, more efficient, more productive. And so I'm welcoming back Greg Howe. Greg's the founder of Gimme Info. Greg's joined us before to talk about custom GPTs, but Greg has been spending some time literally doing what it is that you were asking for more information of, which is connecting the Dots, simplifying workflows, understanding what people need and then putting together some automations to make it happen. So we really great conversation with Greg today. I think you're going to find a ton of value and we talk about maybe doing a workshop together. So here's what I want you to do. If you are interested in learning when that workshop might be, listen to the episode. And if you're Interested on the ChatGPT experiment website, chatgptexperiment.com@the Footer, you'll see. Join the newsletter. Just sign up if you've downloaded a guide. By the way, you're already on it. So if you've downloaded already the five tips or the custom GPT guide, you're already in the mailing list. So no need to do that. But if you haven't downloaded any of the guides and you want to be on the list for either more information, articles and you know, I have people, I keep people updated on tips and writings and events and that kind of stuff. So if you want to join the mailing list on the footer of the website, click Join the mailing List and I will let everybody know when this workshop, this interactive workshop will be ready. It'll be a live thing. You can join in. We'll put the details together sometime in August. But yeah, really great conversation with Greg. I think you're going to find some terrific value and you know, on the miracle of automation. This episode is scheduled ahead of time. I'm on the road this week. I'm down in Washington D.C. and then down in Maryland visiting family, but produced this before I went, so, so this will go live while I'm on the road. But yeah, automation, workflow, how to make things happen in and outside of ChatGPT, including email by the way. And so he answers some questions on some really, really fantastic ways on how you can make sense of email automations and replies and logic and all that kind of stuff.
Greg Howe
Okay.
Kerry Weston
So I hope you find it valuable. Join the mailing list if you want to learn more about the upcoming workshop. And without further ado, here's my conversation with Greg Howe.
Greg Howe
Talk soon. Well, hey Greg, welcome to the show, my friend.
Hey, thanks Cary. Appreciate you having me on.
I appreciate you coming back. Last time we talked about some custom GPTs. But just for the sake of perspective and for the sake of folks understanding what they're going to be listening to, why don't you do a quick intro as to who you are, what you're doing and who you're doing it for? These days.
Yeah, sure. So again, my name is Greg Howe. I'm the founder of a company called Gimme Info. And one of the things that I've really been working hard on is making data accessible to people. And the current push on things is of course AI. AI is everywhere, whether it's a GPT or whatever. What I've been doing is helping small businesses and high end solopreneurs get AI into their business. A lot of times using AI automations to do the work and it's really about simplifying tasks and making things more effective for people. Not creating the next unicorn multimillion dollar exit strategy but really focusing on the day to day stuff that truly makes a difference, I believe truly makes a bigger difference for people on a day to day basis.
So cool. And that's why I asked you to come back on. I just finished a listener survey and had a few hundred folks contribute which is awesome because you get to hear from folks that listen to the show what they're interested in, what their goals are, what they're challenging with and what they'd like to hear more of. And I was able through some analysis to break down. It's about eight different categories I would say. And two of the categories are kind of similar and one of them being I had so many people say I want to help get help, time saving help, which is what you hear a lot. Right. Email and communication management, like how do I scan my inbox, how do I prioritize what's going on, how do I really just make best use of all that stuff that's coming at me on a daily basis? You know, people. I had one person say I wake up to 100 plus emails across various clients. How do I triage this? Right. So to your point, it's not about how to create a better system, it's really how do I survive and become more efficient in the systems that I have.
Yeah, absolutely.
And the second one that's connected is a lot of people said workflow integration and automation connecting things like Gmail and Forms and Slack and their CRM and all that kind of stuff. Right. And so I know that's a space that you're living in. Yeah. So I figured if we could walk through at an eighth grade level because that's basically where I feel I'm at sometimes, I think that would be helpful to me and the folks.
Yeah. You'll be able to keep up with me, Kerry, for sure.
Yeah, I appreciate that pretty much. Well, let's start with, let's start with the work that you're doing just for context, because you said you're trying to work with people to make their life easier. What's a for instance? Like what, what's a common for instance that folks are doing just, just to kind of give context where we're going to go?
I mean, it's, it's funny because it's a lot of the same things that you just said. It's the same things I'm hearing whether, like, I had a, a client talk to me from, like, Paris, France. So it's like, it's not like a particular industry or anything. A lot of these things are common. Right. So the for instance that I always use is you have things that are ROI driving. There are things that are time savings and there are things that are productivity enhancements. And people are around the same things down to, you know, email or. One of the big things that I've been working on lately is sentiment, like being able to adjust a reaction to something that has a sentiment to it. So you receive a customer service email, if the person's really happy, maybe you want to contact you. Put something like a slack message out and have somebody contact them. If they're really upset, maybe you want to send it to a customer service manager so they can triage it. I think a big thing that I'm seeing right now is how do I, like you said, how do I do the daily overwhelm of stuff but also not lose the human connection? So how can I, how can I integrate AI while maintaining the places where human connectivity is really necessary? And that's something that's hard for people to think about. So I actually spend most of my time strategizing rather than building something. Because once you get the strategy down, you're like, oh, I see how I can do that. So that's a lot of it right now.
Okay. So I came in before we started talking. I was of the mindset that the AI tools within these certain platforms, like Gmail and Microsoft's suite, Right. Have their inherent AI, which is probably, in my mind, it was the preferred place in which you would use AI to manage your inbox. But you said that you had a different perspective, and I appreciate that because you're spending time here. So let's start with email, because that tends to be. That's the most popular one here. So let's start with, well, you know, I wake up to 100 emails across clients and let's, let's filter in what you just said. There's some sentiment things here. So Open my eyes as to what's actually possible and how hard is it to utilize an AI tool with my inbox to make some sense of it.
Sure. So there's two parts that I want to, I want to mention. The first one is your take on it is actually really quite good. There are some really good tools inherent in, you know, the Google Workspace or the. Whatever it might be, and they're getting better on a daily basis. Like, for my Gmail stuff, the summarization that it's giving me on a whole conversation thread is fantastic because it only shows me, which I think is fantastic. The latest, you know, like, if we're back and forth with a customer and you're having all these details, it's tough to where was that note? You know? But it's like this thing summarizes it and only gives you the latest update. So really nice, helpful thing. But it doesn't handle the. When you have 100 emails. So what I want you to look at with that is there are tools that you can use. Like, and this is one of the things that I'm sure we'll talk more about, but there are workflow managers, there are tools like zapier, make.com and n8n.IO.
Say that last one, n8n.
It's the letter n, the number 8 and the letter n as in Nancy. So n8n. So it's a big thing right now. If you're, if you're, you know, in that, you know, on LinkedIn and you're seeing a lot of people talking about workflows, they're talking about N8N. But I would, I would, for this audience that we're talking about, put that one aside. It's too, it's too detailed, too technical. It's really made for developers and workflow. But there's tools like zapier. Yeah, the N8N1, it's so powerful. But it's one of those things where it's, it's like what you'd expect. It's, you know, where there's more power, there's more complexity.
Yeah.
So what I think people need to do is really simplify and look at the simple tools like a Zapier or Make. And I use make personally because I find it has flexibility.
And that's make.com maybe.
Yeah, make.com.
Okay.
And what it has is integrations with lots of different softwares the same way Zapier does. And it allows you to then connect them together where it, where it benefits more than, say, Zapier. Is it allows you to kind of branch your flows. You could say, like for example, with sentiment. If I'm getting email, 100 emails and I want to be able to. And this is, this is one that I use personally. I get hundreds of emails every day, some of them, most of them trying to sell me something. But what's always tucked into those ones that people are trying to sell me something is that gold nugget of information that somebody found and they're simply giving that value back. And so I'm like, I don't want to read this person's 18 emails to find the one that has value. So what I do is use make to read. I have Gmail say, send every email from this person into this tag, which puts it into a folder, right? Then you use make to say, anytime something hits that folder, look at it. Use AI to look through it, get a feel for it. And I literally say, if it's trying to sell me something, put it in a different folder. If it's trying to give me value, put it in a third folder.
So I have one inside Gmail itself.
Inside Gmail itself. So I have three different folders. One called Research, one called Research, two which is the real value, and one called Selling. So everything goes into research. Like I've got to have 50 to 100 different emails that come in and get dropped into this one folder.
Okay.
My AI then goes and looks at it, parses through it, and separates it out into real Research or selling me something.
And then I can choose for a second. Yeah, that's great. So let's just pause for a second because I want to make sure that I'm crystal clear and that we kind of don't overlook anything. So inside Gmail itself, which I've done before, you can create your own rules. You can create folders and say if this person sends me a mail, send it to this folder. That kind of thing. That's pretty straightforward. But what you're saying is that's all inside Gmail. So you've identified, in other words, you've identified the people that you're putting into that folder to begin with.
Right.
So you've said you've got 50 different people or senders or whatever that are going there now. Are you ever deleting them and saying, stop this, I don't want them, I don't want their emails anymore.
Oh, absolutely. Yeah.
This is an ever changing kind of. You're managing that offline.
Okay, yeah, I can bring that up and mention that, that it's a. It's part of a bigger process. These, these workflow automations are really, you know, one part of a, a bigger process of how you want to manage. Okay, emails in or out.
Yeah. And then you have, through make.com, you said you have AI now. What, what would be AI in this case? You said you AI read it.
Yeah. So literally it has a Gmail module, it reads the email and then I have next to it an OpenAI module. So it uses chat GPT stuff.
Okay.
You bring the text of that body in and then you say, you give it a prompt that says, here's how I'm going to determine what's what.
Okay.
And then it literally, it literally just comes back with research or selling.
Okay. And I asked, I asked that question because I, people use AI as a product sometimes. And so I just want to make sure. And so you're literally using chat GPTs a lot. You're using Chat GPT on one side, you're using Gmail on the other, and the, the bridge, if you will, is make.com.
Yeah.
And so you're saying make, go over here, do this, come over here. It does chat GPT stuff and then it goes back and put stuff into the. Okay, that's awesome. So let's say, let's say we're not doing the selling and research thing. Let's say we have employees or clients. Right. And so we've got a client folder. And then that could come in and you could say, is this urgent? Like, you could put rules, I would imagine. Does this appear to be urgent? Does this appear to be time sensitive? Is there a question? Right. If so, move it over here. If not, go over here. Now can I go one step further? The whole goal of this is to get out of my inbox or to be more efficient in my inbox. Could we go so far, in your experience, could we go so far as to alert me another way if an email comes in and I'm not looking at it and this chatgpt through make is determined, this is important or this is time sensitive, could it send me an alert in some other way?
Absolutely. So this, that's why these, these tools like make and, and whatnot are so critical in the next thing for AI. AI is just one part of it.
Right?
So AI is going to determine sentiment, like you're saying, like what's important, what's critical, what needs to be acted upon. So you have the input which you know is, is Gmail, and then you have AI doing its part, whatever you do after that you can pick any thousand things and, and you can do them in any combination you want. So it's like if you want a slack message, if you want an email to go like I get my, I send emails back to myself. A good example is when I get a form submitted from somebody who's interested in my stuff. I look at it, take the whole thing and AI looks at it and looks at the whole combination of the answers to the questions, makes a determination about whether I should be contacting them immediately or, or not. And if it is, it sends me an email that has the summary and you know, information about this person that basically says, hey, you know, that way you don't have to look at every form request you, you can later on in the day. But like if you want to act on something, find what works for you. It might be an SMS message, it might be, you know, it can be any number of.
So you could literally take, and this is eye opening for me. So you could literally take an email, have ChatGPT read it, evaluate it and say, holy smokes, this is important. I'm going to text Greg and say, Greg, you might want to look at this right now. Yeah, and then that would be real. Like that's a, that is a now the things that you're talking about right now seems super programmatic. It seems like you got to have some experience. Is this an easy interface, this make.com?
Yes, it's. So first of all, make is a workflow for tools that you already have or need to purchase. So Gmail you have. If you're going to do text messaging, then you need something that does text messaging. You need to have Twilio or whatever it might be. This make.com is the connector of all these things in a strategic way for you, a tactical way for you, the.
General contractor of sort, you're still going to have your.
Exactly. So but it's not programming, it's like there's, there's connectivity set up, but those are the things that you're going to know anyway. Username, password, there are some things in there, but they're just property settings.
Okay. So you have to know there's a lot of drag and drop fillings, that kind of stuff.
Okay. Yeah. And even, even when it starts to get a little bit more complex, it is drag and drop. You. So where you're going to spend your time is an understanding what make can do for you in a broad general sense. And then when you get to something like where you know, if it's this kind of condition, I want to do something. If it's a different type of condition, I want to do something else. You simply have to learn how make does that. So it's not really programming and it's not. You don't have to be overly technical but you do have to think it through. You have to think what would I do here? And then make will do that for you and you just have to give it the rules. You have to say here's my login, here's my what I want for a subject line on that email. Here's what I want for the body and fill in the blanks here. So it's almost like when you used a template out word. You know you could, you could send a bunch of word documents that are all just pre filled. You have this pre filled template, same thing, you just pre fill your okay.
Whether it's a slide, a lot of us have a lot of folks listening have customer relationship management tool what we call a CRM. It could be HubSpot. Some people use Salesforce. These are things and they typically have automations and auto drips or auto flow. So if a forum comes in this happens. This kind of seems like this is on steroids. Where the if then in the CRM typically it's I'm going to send this email or do this thing. It's very limited. But it's automated, right? Quote unquote. It's automated and it sounds like the same logic but if you have access to different tools this make.com can go and be an assistant and do all of these things on an if then basis, right? You're saying there's all kinds of if thens that's happening here that will I.
Think is important though you bring up a good point. Like say for example you have HubSpot and you've got an automation in there. HubSpot also gives you forms that you could put onto a website. And so you have this sort of. I have this whole environment in HubSpot and I have people that go to my website, they fill out the form. It's all within HubSpot. Everything's great. But as soon as you want to do something else like drop a slack message. Yeah, what are you going to do? You got to hire a HubSpot consultant or whatever. It's like what's the beautiful thing about make is is that it doesn't take away from the value that you have in HubSpot already written. You can enhance it by saying maybe I have this other form over here that's not tied to HubSpot, but I'm going to siphon the data over, drop it into the same HubSpot form that's used on the website. So now you're just dropping it into HubSpot through the same form mechanism that's always worked for you. So all the workflows, all the form entry, all, everything can then happen that way. But that's what I mean. You have to think about it. Do I want it to go through the same form? Do I want it to be a separate process? Do I want different notifications? And so you just allow the thought to percolate about what makes sense for your business as far as where the data is coming from and going to. Don't lose the investment that you've already put in. You can literally enhance it or expand on what you have.
Okay. And I think this is where we need to pause for a second because one of the bullets that I got in the analysis was automating repetitive admin workflows. There's a lot of just rinse and repeat kind of things that people do in the office. And they either hire an assistant or they do it. You know, they delegate it and. But you're saying that there's an ability if we just take a step back, much like onboarding a new employee, Although we do. I've seen this happen. We wouldn't hire an employee, put them at a desk and say, good luck. Right. I hope you figure it out. That has happened. I know it has happened. And that's where we get into some dissatisfaction at the workplace. Right. But let's say you were going to hire a new employee and you put them down at the desk. You're going to spend a couple weeks just getting them to understand the basic workflow of what happened happens in the office. And so what you're saying is, let's commit a fraction of that time maybe to teaching a tool like make.com what's happening so you can understand your own workflow. And that's. And I'm going to pause right there too, because a lot of people don't take time to analyze their own workflow. They just do it instinctually. They're just busy, you know, so there is no master diagram. And that's because we just do it. Right. But if I think what I'm hearing you say is if we could pause for a second and create our own kind of master diagram, like what's really happening so we can understand the steps, that information could be really valuable when we bring in a tool like make.com to say, maybe there's a way of creating that pattern in make.com.
Yeah, let me blow your mind here, Kerry, for a second. Let's really look at this. This is something that I think is very important that people need to understand because I know half the audience just went, oh God. So here's one of the things that I use that has been extremely beneficial. So everybody's ears perk up a little bit on this because you can use chat GPT, you just brain dump. Just here's what I do. Blah, blah, blah, blah, blah. All the ums and odds and I'm not sure. And wait a minute, let me change my mind on that. Just chuck it all in and then say, create me an SOP for this. Enable.
Hold on. We have an acronym rule on the show. What's sop?
Okay, sorry, Standard Operating Procedure. The rules. The rules of how you do something. The step by step. And that can be done any number of different ways. Or you know, you can do loom videos. There's a product called Loom that lets you record screens and you can add additional text to say, this is what I actually do. So don't let the, the blah of an, of an SOP or instructions hold you back from actually producing that for onboarding. You need that for the onboarding. Just do the work and let a loom video show somebody. What you actually.
What would you. You just introduced the loom video and I know there's Vidyard and other ways of recording your screen, but what you introduced that so kind of nonchalantly. What, what would I do with a loom video while we're doing this?
So like for example, one of the things, let's just say you got a busy CEO, somebody who says, I just need somebody to help me do this work. I do this. It takes me, you know, half hour a day. Drives me crazy. Basically, instead of trying to explain what you do, you do it, go to your screen and start actually logging into the different tools. Go to your different web pages.
You're recording it while you're doing this is what you're saying. With a Mac, you could actually record the screen. Right? You're just talking about a screen capture.
It's a screen capture video. Yeah, there's scribe.com, that's another good one. There's several of them that will do it for you. And each of them have pros and cons. It can show button clicks and things like that, which I think is great. So it's doing like it's creating a PDF for you of step by step. Go over here, mouse over there.
Okay, so. Okay, so that. Let's. All right, I'm going to pause again because I want to make sure that you talked about workflow, we talked about zapier and make, and then you had the developer heavy, which is the N8N. We're now talking about creating the standard operating procedure, the instructions, the rules. And we want to do it on a computer and we want to log in and do these certain things, but we want to record it as we're going. And so what tools. Let's, let's stop right there. What tools would you recommend that are capable of doing this kind of thing if we were to record?
Yeah, I mean, there's, there's a scribe. Scribe. Loom. Loom is nice because it's video. You actually seeing the screen. Scribe is nice. If you're trying to create a static document of screenshots, it knows when you. It just, you just still use it, but it's, it's taking screenshots here and there and then supplementing it with information that it's hearing. You say there are lots of, of.
Tools out there I don't use, but those are two. What would you call. If I was gonna, if I was going to search for one, what would you call it? What is the category called process recording software?
Yeah, screen recording would be that. That will. That might do it. Yeah, it's process recording.
So screen and process recording software or tools or something.
Yeah, something like that.
Okay.
Okay, we'll do it. But if you, if you get a list and you start seeing things like scribe and loom, then you know you're.
In the right place. Okay, so let's say I want to a new customer, you know, new customer inquiry or. And whatnot. And I'm gonna, I'm gonna go through the fact of using three or four pieces of software, whether it's setting up a new invoice or whether it's setting up a meeting or whether it's, whatever it might be sending an email. You said the busy CEO is going to record. So let's say we've picked something like loom and we're. Now we're recording and we're opening our windows and we're logging and doing our things or what. What's happening now.
So. And again, this has nothing to do with AI, but this is one of those things that it's important pre work that you need before you start chucking AI at a lot of different things. These are the things that are process enhancement. So the CEO is going to dump all this stuff and then it's a matter of you can give that to somebody to say, you know, say you get a new employee and you say you're going to help the CEO, you're going to do this process for them. They'll be able to follow it through and then be able to say, I don't understand this and ask that one question. Okay, Give that new employee the responsibility of updating the document and saying, okay, here's what actually is happening or here's, here's an additional step. We'll remove the step, that type of thing. So they basically, the CEO encompasses the knowledge so that somebody else can maintain the real process and that now drops that process out of the CEO's hands and delegates it effectively to somebody below them where it should actually get done.
So this is homework. This would be background in homework. AI is not involved. This is. I just really need to understand what is that you do. What problem we're trying to solve, what process are we trying to enhance? And rather than asking somebody what do you do next? You're just saying, because we don't really think of it, we just do it sometimes. Right, let's record it once, give it to somebody so they can watch it and analyze it. They may have a few questions, but it's a lot easier that way than asking all the questions all the time.
Right, exactly. So we're just. Sorry, I want to mention this part because I think it's important is that that's not sort of off to the side from the AI workflow stuff we've been talking about. It's more foundational because what's going to happen is you're going to take the knowledge and, and be able to, as we were talking about before, you can just dump it into chat GPT and say, generate me an sop. So if you don't even use LOOM or any of the screen recording stuff and you just want somebody to know what the process is, then you can put it into Chat GPT and say, make sense of this for, for the.
Next person you could not take. I just want to make sure we're clear on this one. You couldn't take a LOOM video and give it to ChatGPT and say, figure this out. There's that technology really isn't there yet.
I think there is. I think you can natively though, you'd have to dump it out to like you know, a video, an MP4 video or something and upload it. But ChatGPT does a great job with doing transcripts now on videos. So you can drop in, you can.
Upload a video to ChatGPT and have it figure it out. Okay, so if there were. No, let me go a step further because I guess you're assuming at that point that there's narration happening while this is like if it was a silent video.
Right.
You'd really, you couldn't, there's really.
No, Yeah, I wouldn't trust that at that point.
Yeah, exactly. Right. So, okay, so foundational work. We're setting up the workflow, understand the workflow, someone's going to watch this, document it, and then once we have that documented, you're saying you can, you can give that documentation to the next step, which would be chatgpt perhaps.
Yeah, or like I said, you know, I, I use it to help me strategize and create documents. I had it literally the other day. I, I spent a few hours literally just coming up with what I'd want to do with this AI automation stuff for my business. And I, it treated, I treated it like a strategy kind of advisor. At the end of this huge discussion, several hours worth of just hemming and hawing and throwing notes out and every once in a while throwing something in, I said, can you please summarize this entire thing for me? It created. And I thought I would get like a one pager. It created a 13 page document with action items, what the next quarter looks like, My, my next steps. And it literally carried me through to January 2026 when it said at that point it gets a little vague like, oh my goodness, you know, I mean, those are the types of things that people need to take advantage of. And it has nothing to do with the workflow, but it's like use Chat GPT to provide something that then can be usable. And those are the places where workflow can come in and take a look at that 13 page document. Now feed it back in and let workflow say, send Greg an email that summarizes what he has to do today. Yeah, so it gets kind of meta on you, but it's basically one of those things where.
Yeah. So you know what's going through my mind right now is eighth grade.
Yes.
Trying to remember her name. But anyway, the teacher had us. I didn't know we were doing technical writing. I had no idea what technical writing was in eighth grade. But we had to do the create step by step guide for making a peanut butter and jelly sandwich so you could pass the notes off. To somebody else so they could follow your instructions step by step, right?
Yeah.
And so you forget to take the jar to open the jar, right? So you got people stabbing a knife into a closed jar of peanut butter because you didn't give them that step. And what you're doing, what you're talking about now is really breaking something down that we might do instinctively over and over again without thought. We're just kind of doing it. We're breaking it down into a technical pattern so that someone else could replicate. And in this sense, someone else might be a tool like make.comchatgpt and these other tools. That's really where we're going with this, right?
Yeah. So that's a great example. Let me throw this to you. This is the way I'm picturing it. I think this will help solidify it. With a real example like that. If we were to take the peanut butter, you know, open a jar, peanut butter thing, we will miss things because we're so used to our day to day. The first step is get whatever you have in your head out. Then you feed it into chat GPT and say, where are the gaps? Because it has all that history behind it of millions of people that have opened jars of peanut butter. And you say, what am I missing? That was the thing that gave me 13 pages of information, was it was throwing things at me like you can say to it, where, what are my blind spots here? You know, if you want to take a product idea, what are my blind spots? So it can feed that information. You just keep revising it and feeding it back in saying, now what now.
I love that part. So you're going to appreciate this because I know, you know, in a previous life, you and I worked in a programming company together. And, and there's a difference between how normal people think and how programmers think, right? And it's, that's a true statement. That's a true statement backed by science and data. And so I am working with somebody who is working with or wants to work with a programmer. So I'm the middleman because I can speak just enough programming to get dangerous and I can speak English and like a translator, right? And so I had a conversation where I'm listening to what he wants and I'm talking to the programmer. So I get to figure out how they think. So I use ChatGPT and I said, I need to give them the specs. I need to give the programmer the specs so they know what to build. Interview me. Okay? So it did. It interviewed me. Well, the brilliant part's next. The part you're going to appreciate is now, I won't call it brilliant. The part you're going to appreciate is next. So it interviewed me and it got the English out of it. Okay. And so it said, great, here's what I think you're building. Here's all the function features. Do you want to send it? And I said, no, not yet. I said, now I want you to think like a, Are you familiar with the disk profile?
Yeah.
So high C, right. I need all the information, need all the details before I do it. So I want you to think like a high C programmer that is really in the weeds. I want you to take on the perspective of that person. Now look at my spec. What questions have I missed? What details have I missed? What, what questions you're going to have for me and ask me that. And it was, and it asked me stuff. I was like, holy moly, like, I didn't even think of that.
That's it. Exactly.
And so we did that and we went through ChatGPT, played that role, and it filled in those gaps. And I did, I ended up with, you know, maybe an eight page spec of features, questions, considerations, you know, all that kind of stuff, and sent it off. And, you know, it's been a week and I haven't heard back from the programmer yet. So I'm wondering if I did A, a great job or B, I kind of, you know, scared him away. I'm not quite sure where we're going with that one. But the ability to fill in the gaps, I think is so key and so important. And, you know, it's also important to understand that ChatGPT out of the box is wonderful. But ChatGPT, when you've spent some time building some of these foundational lists and processes and attribute lists is amazing, you know, and that's really where the differentiator comes, when you can understand what it needs to understand. And so your, your explanation is great. I've never thought of recording like on a scribe or a loom, so I've got all kinds of ideas going through my head. So thank you for that.
Let me bridge this for you one more, one more notch up, still keeping it at an eighth grade level, because here's, here's the thing. So what we've talked about is build foundation, understand your processes, get them in a place like you said, create. Like, if anyone is familiar with projects, if you, if you have the paid version, you can do projects and they're really, really useful. Create a Project for your business within Chat GPT and start feeding it your processes, feeding it how you think, feed it your voicing on emails, give it the data that it needs. Now you can go and from make, you're going to connect to Chat GPT. So all of this meat that you've put into Chat GPT you can hit through make. So if you want to have make, let's hear a real example. You get an email, it's, it's a return of a product. Somebody's upset, but you don't necessarily need to contact them to us through a slack message through somebody you know calling them. You can generate, let's say you have a template for handling this sort of response, but you want to respond immediately. You can have AI generate the text based on what you put into Chat GPT for your voicing your emails you've done before, how you, you know, how you handle certain situations, whatever the case might be. MAKE has access to the whole Chat GPT world that you've been building. And so now make can say, oh, I just found an email that somebody is complaining, they want to return a product, let's get them back and let's get back in touch with them directly. And you can actually make things like imagine this kind of rule where you say, if it's this particular product that we've had problems with, let's just say, and it's, someone's really upset, let's give them a 50 off coupon on their next order. Whatever you can, AI can make this determination on the fly, pop it into an email, send the email on your behalf, done. You're never even involved in the middle of it. And it's, and it's trying to do things appropriately. Now you have, that does take a little bit of extra effort on the programming side. It's making sure.
Yeah.
You know, making sure it's working for you. But those types of things like, that's what, that's what I'm trying to get to the, the whole goal of what we've been talking about is with a solid foundation, your processes getting documented, become input back into an AI type of system that says, this is how I work. And so now AI can go, oh, I can do that, I can help you with that.
And then you can even, you've got my idea treadmill spinning here a little bit. So let's go back, let's, let's go full circle. Let's talk about the Gmail folders. Yeah, okay. And so you've got a, you've done Your work and you've created, you know, the project, as you said. And here's some examples of emails. Here's some if thens, here's how I handle these basic customer inquiries, complaints, questions, whatever it might be. You could have templates set up, is what you're saying. If it's clear that this is the question or this is the concern, or this is the scenario, you know, it could come in and create this email. And even an email could even say, you know, I want to get back to you quickly. You know, this is an automated message. I want to get back to you quickly. But based on the question that you've asked, here's the three most likely resources or answers that I would give you. Right. And then that email could go into a folder.
Yep.
Right. Need to review folder kind of a thing. So make could say, hey, I sent these 10 emails because there's people nervous right now that they're speaking on your behalf without you knowing about it. Oh, absolutely. You could send a safety email that say, hey, in the interest of getting back to you quickly, you know, this is an automated response. I've understood your question to be this, your issue to be this. Here's three answers. Resources, links, articles, whatever that will help solve your problem right now. And I will take a look at your email and if there's additional detail, I'll get back to you. And then that email could go into a folder you're saying that could be like, to be reviewed, Something like that.
Yep. Or when you want to drop it into a HubSpot event or task, you just basically can say, hey, here's a ticket.
Wow. Okay, so you could, you could come in twice a day and there could be a folder there or a task or whatever that says, hey, you've got 10 things that have already happened without you knowing you want to review them here. Right. And then you could say eight of them are good. Okay, this one, I want to add a little different context. But as long as you've kind of set it up the right way, you're kind of being in two places at once, solving problems, instantly, providing value. And then you get to verify trust.
But verify exactly that Right there is where I want people to get. That's what I'm out here for. I want to simplify all of this stuff to kind of look at it from a strategic standpoint and say, here's what you could do with these emails. Because think about what that CEO or whoever it is, they've now been freed up. They're not watching Emails, they're not typing out new replies, they're not, you know, and the system is just flowing and you're doing more of a review kind of concept.
Correct.
The rest of your day is on high value type work. Things that you can actually make a difference in expanding your business or whatever it might be. You don't have to deal with that drudgery type work. Even though it's important work, it is still like monotonous. Okay, I got to do another email. Okay, I got to send this thing out. Okay, I got to look at this. It's like you don't have to.
So I think that's a great place to pause and end today because I think you've opened my eyes. I know you've opened listeners eyes to a number of things here. And the software you kept referring to was make.com and so I know that you've actually, you shared with me the other day, go take a look at it and haven't been able to log in yet, but, but I'm looking forward to it.
Yeah.
If people want to find out more about you, I, I'm, I'm sensing there's going to be some feedback on this episode. So we may even want to do maybe a workshop.
Absolutely, that'd be the best.
Okay, great. We can kind of demonstrate this a little bit. But if people want to find out more about you and what you're doing, where do they find you?
So actually I'm, I have a, a free school community called AI Automation Launchpad and it's all this stuff we just talked about. Okay. It's just been launched. It's not completely full of all my content yet, but it does have three templates. I basically do something through a product called Voice Flow which does chatbots and phone and whatnot. So there's that. Okay, that space. It also has a free template for make.com to handle the flow of data coming from those chatbots and it does the sentiment that we talked about and instead of dropping it into HubSpot or something like that, because I don't know what everybody has, I actually drop it into a Google sheet. So you can use a Google sheet as a little mini CRM and even on these like little one off type things, it's like just drop it into someplace and then you can review it. And so that's the actual thing is that when it does the drop into the spreadsheet, if someone has a very positive response or a very negative response, it sends an email as well.
Where can people find out about you and what you're doing in that regard.
In my stuff, Basically, I'm on LinkedIn. LinkedIn is a big thing. If you just look up Greg Howe on LinkedIn or LinkedIn, slash in slash, gimme info, G, I, M, M, E, E info, you'll find me out there. And I'm always open to dm, throw me an idea. Hey, I'm thinking about this. Can I do something like that? Totally open to stuff like that. Love to help people kind of get their ideas going, like helping Carrie, like he said, because I want to set him free. I want to, you know, I want to set people free to be able to build bigger things.
Yeah, yeah. I'll put those links in the show notes as well. And would you be up for it if we did kind of a listener workshop where we kind of show some things? All right, cool. Let's schedule that in the future.
Yeah, let's do it.
Kerry Weston
Good.
Greg Howe
Greg, thanks for joining. I know this is terrifically helpful. You've equally excited and confused the listeners, just like. I hope so because. Because you get to the point where like, oh, my God, look at all the things that are possible. So I gotta go figure this out.
But I really appreciate that's, like, the littlest little piece to where my head goes with all this stuff.
Well, I appreciate you keeping it at eighth grade level, too, by the way, because that's about two grades ahead of where I am on most times. So I appreciate it very much. Yeah, I'm happy. Thanks for joining us. Yeah. We'll have you back as well. And let's circle back on that workshop, too.
Yeah, thanks a lot for having me, I hope. And honestly, people, keep just getting your hands dirty. It'll. It'll pay off.
Very good. Stay curious. All right, we'll talk soon. Greg, thanks.
Thanks.
This episode is brought to you by Progressive Insurance. You chose to hit play on this podcast today. Smart choice. Make another smart choice with Auto Quote Explorer to compare rates from multiple car insurance companies all at once. Try it@progressive.com, progressive Casualty Insurance Company and affiliates. Not available in all states or situations. Prices vary based on how you buy.
Amazon Music Ad
Marketing is hard, but I'll tell you a little secret. It doesn't have to be. Let me point something out. You're listening to a podcast right now, and it's great. You love the host. You seek it out and download it. You listen to it while driving, working out, cooking, even going to the bathroom. Podcasts are a pretty close companion. And this is a podcast ad. Did I get your attention. You can reach great listeners like yourself with podcast advertising from Libsyn Ads. Choose from hundreds of top podcasts offering host endorsements or run a pre produced ad like this one across thousands of shows. To reach your target audience in their favorite podcasts with Libsyn Ads, go to Libsyn ads.com that's L I B S Y N ads.com today.
The ChatGPT Experiment - Ep 78: Automating AI Workflows With Greg Howe
Release Date: July 22, 2025
Introduction
In Episode 78 of "The ChatGPT Experiment - Simplifying Chat GPT For Curious Beginners," host Cary Weston delves into the realm of automating AI workflows with special guest Greg Howe, founder of Gimme Info. This episode, titled "Automating AI Workflows," explores how integrating AI tools like ChatGPT with workflow automation platforms can significantly enhance productivity and streamline daily operations for individuals and businesses alike.
Listener Insights: The Demand for Automation
Cary opens the episode by highlighting feedback from a recent listener survey, emphasizing that many subscribers are eager to learn about automating workflows. Listeners expressed a strong interest in understanding how to connect various tools with ChatGPT to make their lives more efficient and productive. This sets the stage for an in-depth conversation with Greg Howe, who specializes in helping small businesses and solopreneurs harness the power of AI for everyday tasks.
Guest Introduction: Greg Howe and Gimme Info
[04:35] Cary Weston: "I'm welcoming back Greg Howe. Greg's the founder of Gimme Info. Greg has been spending some time literally doing what it is that you were asking for more information on – connecting the dots, simplifying workflows, understanding what people need, and then putting together some automations to make it happen."
Greg introduces himself and his mission:
[04:44] Greg Howe: "I'm the founder of Gimme Info. One of the things I've been working hard on is making data accessible to people. I'm helping small businesses and high-end solopreneurs get AI into their operations, focusing on simplifying tasks and increasing effectiveness without aiming for a unicorn exit strategy."
Automating Email Management
A significant portion of the discussion centers around managing overwhelming email influx using AI. Greg shares his personal strategy for handling hundreds of daily emails:
[12:43] Greg Howe: "I have one inside Gmail itself. So I have three different folders: Research, Real Value, and Selling. All incoming emails are initially tagged and placed into the Research folder. Then, using Make.com, AI analyzes each email to categorize it into Research or Selling."
Cary probes deeper into the mechanics:
[13:03] Cary Weston: "You've identified the people that you're putting into that folder to begin with. Are you ever deleting them and saying, stop this, I don't want them any more?"
[13:38] Greg Howe: "Absolutely. It's an ever-changing process managed offline."
This method allows Greg to filter valuable information from promotional content, ensuring that only pertinent emails reach his primary folders.
Tools for Workflow Automation
Greg emphasizes the importance of choosing the right automation tools based on technical proficiency. He contrasts developer-centric platforms like n8n.io with more user-friendly options:
[10:54] Greg Howe: "Tools like Zapier or Make.com are ideal for most users. They offer drag-and-drop interfaces that simplify the connection between different software without requiring extensive technical knowledge."
He personally prefers Make.com for its flexibility and robust integration capabilities:
[11:27] Greg Howe: "Make.com allows for branching workflows, enabling more complex automation scenarios compared to Zapier."
Documenting and Analyzing Workflows
Understanding and documenting existing workflows is crucial before implementing AI automations. Greg suggests using screen recording tools to create Standard Operating Procedures (SOPs):
[22:44] Greg Howe: "Use ChatGPT to brain dump your processes and then create an SOP. Tools like Loom or Scribe can help record and document these workflows visually."
He shares a relatable example from his own experience:
[31:19] Greg Howe: "Think back to creating a step-by-step guide in school, like making a peanut butter and jelly sandwich. Documenting workflows ensures no critical steps are missed, which is essential for effective automation."
Creating SOPs with AI
Greg illustrates how AI can assist in refining and expanding SOPs:
[32:08] Greg Howe: "By feeding your documented processes into ChatGPT, you can identify gaps and enhance your instructions. This leads to more comprehensive and actionable SOPs."
He recounts an experience where ChatGPT helped him develop a detailed specification for a programmer:
[34:02] Greg Howe: "I used ChatGPT to interview myself and generate a comprehensive eight-page document outlining all necessary features and questions for a programmer. This process ensured nothing was overlooked."
Enhancing Business Processes with AI
The conversation shifts to practical applications of AI in business operations. Greg provides a concrete example of automating customer service responses:
[37:48] Greg Howe: "With Make.com and ChatGPT, you can set up workflows that automatically categorize incoming emails based on sentiment and generate appropriate responses. For instance, if a customer is upset, the system can send a tailored apology and escalate the issue to a manager."
Cary elaborates on the benefits:
[40:30] Greg Howe: "This automation frees up time for high-value tasks, allowing business owners to focus on expanding their operations rather than getting bogged down by monotonous tasks."
Greg Howe’s Offerings and Community
Towards the end of the episode, Greg introduces his AI Automation Launchpad, a free community designed to help individuals implement AI automations in their businesses. He highlights available resources, including templates for Make.com and integrations with chatbots:
[41:22] Greg Howe: "Our community offers free templates and guides to help you get started with AI automations. You can find me on LinkedIn by searching 'Greg Howe Gimme Info,' and I'm always open to collaboration and ideas."
He also hints at future interactive workshops:
[43:11] Greg Howe: "We're planning to conduct workshops where we can demonstrate these automation techniques in real-time, providing hands-on learning opportunities for our members."
Conclusion and Future Plans
Cary wraps up the episode by acknowledging the transformative potential of automating AI workflows:
[43:39] Cary Weston: "Greg, thanks for joining. You've opened both my eyes and the listeners' to a number of new possibilities. We're looking forward to scheduling that workshop and continuing this valuable conversation."
Greg echoes the sentiment, encouraging listeners to explore and experiment with AI tools:
[43:53] Greg Howe: "Keep getting your hands dirty. It'll pay off. Stay curious."
Key Takeaways
Workflow Automation Tools: Platforms like Make.com and Zapier offer user-friendly interfaces for connecting various software applications, enabling complex automations without deep technical expertise.
Email Management: Utilizing AI to categorize and manage emails based on sentiment can drastically reduce inbox clutter and ensure timely and appropriate responses to important messages.
Documenting Processes: Tools like Loom and Scribe are invaluable for creating comprehensive SOPs, which serve as the foundation for effective automation.
AI-Assisted Documentation: Leveraging ChatGPT to refine and expand process documentation ensures thoroughness and minimizes the risk of overlooking critical steps.
Community and Learning: Engaging with communities like AI Automation Launchpad and participating in workshops can accelerate the learning curve and provide practical, hands-on experience with AI tools.
Focus on High-Value Tasks: By automating repetitive and administrative tasks, individuals and businesses can redirect their efforts towards more strategic and impactful activities.
Notable Quotes
Greg Howe [04:59]: "AI is everywhere, whether it's a GPT or whatever. What I've been doing is helping small businesses and high-end solopreneurs get AI into their business."
Greg Howe [12:43]: "I have three different folders: Research, Real Value, and Selling. All incoming emails are initially tagged and placed into the Research folder."
Greg Howe [22:44]: "Use ChatGPT to brain dump your processes and then create an SOP."
Greg Howe [37:48]: "With Make.com and ChatGPT, you can set up workflows that automatically categorize incoming emails based on sentiment and generate appropriate responses."
Greg Howe [41:22]: "Our community offers free templates and guides to help you get started with AI automations."
Further Resources
Embark on your journey to streamline and enhance your workflows with AI by tuning into this insightful episode. Stay curious and transform your productivity with the expert guidance of Cary Weston and Greg Howe.