Episode 444AIGo-to-MarketPrompt Engineering

How to use AI across sales, marketing, and customer success with Richmond Taylor

Richmond Taylor, founder of an AI automation and education business and the startup promptanything.io, uses the Feynman technique to collapse the entire go-to-market motion into a frame a five-year-old could repeat: sales is how you speak, marketing is how you look, and customer success is how you get the second date. Inside that frame, AI now handles roughly 80 percent of the work in each function, from custom outbound and meeting booking on the sales side, to scripts, frames, videos, tone, voice, and brand on the marketing side, to inbound responses and tier-one customer support. The unlock is prompt engineering, which Richmond compares to landing in a small Japanese town with no shared language: without it, you cannot navigate the AI world. He breaks prompts into four categories (system, user, developer, assistant) and explains why one big detailed prompt outperforms a long thread of short follow-ups, where each new message burns tokens, bloats context, and pushes the model toward contradictions and hallucinations. His closing recommendation is to start by mastering the prompt itself, because every other AI capability (agents, automations, web apps) is downstream of how clearly you can communicate with the model.

Richmond Taylor

Richmond Taylor

Founder, AI Automation and Education

14 min

Key Takeaways

  • 1Use the Feynman technique to collapse go-to-market into three connected functions a five-year-old could repeat: sales is how you speak (the dream you sell), marketing is how you look (showing the world the proof), and customer success is how you get the second date (renewal and expansion through nurture).
  • 2Treat AI as a force multiplier that can absorb roughly 80 percent of the work in each go-to-market function, including outbound sales and meeting booking, content scripts and brand-voice production, and inbound customer support, while you keep the creative 20 percent that AI cannot replicate.
  • 3Master prompt engineering before anything else, because it is the language layer for the entire AI world (agents, automations, workflows, web apps) and without it you are like a non-speaker dropped into a small Japanese town trying to find the bathroom.
  • 4Learn the four prompt categories (system, user, developer, assistant) so you can shape the model's role, the user request, the desired output structure, and the assistant behavior together in one frame instead of guessing your way through follow-ups.
  • 5Start with one big detailed prompt instead of a long iterative back-and-forth, because every follow-up message burns tokens and bloats the context window until the model contradicts itself and starts hallucinating.

About this episode

Most people use AI like a chatbot: one short prompt, a back-and-forth, and a mediocre output that gets worse the longer the thread runs. In this Content to Close episode, Richmond Taylor breaks down a smarter way to think about AI across the whole go-to-market motion. Richmond uses the Feynman technique to simplify go-to-market into three connected functions, sales is how you speak, marketing is how you look, and customer success is how you get the second date, and explains where AI can take over 80 percent of the work in each. He digs into why prompt engineering is the single skill that determines whether AI helps you or hallucinates on you, walks through the four prompt categories (system, user, developer, assistant), and explains why one big detailed prompt beats twenty short follow-ups every time. If you want a practical view of where AI fits inside a real business cycle, and how to stop wasting tokens on prompts that contradict themselves, this episode is worth your time.

Topics covered

  • Feynman technique applied to go-to-market
  • Sales, marketing, and customer success as one cycle
  • Where AI absorbs 80 percent of the work
  • The four prompt categories: system, user, developer, assistant
  • Why one big prompt beats many short follow-ups

Notable quotes

He was one of the lead scientists behind the Manhattan project, knew Einstein, was a smart guy. But Richard Feynman had this thing called the Feynman technique where you could explain such a complex subject like quantum physics to a five year old.

Richmond Taylor(2:34)

Sales is how you speak, marketing is how you look, and customer success is how you get that second date.

Richmond Taylor(4:09)

I do 80 % of the work myself and that's all thanks to AI.

Richmond Taylor(5:21)

People will start with one sentence and they go down this long rabbit hole and then your context window's totally just so robust, then the AI starts getting confused on what you said before and you contradicted yourself. And then it starts hallucinating and manipulating the information.

Richmond Taylor(10:38)

Resources mentioned

  • Framework

    Feynman Go-to-Market Wheel (Sales, Marketing, Customer Success)

    Use the Feynman technique to translate go-to-market into a single sentence anyone on your team can repeat. Sales is how you speak (the dream you sell at the start of a relationship). Marketing is how you look (the visual proof that the dream is real). Customer success is how you get the second date (the nurture motion that earns renewals and expansion). Whether you are a solo founder running all three or a larger company hiring for each, the cycle is identical, just deployed at different budgets. Use the wheel to decide which function to invest in next, and to clarify which problem each new hire or AI workflow is actually solving.

  • Playbook

    One-Big-Prompt Beats Twenty Follow-Ups

    Stop iterating with short messages and start with one detailed prompt that defines context, role, examples, output format, and constraints up front. Every follow-up message burns tokens, bloats the context window, and increases the chance the model contradicts itself or hallucinates. Learn the four prompt categories (system for the model's role and rules, user for the actual request, developer for output structure and constraints, assistant for desired behavior or example responses) and use them together in plain English. Treat prompt engineering as the literacy layer for the AI world, the same way English is the literacy layer for navigating an unfamiliar country, because every agent, automation, and workflow is downstream of it.

  • Tool

    promptanything.io

    Richmond's startup, built to make prompt engineering easier for non-technical users by generating customized prompts based on what you do, who you serve, and the workflow you want to build. The pitch is to get the first 80 percent of the prompt assembled for you so you can spend your time on the creative 20 percent, whether you are building an MVP, a sales agent, an email organizer, a script generator for a sales team, or a mentor agent. Even just opening one of the generated prompts is useful as a study aid for understanding how a strong prompt is structured.

Benjamin Ard (00:55) Welcome back to another episode of Content to Close. Today I'm joined by Richmond. Richmond, welcome to the show. Richmond Taylor (01:00) Glad to be here, my man. Benjamin Ard (01:01) Yeah, Richmond, I'm excited. This is going to be a fun conversation, extremely timely. We're going to dive into all the fun stuff when it comes to go to market and AI, all that good stuff. But before we dive in, let's have you introduce yourself to the audience so everyone knows a little bit about who you are. Richmond Taylor (01:17) Yeah, definitely. Thanks for asking. I love the one thing that we love talking about is ourselves, right? But it's always good to hear like some different stories out there. And that's something I love to bring to the table. But a little quick history behind me, little elevator on it. I played professional soccer until I was 26. From there, I used all that energy, discipline, winning, whatever I learned in the daily, the daily rituals of playing sports to learn sales, learn marketing, learn customer success. My dream was always to, how can I run a business and how can I make an impact in this world? From there, waking up every single day and focusing on that, I got the skill sets and then I opened up my own business eventually in the AI automation and education space. And we also have a startup into that space as well, which is helping us do our job and helping our clients even do their job a little bit better, faster, and more detailed. So it's been very exciting. Benjamin Ard (02:09) I love it. And there's some really cool things that you're building into this system. So let's talk about all sorts of cool stuff that we're going to today. Really, we're talking about the whole business cycle from sales to marketing to customer success, and then how all of that kind of connects into AI. So walk us through the framework. What do you mean when we're talking about a business circle or a cycle, how there's this continuous handoff between those three different go-to-market departments? Richmond Taylor (02:34) Yeah. And a lot of the times go to market seems like such a big deal, right? I mean, it's a, such a huge subject. How do you start a business? How do you grow a business? If you're an enterprise, how do you bring a product to market and go to market? But in the past, I've done a little bit of research in my end and I fell in love with this one guy named Richard Feynman. He was one of the lead scientists behind the Manhattan project, knew Einstein, was a smart guy. But Richard Feynman had this called the thing called the Feynman technique where you could explain such a complex subject like quantum physics to a five year old. And that's what we look to do with go to market. At least that's what I did for a while. I was consulting into that and got into the AI space, but really bulleted under three different topics or subjects. And that is sales, marketing, customer success. Now picture this. You have a sales at the top on the right side, you have marketing and then customer success as sort of a wheel. If you're a founder or going to market and figuring some new things out and testing the waters, you always start with sales. It's really how you talk and how you speak and how you sell the dream. From there, whether you're a founder, you're doing it yourself, or you already have a team, you go to marketing, you reinvest into marketing, and then you could show the world. It's not just an idea. I can just tell you, I'm going to show you. So you can get your visual. It's how you look and then you can speak and how you talk and how you tell the story. And then from there, after you get customers and they're really bought into the whole entire world, that goes into customer success, which is how you get your second date. So sales is how you look, is how you look, or sales is how you speak, marketing is how you look, and customer success is how you get that second date. If you're a founder, you do it all at once and then you can hire someone to do sales so you can focus on the marketing side and the customer success. Same thing goes with marketing. Then you can find someone to help run and be your CMO. And then you go into customer success and have someone really just nurture and lead. And then you can look into good or product. You can really work on the business and work not in the business. And that's how the whole circle works. No matter if you're low investment and low budget all the way to the highest, it all works in that same circle, just different ways. Benjamin Ard (04:42) Yeah, I love that. So like you were talking about, as things expand, you can invest in different areas there on the marketing, sales and success side. Traditionally, what we did is we went out and we hired new people to manage new elements in those different departments, but it feels like AI is disrupting that maybe just a little bit. Where are those kind of things changing? How is it being built in AI, what are you seeing in that space from an AI perspective that may be disrupting the whole idea of we just hire to fill problems as opposed to using AI to kind of help solve in some of those spaces. Richmond Taylor (05:15) Well, if you're really a one-stop shop, my understanding, and this is what I do now, I work with enterprise all the way up to mid-market, even some SMB, depending on the fit. But it's really just myself and I outsource some development and that's it. But I do 80 % of the work myself and that's all thanks to AI. Now, if you're in a larger organization, AI could be the people in helping you around you facilitate your job by itself. The more and larger you get, bigger and larger you get, the more intricate your processes you become. And that means the more intricate the AI can actually do to facilitate whatever you're looking to achieve. Now, if you're a one stop shop and you're a one person show, you can have AI help with your sales and book meetings on your calendar, making custom personal and relatively act as more like that outbound lead generation tool. With marketing, it could develop your scripts, your frames, your videos, your content itself and your tone, your voice and your brand and whatever you want to be able to achieve, where all you have to do is either review it until you're happy with it. You have nothing to review customer success inbound leads, being able to reach out. You got to people respond to emails, respond to inquiries. Now, if it's a little bit more intricate, yes, I could route it to more level two, level three customer success agent, but most of the bulk, the 80 % like I mentioned, it's now automatable. Benjamin Ard (06:38) I love that. And you're talking about on your side of things, you spend a lot of time helping educate how to use AI, how to take full advantage of it, set up these systems, things like that. What does the typical journey look like? Someone who may, you know, obviously I think most of us at this point in time know how to get into AI and write a blog post or a LinkedIn post or use it for content or do some research. How does it turn away from just content writing to something a little bit more productive? Richmond Taylor (07:05) No, I love this. And even like during content writing, it's going to have that forbidden M dash in there that we were all terrified to see. It's like, that's obviously AI or it says game changer. Like my grandpa would say, I'm like, yeah, yeah, buster. Gotcha. Love to see that. No, we don't want to see that. It's not in our tone, our voice under the brand from contents, from copywriting to where human or humans are creative the most. AI is not great at replicating it, unless you prompt it correctly. And the same thing goes with actual backend automations, whether you have an inbound sales agent or a customer success agent or a chat bot, or even a front end making the chat bot look pretty or your whole entire website, being able to fill, facilitate orders and e-commerce, but making it custom and personal to them, showing them the right kind of dress that they want or even thought about. And all of this boils down to is the prompt. The prompt is how we communicate with AI. It's how do we say exactly what he wanted to do, give it examples, show the output format that we want. And that's a skill. That's a skill that is tough to learn and master. But I did mention the startup that I have, but we built a tool that does it for you, where you can make this content custom to who you are, what you do, whatever kind of dog that you like to whatever application workflow, build an MVP, get the 80 % done so you can focus on being creative the last 20 % custom to your tech stack to what area of the world you're in, whether you guys like coffee or vegetarian, if you're selling some food. Benjamin Ard (08:41) I love that. So as you're doing this education process, I mean, there's a lot that goes into prompting. Where do people start? Like, where do they go? Obviously, there's a great solution in yours to be able to go out there and figure it out, but if they're wanting to really see how do I take myself up from one step to be better at prompting before maybe I need to implement a new solution or something like that. What are some tips on how to get better at prompting to make sure that AI is helping you honestly fill some of these gaps and you go to market circle and cycle. Richmond Taylor (09:09) My advice is really just jump in the rabbit hole and see where it takes you. We all have a different journey into this. I mean, myself, I want to learn how to do automations from there. Brought me to prompting from there. It brought me to more automations, to agent building, to web app, all these different things. But the number one skill, the one place you have to start is prompt engineering, communicating. It's like if you go, if you're not Japanese and you go to Japan and you don't know anything about the Japanese language and you go to a small town, that's all they speak. And are you going to be able to navigate or even find the bathroom, go on the train and find your next destination? No, no, it's going to be very difficult. You're going to have to find someone that speaks English. The same thing that goes to navigating the AI world. You got to learn how to prompt and speak to the agents, build your workflows, your web apps. And if you can learn how to do that and plain English, it's literally like coding, but a translation into just text we all can understand. And we could since we were five. Shout out Richard Feynman. Benjamin Ard (10:02) I love it, and I think that's amazing. So one thing that I always find interesting when we're on these podcasts, especially with someone who's spending all their time and energy in something very specific, they're getting into the nuance and things like that, you're also working with a lot of groups, helping them implement AI. I'm sure you're seeing things that are less than ideal practices, maybe some things that are pitfalls that people shouldn't be doing. Anything that you're seeing that's pretty typical of maybe some roadblocks that people run into some advice to say, Hey, please don't do this. This is going to go down an area. We don't want to go down anything like that. You're noticing. Richmond Taylor (10:38) Yeah, I mean, definitely. I'm still going to go, I'm always going to go back to the prompt because if you master that you can master AI. I mean, people will start with one sentence and they go down this long rabbit hole and then your context when those totally just so robust, then the AI starts getting confused on what you said before and you contradicted yourself. And then it starts hallucinating and manipulating the information. So then you look at AI and you're like, I can't use this. I don't want to use this anymore. If you get, or if you go and put a video and I want to make a video of a kangaroo jumping across, walking on water, it's, it's going to look a little bit goofy. Maybe it's got to what you want, but usually not. And the one thing that I do see that people really struggle with and kind of make those mistakes is that they, they start by just one by one, iterate and reiterate. And yeah, eventually it could look pretty good, but that's not how AI works. Every single one of those messages cost you a token and money. Hurts the environment even if you care about that. But it also gets so robust, the AI is now confused and it doesn't perform as well as starting with a general huge prompt. So if I could encourage anybody, really learn prompting and the four different categories of prompting and the whole of the message is your system prompt, your user prompt, your developer prompt, and your assistant prompt. I know you're a developer in there. Don't worry. It's really just like your, your output and like what you want it to look like. It's not, you're actually being developed. It's again, plain English, five year old could, could speak into AI and build probably just as well as somebody else. If you just know how to. Benjamin Ard (12:10) I love that. And why I think that's so cool, because a lot of us, myself included, I'm really good at the back and forth with the conversational side of AI. But I love the call out here of, start off with a much more robust, detailed, helpful prompt. And that way, you can have that context window be shrunk down a little bit, make sure things stay on focus, and then use those extra follow ups, not for core information, but for refinement. And I think that's awesome. That's so cool. Well, Richmond, we have run out of time. These episodes are short on purpose so that people can get back to their daily grind. This is amazing. Richmond, for anyone who's intrigued by everything that we're talking about and wants to reach out and connect with you online, how and where can they find you? Richmond Taylor (12:50) Yeah, so connect me with me on LinkedIn. I post a lot about automations and prompting and web applications and really just being able to one do your own agency or to build your own automations for your team that you have. It's easy. And I encourage you the next thing you ever prompt if you want to make an agent for a mentor or an agent that helps your emails and organizes or builds scripts out for your sales team. Anything, I named the tool promptanything.io for a reason. Go look to prompt on there in your next one. And trust me, you're gonna learn a lot about how a prompt's built by just looking at it, and you're not gonna have to spend nearly as much time building it yourself. Benjamin Ard (13:32) I love it. Perfect. For anyone listening, regardless of what platform you're on, scroll down to the show notes and inside of the show notes, we'll have all the links to all the different resources so you can click and connect with Richmond. Again, Richmond, thank you so much for the time and energy and for all of your insights today. It's been amazing. Appreciate it. Richmond Taylor (13:47) I appreciate you, Ben. Until next time.

About the guest

Richmond Taylor

Richmond Taylor

Founder, AI Automation and Education

Richmond Taylor played professional soccer until he was 26, then channeled the discipline of daily training into building skills across sales, marketing, and customer success. Today he runs his own business in the AI automation and education space, working with clients ranging from enterprise to mid-market and select SMB. He outsources only a slice of development and personally handles roughly 80 percent of the work, all because of how he uses AI. Richmond is also the founder of promptanything.io, a startup he built to make prompt engineering easier for non-technical users so the bulk of the work can get done in the first prompt instead of through endless follow-ups. His view is simple: AI is not a replacement for creativity, it is a force multiplier for anyone willing to learn how to communicate with it.

Connect on LinkedIn

Continue Exploring

Sales Enablement Software Buyer's Guide

A detailed guide to sales enablement platforms, rollout patterns, and what actually improves rep performance.

Read the guide

Get new episodes in your inbox

Join listeners who get episode summaries, key takeaways, and content strategy insights every week.

Frequently Asked Questions

Richmond borrows the Feynman technique, the rule that you should be able to explain a complex subject like quantum physics to a five-year-old, and applies it to the entire go-to-market motion. He collapses sales, marketing, and customer success into one sentence: sales is how you speak, marketing is how you look, and customer success is how you get the second date. The frame works whether you are a solo founder doing all three yourself or an enterprise company hiring leaders for each function. The cycle does not change with budget, only the way you staff and automate it.

Richmond estimates AI can absorb roughly 80 percent of the operational work in each function. On the sales side that looks like outbound lead generation, custom personalized outreach, and meeting booking. On the marketing side it produces scripts, frames, videos, content, tone, voice, and brand application, with a human reviewing until it is ready to ship. On the customer success side it handles inbound replies, email responses, and tier-one inquiries, while more complex issues are routed to a level-two or level-three human agent. Your job is the creative 20 percent and quality control on what the model produces.

Richmond compares prompting to landing in a small town in Japan with no Japanese, no English speakers around, and no way to find the bathroom or your next train. The same dynamic plays out in the AI world: agents, automations, workflows, and web apps are all built on top of prompts, so without prompting fluency you cannot navigate any of it. The good news is that the language is plain English, not code, which means anyone who could speak as a five-year-old can write a prompt. Master that one skill and every other AI capability becomes accessible.

Each follow-up message in a thread costs you tokens and money, and more importantly, it inflates the context window the model has to keep track of. As the context grows, the model starts noticing contradictions between what you said earlier and what you are asking now, and that confusion is exactly what triggers hallucinations and lower-quality output. Richmond's fix is to use the four prompt categories (system, user, developer, assistant) to write one comprehensive prompt up front that defines the role, the request, the output format, and the desired behavior. Refinement should happen in the first prompt, not across twenty short ones.

EP 44014 min

Why sales enablement is really a revenue execution system with Robin Schweitzer

with Robin Schweitzer

Most sales enablement teams are stuck running training programs when they should be running a revenue execution system. In this episode of Content to Close, Robin Schweitzer, a Revenue and Sales Enablement Executive with a background that spans carrying a bag, running marketing, and leading enablement, makes the case for pulling enablement out of the classroom and into live deals. Robin lays out the pillars she installs when she walks into a new role (rep readiness, pipeline management, deal execution), explains why SKO momentum dies a month later without micro-trainings tied to real deals, and shares the signal that helped her team lift close-won rate by 12 percent: a discovery-to-proposal ratio so lopsided it exposed a three-part discovery problem hiding in plain sight. She also walks through the data points she watches (threading, stage duration, talk-to-listen ratio, stall clusters), the four-quadrant stakeholder mapping exercise she uses to build credibility across C-suite, sales leadership, and reps, and when to carry a product or market problem back upstream on behalf of the team. If you own enablement and want to stop being treated as a cost center, start here.

April 24, 2026Listen
EP 43621 min

How to Operationalize Discovery So Buyers Sell Themselves with Nick Lopez

with Nick Lopez

Most discovery calls fall apart for the same reason: the seller asks surface-level questions, bounces between topics, and then defaults to pitch mode the second things get quiet. In this Content to Close episode, Nick Lopez walks through a discovery framework that fixes all of it. Nick teaches the 'pillar' approach, where every question you ask drills deeper into one topic before moving on, so you actually uncover the pain instead of skimming past it. He explains the 80/20 listening rule, why personal pain matters more than company pain, and the specific questions that get prospects to sell themselves on your solution. Nick also shares how to build a clean handoff from sales to customer success so the rapport you built in discovery doesn't get lost the second the deal closes.

April 17, 2026Listen
EP 42819 min

Sales Enablement Is a Revenue System, Not a Training Function with Jason Gwilliam

with Jason Gwilliam

Most companies still treat sales enablement like a training department. Jason Gwilliam thinks that's exactly why their reps take too long to close. With 25 years in healthcare and med-tech, Jason has built enablement programs from the ground up at companies like Abbott, and he's seen firsthand what happens when enablement is treated as a true revenue system. In this episode, he breaks down how to measure enablement's ROI through time-to-competency and sales cycle compression, why marketing alignment is critically undervalued, and how AI coaching tools should help reps improve without being punitive. He also shares why fractional enablement roles are emerging as the next big trend.

April 3, 2026Listen

Get new episodes in your inbox

Join listeners who get episode summaries, key takeaways, and content strategy insights every week.

We use cookies to improve your experience and analyze site traffic. Privacy Policy