S6E19 – Future-Proof Your GTM Strategy: From AI Hype to Real Results with Frank Sondors

Sales is changing fast. What once relied heavily on cold calls and manual follow-ups is now being reshaped by artificial intelligence. In this exciting episode of the Grow your B2B SaaS Podcast, Joran Hofman sits down with Frank Sondors CEO at Forge, who is a leading expert in sales and AI integration, to uncover how companies can use AI to build smarter, faster go-to-market strategies. Frank shares how he built his own company to $3 million in revenue in just one year by combining AI with smart sales processes. He talks about new roles like the GTM engineer, how AI agents can support sales teams, and why blending human effort with AI is the key to long-term success. Whether you’re a startup founder or leading an enterprise team, this episode is packed with practical advice to help you sell better in today’s fast-moving market.

Introduction to Frank Sondors

Frank Sondors is known for creating powerful AI tools that boost sales and outreach. After starting his career at Google, he launched his own company focused on making sales teams more efficient. His business hit $3 million in revenue within a year by using AI to streamline sales tasks and improve performance. Frank’s story shows what’s possible when technology and smart strategy come together.

The Evolution of Sales Teams

Frank talks about how traditional sales teams often rely too much on hiring more people, even though only a few truly excel in sales. This leads to wasted time, low conversion rates, and high costs. Seeing this problem, Frank started looking at how AI could help reduce the need for large sales teams while still increasing results. The goal is to make sales smarter, not just bigger.

The New Go-to-Market Approach with AI

Today, more companies are moving away from old-school sales methods. Frank explains how a new role called the GTM engineer is helping lead this change. These professionals use their technical skills to build automated systems that handle key parts of the sales process. This approach lets companies do more with fewer people and helps them grow faster without sacrificing quality.

Blending AI and Human Efforts

AI works best when it supports people, not replaces them. Frank explains how AI can handle repetitive tasks like sending follow-up emails or setting meetings, while sales reps focus on building relationships and closing deals. This mix of AI and human effort creates a stronger, more efficient team that can achieve better results without burning out.

The Role of AI Agents in Sales

AI agents are quickly becoming an important part of the sales process. Frank shares how his team uses these agents to follow up with leads, qualify prospects, and even have early conversations. By testing AI alongside human reps, his company has learned which tasks AI does best and where people still make the biggest impact. This helps them find the perfect balance for their team.

Overcoming Misconceptions About AI

Many businesses hesitate to use AI because they think it will take over jobs or isn’t as effective as humans. Frank clears up these fears by showing how AI is a tool that supports teams, not one that replaces them. He also explains how fast AI is improving and why staying current with these changes is essential for companies that want to stay competitive.

Implementing AI in Your Business

If you’re just starting with AI, Frank suggests beginning with simple tools like ChatGPT. This helps your team get used to working with AI before jumping into more advanced systems. He also recommends hiring someone with technical skills to build custom workflows that fit your business. The key is to start small, automate low-value tasks first, and scale over time.

Scaling from 0 to 10K MRR

For early-stage startups, Frank shares clear steps to reach $10,000 in monthly recurring revenue. Start by talking to customers, testing your idea with pre-orders, and focusing on solving real problems. It’s important to learn directly from your audience, adjust quickly, and build a strong foundation before trying to scale. Community support and early feedback are crucial during this phase.

Growing from 10K to 10 Million ARR

Once your business has momentum, the focus shifts to scaling. Frank explains that reaching $10 million in annual revenue requires building a solid team, hiring the right people, and making sure everyone is aligned with your mission. He also highlights the importance of content marketing, automation, and doing things that grow over time. Growth comes from doing the right things consistently and with focus.

The Future of AI and Human Collaboration

Looking ahead, Frank sees a future where AI and humans work side by side. AI will handle the repetitive, time-consuming tasks, while humans focus on strategy, creativity, and relationships. Sales teams will become more technical and more focused on value. This shift will make sales more efficient and open new opportunities for growth and innovation.

Final Thoughts and Takeaways

Frank’s message is simple. AI isn’t here to replace people. It’s here to help them do their jobs better. Businesses that embrace AI now will have a major advantage in the future. The key is to stay curious, experiment often, and build systems that make the most of both human talent and smart technology.

Connect with Frank Sondors

If you want to learn more about Frank’s work or get tips on how to use AI in your sales process, connect with him on LinkedIn. His insights are especially helpful for founders, sales leaders, and anyone building go-to-market strategies in today’s digital world. This episode is packed with actionable advice on how to use AI to improve your sales process and grow your business. If you found it helpful, please leave a review, share the podcast, and join us next time as we explore more ways to win in today’s business landscape. Thanks to Frank Sondors for sharing his time and expertise.

Key Timecodes

• (0:00) – Introduction to hiring challenges and sustainable business
• (0:51) – Guest intro: Frank Sondors and his ecosystem of tools
• (1:36) – Building an ecosystem similar to Apple and Frank’s sales journey
• (2:34) – Profit margins and the role of AI in business scaling
• (3:43) – Rapid growth and scaling with minimal headcount
• (4:41) – Old vs new go-to-market strategy leveraging AI
• (5:41) – The role of GTM engineers in sales and pipeline generation
• (6:45) – Sales development engineers and their responsibilities
• (7:40) – Increase in tech usage in sales and admin work reduction
• (8:32) – How AI can craft personalized emails
• (9:23) – Assembling custom workflows using tools like Zapier
• (10:15) – Voice AI agents for customer follow-up
• (11:06) – The value of human interaction in sales
• (11:55) – Decision making: AI agent vs human interaction
• (12:56) – Testing AI agents after hours and A/B testing strategies
• (13:57) – The misconception of automating all sales processes with AI
• (14:56) – Evolution and effectiveness of AI in business
• (16:10) – The resistance and acceptance of AI in the market
• (17:11) – The necessity of being profitable without VC funding
• (18:05) – Resistance to AI and the inevitability of change
• (19:39) – Implementing agents at scale and the role of NA10 people
• (20:32) – Encouraging team-wide automation and innovation
• (21:28) – Using WhatsApp as a channel for customer communication
• (22:28) – The importance of first impressions and human interactions
• (23:28) – The role of WhatsApp in increasing response rates
• (24:40) – Integrating WhatsApp with Slack for seamless communication
• (25:42) – Being available across multiple channels for customer engagement
• (26:08) – Feeding the right context to AI agents for effective interaction
• (27:10) – Starting with ChatGPT for AI integration
• (28:03) – Treating ChatGPT as a business coach
• (29:02) – Encouraging team use of ChatGPT for problem-solving
• (30:02) – Optimizing repetitive tasks with AI
• (31:09) – Personal pain points and the role of AI in alleviating them
• (32:05) – The focus on building a great product and selling it
• (32:38) – Testing AI agents vs humans and employee concerns
• (33:10) – The future of labor distribution and AI
• (34:09) – The role of humans in supervising AI agents
• (35:08) – The evolution of sales roles and the end of email templates
• (36:20) – AI agents communicating with each other in business
• (37:30) – The importance of context in AI interactions
• (37:45) – Advice for SaaS startups from 0 to 10K MRR
• (38:49) – Speaking to customers aggressively and leveraging networks
• (39:44) – Building in public and social selling strategies
• (40:33) – Doing things that don’t scale and leveraging communities
• (41:31) – The importance of hiring the right team from 10K to 10M ARR
• (42:22) – The challenge of hiring and firing for growth
• (43:14) – The value of hiring autonomous employees
• (44:06) – The problem of babysitting employees and hiring practices
• (45:06) – Compounding business strategies for growth
• (46:55) – Ensuring cofounders have the right competence
• (48:47) – The importance of pricing based on consumption
• (49:49) – De-risking for growth and the role of sparring partners
• (50:48) – Summary and closing remarks
• (53:19) – Contact information and closing
• (54:06) – Final thoughts and call to action for the audience

Transcription

– Frank Sondors

When it comes to hiring, and I’m just general attrition problem is that for every 10 salespeople that you’re hiring in the sales team today, only one person, best case, one person is a natural in sales. You hire them and they’re really designed for sales. They’re born to be a salesperson or whatever. Why wouldn’t you be profitable from day zero and build a sustainable business. If sustainable business means that you have to apply agents, then you apply agents, right? I don’t know actually many businesses where it just run purely on humans and they are sustainable. Most companies actually burn cash. In the world of business, I think only two things really Build a great product and sell that product. That’s it. So we should be building a better product, more of that better product, and we should be selling that product and looking after customers. Anything else that we do, admin work or something Today, we’re going to talk about how to leverage AI within your go-to-market strategy to book over 500 meetings per month.

– Joran

My guest today is Frank Sonders. Frank is building an ecosystem of tools which helps 10 actual outreach. He’s building seven tools at once. Salesforce, sales acquisition tool, Leadforce, search engine for leads, Primeforce, Google and Microsoft infrastructure, Mailforce, distributed email infrastructure, Infraforce, private email structure, Warnforce, email delivery tool, Agent Frank, which is an AI layer which works on top of all the different platforms. They grew from zero to 3 million within a year and are currently booking over 500 demos themselves on a monthly basis by combining AI and humans. What we’re going to do, we’ll dive into how you can leverage AI within your go-to-market strategy and also, of course, talk about their example. Welcome to the show, Frank.

– Frank Sondors

Thanks for having me. It’s good to be here. Yeah, there’s a lot of softwares, but think of it, it’s a bit like Apple. Apple also has a lot of products. We’re also doing something similar, but there is essentially what we call ecosystem benefits. If you’re using multiple products, you’ll feel that ecosystem play just like in Apple. So Apple has Apple ID, we have Forge ID, and it’s following a concept called Software Compound Model. We’re just building really cool Cool software for folks out there, specifically for sales purposes, helping them to build maximum possible pipeline with least possible headcount. I’ve been myself in sales for over a decade. Started my career in big corp in Google. Spent there for a bit, for two and a half years, and then transitioned to work for a bunch of cool software companies, SimoWeb, BlackRole out of US. Then very much two, three years ago, I became a VP sales for a sales team of 50. That’s where I realized actually the headcount problem, essentially, we were all scaling, and there are still companies There’s a lot of companies scaling with a lot of headcount. I was like, damn, that’s not the right way to think about things.

– Frank Sondors

We are a capitalist society, and why wouldn’t we want to increase our profit margins? Why wouldn’t we want to improve our CAC? You can see that there’s a lot of companies, cool companies today, that do build software in a way where they’re scaling revenue without necessarily adding headcount. Recently, there was a company where I think there was the one guy in Israel, I believe, that built a software, and he just sold that software after six months with 250,000 users for $80 million million dollars to wix. Com. This is just to unlock your imagination that you can build a really great company without necessarily having to hire humans. However, humans is the necessary what I call evil, just generally. Where you can’t get rid of all the humans, but you can become significantly lean and meaner, meaning small team, very powerful team. In tandem, you’re adding various automations, AI agents, for example. You’re blending, you’re putting that together. This is how you achieve fast security outcomes at the lower cost per unit, that’s booking a meeting or whatever else. That’s what we’ve been up to. Like you mentioned, we scaled in the first year, less than 12 months to about 3 million bucks in annual recurring revenue, profitable since day zero, even though we did raise half a million bucks in the round.

– Frank Sondors

The team right now, we scale from three cofounders to about 40 employees right now across the world, predominantly based in Europe. I think the fun story there is just that we scaled to a million bucks in the car with no employees. Then $2 million with no people in sales, marketing, and account management. We only added a few engineers and support people. At $2 million, I had enough of doing 20 meetings a day, every day, Monday to Friday. I was like, Okay, it’s time to hire some people. That’s an example. When you hire, then you pass on the process to individuals.

– Joran

It is interesting as you say, we want to do, I guess, less with less people, right? But you’re doing even the opposite, like you’re building multiple products, trying to be lean as possible. We’re going to dive into that. But the topic is about leveraging AI within you go-to-market. Maybe even super basic question, how do you see the difference between the old go-to-market and the new go-to-market when you actually leverage AI?

– Frank Sondors

The old go-to-market approach is all based on a predictable sales model, where you have SDRs, then you had account execs, and that was a typical thing that companies were implementing. That is all fine and dandy. But what happens over time, I think, salaries are rising. They were all competing for great sales talent. There’s a supply shortage for really great salespeople. Then as a sales leader, you just end up hiring a lot of people that shouldn’t be in sales. Just to give you an average stat in sales, when it comes to hiring and just general attrition problem, is that for every 10 salespeople, people that you hire in your sales team today, only one person, best case, one person is a natural in sales. You hire them and they’re really designed for sales. They’re born to be a salesperson or whatever. Then 2-3 people can be trained up to be somewhat good as the person that’s really natural in sales. Then Basically, 60% is what I typically call dead wood. These people should not be in sales. They are not accurate to your business. The reason why you need them is you need to hit your target and you are forced, typically, to hire these people.

– Frank Sondors

But typically, majority of the revenue, a part of the generation, sits within about 20% to 30% of your sales employee, and the rest shouldn’t be there, really. This is why I thought to myself, what if there was a way to reduce the 60% down to 0% and increase the output levels to 40%, this is how companies would become significantly more efficient and would increase their profitability as well along the way. Now, the new way of doing things, you could probably answer it in multiple different ways. But one of the things we have seen slightly the rise of a new title in the space called the GTM engineer as an example. That’s one way to answer the question. What does that mean is that instead of, say, having two, three, four, five SDRs, one go-to-market engineer, essentially a person that’s more technical, that person is able to do something with APIs, meaning hook up different API endpoints from different databases, clean the list at scale, scrape stuff, use AI, open AI to then qualify the leads. That doesn’t have to be done anymore by X number of humans manually in a spreadsheet. That can That’s what we’re doing.

– Frank Sondors

We’re done literally by one guy that just hooks up APIs and does it at scale. There’s a lot of naturally things that these GTM engineers do. I call them sales development engineers because for me, also I remember a lot of people haven’t been in sales. Yes, there were some sales development reps, SDRs that were slightly more technical, and they would do the different bits. Then they would promote it to sales operations managers, revenue operations managers. Sales ops and rep ops is a different beast to these GTM engineers. Gtm engineers are typically responsible just to build pipeline. That’s one way to answer. But the other way to answer is that today we’re seeing a lot more tech that’s being used by salespeople, specifically in the iSpace, to increase the output levels, do the work faster, kill the repetitive stuff or the admin work that happens, say, in the CRM or the sales execution layer like Salesforce Research, etc. We’re seeing a lot more of that. That means unlike two years ago when Salesforce was reporting that majority of the time spent by sales reps is actually not on selling, not on prospecting, it’s actually on an admin work.

– Frank Sondors

Today, we’re killing that admin work a lot with everything that’s happening on the automation space, and suddenly salespeople actually have more time to do the selling. I’m looking forward to seeing more data from Salesforce on that front personally. But what we’re seeing, not just in our own business, we are probably very much what we call a top 1% company for using AI, etc. But what we’re just generally seeing across many companies in the US and the Bay Area, but also coming to Europe a bit as well, is implementation of a lot of AI tech within their sales teams. What we do, for example, on the Agent Frank. We use our own agentfrank. Ai. You can go there, you can get a quick email from the agent by just entering your email address in LinkedIn, and you can see an example of what the AI can do, which means in this case, is crafting a unique email that will be delivered to you within seconds, literally within seconds. The way that it works is we’re essentially looking at the context of what we’re selling. Go on the web, check out what we know about you, Jorn.

– Frank Sondors

We’re going to go to your LinkedIn profile. We’re going to go to your website. Another data source says we match these two data sets and then we’ll compute and we’ll craft a great email that we deliver to you. That’s how it works. You can have a look at an example of what the AI can do. It can help you on the copy front. You can We’re not going to naturally use the whole agent to do the whole end-to-end prospecting. But what naturally the agent is what I call an off-the-shelf solution. It’s specifically designed for this end-to-end process, the booking meetings, and specifically in email, called email in particular. We are going to be adding LinkedIn. But then there’s other agentic stuff that I want to do for my business, and our software doesn’t do that today. If you’re unable to find an off-the-shelf solution, then you should build it, custom build it. But not where an engineer has to code it up, actually, but it The more, again, it goes into this world of GTM engineering. You want to assemble a workflow, a custom workflow. The tools that do that would be things like Zapier or make.

– Frank Sondors

Com. Or in any time, it’s what we use. It’s the most advanced solution. The way you can implement the agents, LLMs, It’s workflow, it’s connected to your Gmail account, your sales, we can connect it to everything. It’s a workflow tool. Definitely, you need to learn stuff in there. Let’s say you go to Salesforce. Ai today, you sign up for a trial, and then you decide you don’t want to buy anything from us. Fair enough. Maybe something happened, maybe it wasn’t the It’s your time, whatever. But instead of asking my sales team to go and give you a call or go and chase you, what we do is we have an agent, a voice AI agent that would call you after a couple of weeks and we’ll say, Hey, you’re on. How’s it going? I’ve seen that you started trial with Salesforce a couple of weeks ago, I was just wondering, how was your experience? Was there something missing? Why you didn’t end up progressing? To do this call, and if an actual call happens, the whole conversation cost me less than a dollar for our business to have that conversation going. Then two things typically happen.

– Frank Sondors

One, it happens always there’s an outcome of the call where it went straight to voicemail. There was a conversation with you, and there’s some essentially output, there’s a summary of the call. Or the second outcome, which is what we’re optimizing for, is trying to get you to book into a meeting. That’s one of the ways how We actually book folks on a call with our customer success team, for example. Why is an agent more powerful apart from the cost? You can feed all the concepts to an agent, your whole brain of your organization. So your whole knowledge base, playbook, pricing, whatever it is that you want to feed to the agent to represent your company, including conversations with customers. You can pull conversations with your customers from your CRM, from your customer support. We use GRRIPP for that, for example. We just pull all the conversations, we just give it to the agent. So agent can answer anything, any question on the call. Then you can instruct what you don’t want the agent to answer. For example, in our case, the agent never discusses pricing. The agent says, Oh, pricing, that’s not something I can discuss.

– Frank Sondors

It’s something best to discuss with the individual. If the agent can answer any question, can answer it very fast, the whole process is very cost-effective. It cost me less than a dollar to have a proper We’re in a conversation, why wouldn’t you implement it? For this particular step in the journey, if you use it, we’re saying this has to be an agentic stuff. It makes sense. Let’s have a look at another example where we do calls of don’t use AI agents at all. It doesn’t make sense to use AI agents there. When a sales rep starts their shift, let’s call it, or their workday, which isn’t in the morning because we’re in Europe, they actually start in the afternoon. They actually start after lunch, and they work till 10, 11 PM. The reason for that is because most of our customers are in the US. We work in the time zones of our main market. But then what we do is if you go to our website during those hours, there will be a widget bottom right where it says, Hey, do you want to chat to me? Do you want to talk to me?

– Frank Sondors

Essentially, it’s an ability to connect a human on one end, the prospect’s history, with the team on our end through web interface. I think of it as having a Zoom call on the website, essentially, in real-time. You could do a real-time demo if you want, answer any questions. We do about 200 human calls a month just talking on the site with our prospect. Why don’t we use AI agents? Because humans want to talk to another human. Intrinsically, that’s what you want to do in that particular case. We believe that a human is best fit to have that conversation. You You said you’re going to book the calls after a demo, right?

– Joran

You try to get them to book a demo. That’s the call when you actually call them after a sign up. Here you have that they can jump on a call like ad hoc on the website. How do you make the decision? Like AI agent versus not AI agent. I understand they have an intention when they’re on the website, they might be able to convert them better than giving them a call after a sign up. But what is your decision making behind it? Like when to use it or not?

– Frank Sondors

Good question. If it comes down to, Hey, we need to pitch this. We need to position this correctly, et cetera. Essentially, when it comes to humans, there’s a lot of nuances to things at the very much top of the funnel that humans are just better equipped, I think, in that particular case. There may be some questions about the agent that we don’t have the context against. Imagine if the question would be asked and the agent couldn’t answer, then the person feels left down. This is actually something we We’ll be testing, by the way, on the website in particular. Why would we be testing on the site? Imagine that human does a typically a nine-hour workday. There’s a rest of the hours. I always say A/B test is the if you can. But what we’re planning to do is when the humans are online, we’re going to have humans running the show and actually having conversations. The next test we want to run is when the humans go offline, we’re going to switch on the agents. I want to compare humans versus agent performance. This is what we do in Salesforce. You can do that in Salesforce.

– Frank Sondors

You can run campaigns in Salesforce, the campaigns that you for email and LinkedIn, and then you can run agentic campaigns. You can compare the performance. That’s what I want to do to understand what’s better. I’m always the guy that will test it. Sometimes you just don’t know whether it’s humans better equipped or the agent’s better equipped. But for Typically, for the experience on the website right now that we have, we don’t have yet the ability to switch on the agent. I’m still limited, let’s call, buy the tech for now. But yeah, ultimately, if the AI agent does a better job and it’s cheaper, why wouldn’t you implement? The reason why we decided, for example, to use the agent after sign up is because if they didn’t buy, then we just want to have a streamlined conversation, why didn’t you buy? The setup is very straightforward because they signed up, two weeks have passed, and we just want to find out why they didn’t end up buying and get them back on the conversation with the human. There’s a clearly defined process. The agent is much better equipped when there are these various nuances that happen very much at the top of the funnel.

– Frank Sondors

There’s essentially more control on the setup of the situation. But as the agent tech improves significantly over time, it will make sense to apply it in more and more touch points of the funnel. This is one thing that we do, example, where we would blend both the humans, the agents, but the different stages of the funnel. That’s how I believe the technology has to be built, because it’s not just in the sales, it’s also going to be the same in customer support. It happens today, and it will happen in finance and accounting. That’s essentially different spaces. We have to learn how to work with agents as a team, essentially. For humans, it’s a challenge to get used to it.

– Joran

Are you struggling to find a cost-effective and scalable marketing channel? Check out where it is. We help you to have other people recommend your SaaS, and you would only pay them when they deliver you paid clients. Making it a very cost-effective and scalable marketing channel. Want to learn more? Go to getReditus. Com. I think a lot of companies look at it like, Okay, hey, we’re going to go use AI in our go-to-market, and then we’re almost going to automate everything. I think this is maybe one the misconceptions a lot of companies have. Do you see that as well? What other misconceptions do you see companies having when they think about using AI in their go-to-market?

– Frank Sondors

I think a lot of companies were disappointed by AI last year. And so there’s a lot There’s a lot of feedback in the market that the AI doesn’t work, the AI agents don’t work, etc. The thing is, the whole tech, the whole AI agent space or AI is still in its infancy, arguably. Yes, there’s billions and billions of dollars being pumped in the space to make it better, but the technology is massively It’s evolving at such a pace that after three months or six months, suddenly it works. That’s where we are. We are right now in a stage where the tech actually works. Essentially, there’s no more excuses not to implement. Yes, there was negative feedback back in the market. The resistance is, Oh, it didn’t work. When I tested it, have you tested it now? Three months later or six months later, it suddenly worked. That’s how fast the tech is evolving in this space. Yeah, resistance, one, I had bad experience. I don’t want to use it. Resistance, two, no, human is going to do a much better job. But what people don’t realize is that, unfortunately, due to inflation and how expensive salespeople are, it’s just not sustainable for a lot of companies.

– Frank Sondors

They have to be raising a ton of money to sustain a company. They keep burning cash. Instead, why wouldn’t you be profitable like we are? Why wouldn’t you be profitable from day zero and build a sustainable business? If sustainable business means that you have to apply agents, then you apply agents. I don’t know actually many businesses where it just run purely on humans and they are sustainable. Most companies actually burn cash. They’re not that profitable. We’re trying to be profitable. There’s a weird company, we’re in tech and we’re profitable. There’s not that many. But what AI and LLMs and AI agents allows you to do is actually build a business without the necessity to get VC funding. That was not possible two years ago. Now it is suddenly possible. A lot of companies are doing that. I highly encourage people to essentially consider not to raise money nationally these days because you don’t have to anymore. It’s just much cheaper if you’re able to figure out a business on how to run a fully agentic workflow. I think the other problem is that I’m seeing. Ai agents and this whole thing about AI, AI, is essentially it’s a bit like the internet moment.

– Frank Sondors

There’s essentially resistance by humans. Just back in days with internet and computer, there’s a resistance to change. People don’t like change, but change is inevitable.

– Joran

You see a lot of posts on LinkedIn and these complex screenshots of how I automated my sales flow. Then you see, I think, a screenshot which you can’t read, but it has 50 agents on there or something like that. Maybe that scares people off as well. The thing they’re often doing is selling their course or agent or selling their thing, but not actually growing their business, or at least that feels like it for me. Is that something you see as well?

– Frank Sondors

The whole change has to come actually from top down. I don’t see it coming bottoms up in any organization. What does that mean in practical terms? I’ll give an example of Salesforce Forge. At the beginning of this year, I said to everybody in the organization that we will be implementing agents at scale in every part of the business, and we’re going to be killing repetitive stuff. The way we’re going to do it is we’re going to implement N810, and I will hire an N810 guy or girl to do it. Guess what? I hired an N810 person. I don’t know how to do it. Yes, you could say my engineering team can figure out how to do it, but you have to be in this mindset of being hacky and building agents, the N810 people are very specific type of people. They are technical, but at the same time, they understand the commercial realm as well, and they are able to build something against the vision that you have as a CEO or as a commercial for work. These are not really engineers, but then neither the commercial people, whatever. They’re like this mixed breed. I call them the N810 people, the hacky people that understand both worlds and can put something together.

– Frank Sondors

The dilemma is implementing this. Then, yes, there are courses, et cetera. But I think in a lot of cases, yes, sure, I can also do a course. If I had time, I would do a course and build this. But then, really, just hire somebody. This person doesn’t look full-time for us. He’s part-time, and every week, he builds 5-10 N-8-10 workflows in our business to kill everything that’s repetitive. After three months to I would say 3-12 months, you’re going to start feeling, wow, everything is automated in our business. I had this feeling back in my days working at Google. At Google, everything’s automated. You do nothing. You get a report, this happens, if something good happens, you get a ping, if something bad happens, you get a ping. Everything’s I made it. Humans are the face of the machine at Google. I always wanted to have that feeling of Google back in a small startup. That’s why all the inspiration that I got in the bigger companies, I’m just implementing my little startup using NA10. All I do on a weekly basis together with my team, it’s not just me, but naturally, you have to lead by example.

– Frank Sondors

What I did is we created this channel called NA10 Ideas, and then I started to drop many ideas as the CEO to encourage the team. We even crack in jokes in the kitchen or whatever in the Slack channel. Is this an N810 thing when something is competitive? Essentially, if you’re doing this multiple times a week, you have to ask, Is this an N810 thing? If yes, just put it into the channel. It will prioritize and the N810 guy will build it for you. You’ll be very grateful. Trust me, I think 80, 99% of the time, people are so grateful that this was implemented. Yes, things maybe didn’t go to plan the other 10% of the time, or you have seen no lift in performance, no satisfaction. Fair enough, then you optimize that. But it works majority of the time, and people are happy and happy. You enjoy your work much more, actually. It’s not just, Oh, we’re more efficient. But actually, employees enjoy working this company more and more with every single week. Because I bet anybody who’s listening to this, you work in companies where you’re like, damn, this is horrible. You’ve done jobs that are repetitive, horrible.

– Frank Sondors

It’s something you wish you didn’t have to do. Keep on clicking stuff, whatever, copy-pasting, whatever. It happens in every company. I used to do that. That’s why with the N810 comes in or something similar to just kill that. Then work on really things that really matter. What does it matter? Talking to your customers. This is why we believe if you put the human in front, straight on the website, actually, it gives us, let’s call it the brownie points in terms of them going with us than anybody else in the space. Why that’s important? Let’s say if you go to any of our other competitors, and on a competitor There are websites you cannot speak to a human life like you can in our case. The reason why this is important is because there’s so much software there, so many solutions for the same pain to be solved. When you’re subscribing to your product, you’re not just subscribing to that product or that software, you’re also renting the team behind that. If the team is there, it’s on the side, and it’s available, and you feel like this could be your extended team to your current team, you’re more likely to go with this company.

– Frank Sondors

It’s not that people just buy Salesforce. No, they rent Frank and his team at the same time. That’s how you have to think about it. Don’t shy away to talk to your customer. This is why, again, to answer your question around why humans in that particular, because first impression also matters. Yes, we are an agent company, and yes, we run agents, but at the same time, the first impression that there is access to human to help you run this agent successfully is very powerful from a first impression standpoint. We don’t shy away to talk to customers. The other way maybe to amplify this, not shying away, throwing to customers. So naturally, we talk to customers on the website via email. We’re going to send them onboarding email. Yes, our voice AI agent will call you. So you’ll have an experience even talking on a call. But we even would send you a WhatsApp message. And the reason why WhatsApp is so important. Whatsapp is it diffuses this two companies. When you get your prospect or your user on WhatsApp, we don’t have this barrier in place that we have to be very official. Suddenly, you can get to a stage where you’re having a real conversation.

– Frank Sondors

You’re much faster on WhatsApp than on an email or even on a Zoom call, etc. People open up themselves on WhatsApp a lot more than any other channel.

– Joran

Where do you get this information? Would they leave their actual phone when they sign up? Because personally, if I sign up to something, I can clean up to use my own phone number just, I guess, to avoid cold calls or WhatsApp messages, but I’m maybe not the best person to ask. 90% actually fills out their correct phone number.

– Frank Sondors

Yeah, majority of people do fill it out. Majority of the time… What we do is we check against that number that has been given to us, whether there’s a WhatsApp associated. If yes, we would fire just a quick only one message. Hey, if you got any questions, just let me know. Only one little message. We’re not pitching on WhatsApp, et cetera. We are trying to say, Hey, if you need me, I’m available. I’m here for you. The reason why this is also important is a lot of people just spend a lot of time on WhatsApp, and sometimes they think about a problem, and Frank just messaged you saying, Hey, if you’re not going to get a question, let me know. What happens actually behind the scenes is that we hook up the WhatsApp, let’s call it engine, to Slack. There’s a Slack channel which says WhatsApp, and When the user responds back, then we have a conversation in Slack. It could be me. It could be also another human that’s answering the question. It could be also an AI agent if you wanted really to be. You definitely can also apply AI agents in WhatsApp.

– Frank Sondors

It just depends on your strategy. But the reason why also WhatsApp is fantastic. Also, let’s look from a metric standpoint. You will get at least 10X, reply rate on WhatsApp, then email on inbound. 10x. Now, look at your reply rates and onboarding emails, whatever inbound comes. Imagine if you had a 10X reply. Wouldn’t you want to implement a number on your site. Now, in case it works, because we are selling to salespeople, we’re selling to founders, etc, they now will be in touch. You can definitely test it. Whereas my marketing team hates the idea of having a number on it. It’s always going to ruin the conversion rate. Sure, you can say some people will not want to progress. However, what people will do like you, Yoren, they’ll just put a fake number. Okay, you’ll put a fake number, no problem. But then I won’t be able to provide you that service. My argument in my head of growth is it’s not going to win the conversion rate. But at the same time in sales, it depends on the ideal customer. Sometimes it makes sense, sometimes it doesn’t make sense to have a number. In our case, our audience always has questions, always wants to ask something, blah, blah, blah.

– Frank Sondors

So providing and being available out there in the market is super critical across all channels. Linkedin that we do, all channels possible, whatever the channel they prefer, you don’t know which channel they prefer. That’s why we do four channels in total. Email, LinkedIn, voice, human plus AI agent, That’s it. So four channels that we’re available.

– Joran

Even though you are available, it happens. As you mentioned, what happens? Maybe LinkedIn even happens in Slack as well.

– Frank Sondors

Yeah, and that’s the context that you feed into the AI agent. So when the AI agent calls, it checks if you had any conversations with this particular prospect in the past. And that’s an additional context that we feed to the agent. So it may say, Hey, how’s it going? I saw you did sign up, but you didn’t end up buying. What happened? I saw you spoke to Frank on WhatsApp. And that’s possible today.

– Joran

This is a really important topic, But having the right context when they’re going to actually talk to somebody on WhatsApp, even though you already had a demo with them or they might have a chat on the website. One of my biggest concerns when trying to automate things that we might end up having a begin conversation again, even though the client is somewhere else. So having the right context and having access to the data, I think that’s super important. And that might go really well into my question. If somebody is listening in and I want to future proof my go-to-market and start implementing AI, what would you recommend you mention if you didn’t hire full-time GTM engineer or an engineer did it on freelance? You have this list of things you can automate. What is a step-by-step process? How could they get started and figure their way out?

– Frank Sondors

Sure. I think everything starts in ChatGPT today. Somebody asked me, Frank, How do I start with Faint? Why don’t you start with ChatGPT? Are you probably used to it. If you haven’t used ChatGPT, guess what? This is a good opportunity for you to start using ChatGPT. I think we’re all biased that all people use ChatGPT. I still see a lot of people in my circles that have not even touched ChatGPT. What is this? All this about AI buzz out there. Chatgpt is definitely… The reason why I’m saying with ChatGPT, ChatGPT learns from every single conversation that you’re having with ChatGPT. The more you talk to it, the better the outputs become. I highly recommend that you start treating ChatGPT as your little business coach or your sales coach, and you’ll start to see benefits. The more you talk, the better it gets. They’re like, Wow, he knows everything about me. I can ask him, How would I approach this? How would they think? What are the challenges? It becomes your little coach. You don’t have to pay a coach anymore. That’s where it starts. You can to also provide context about your business. Hey, I work at Salesforce.

– Frank Sondors

These are my problems. These are my competitors. How should I approach this problem? Give me three ideas. This is my budget. Just have a chat with it. Get used to that for every single problem that you have. When you have a problem, go and chat to ChatGPT. For example, we had this We’re back with my cofounders. We’re discussing, I think, promotions of particular employees. How should we do this, et cetera. We were in what’s called the stalemate. I had my opinion. He had his opinion, my CTO. I was like, Let’s chat GPT. Chatgpt is the referee essentially that jumps in and gives you advice. Based on this scenario, what would ChatGPT or the agent do? What is based on the context of our business? It definitely is powerful in different scenarios. You can speak to ChatGPT about specific deals and sales, preparing for team meetings or whatever. There’s so many different ways on how you should train yourself to use AI just generally. You start with ChatGPT. Then naturally, if you want to take it up a notch, you want to be like, Okay, I’m doing all these things, but I wish it would be more We would build this into our business more.

– Frank Sondors

The next step becomes empowering your other people in your company to do it. You should always ask them, Are you using ChatGPT for this? Why are you not using ChatGPT for this? Why wouldn’t you want to make your job easier, better, faster, whatever? It’s amplifying, but very much it’s very important. Top down, you’re telling people why they’re not using ChatGPT. If somebody tells me, Okay, I’m working this problem. I ask, Are you using ChatGPT? No. Why are you not using ChatGPT for this? You’re challenging the way they think about things. I I still challenge even my co-founders. Us building AI agents, we have one problem now. We are hiring engineers, and my CTO is spending 10, 20 hours a week to do hiring. I was like, So you do this every week and you’re spending 10, 20 hours a week to do hiring? This is a fancy Fantastic use case for AI, NA10, whatever you name it. That’s how you have to think about it. Whenever you’re spending a big chunk of your work on something, you have to start thinking about how to optimize this whole thing. Because I told my CTO, yes, Unless you’re spending on hiring, and yes, hiring the right people is important.

– Frank Sondors

But what I want you to do, and I think other people in the company want you to do, is speak more to your customers. You’re not speaking to your customers, really, the product they’re building for. Instead, you’re just telling me you’re hiring. But then we have this burning, and that needs to be doing right, and this needs to be happening. If you’re just spending a huge chunk just on hiring, that needs to optimize. Yes, you could say, Oh, but it’s very important that the human is always in the loop. I agree, but if it’s an efficient for both sides, the employer, essentially, and the candidate for that particular job. If it’s inefficient, then it’s prone to optimization. I tell him to go and look for some AI solution to at least for part of the process in hiring to optimize that. Because spending 10, 20 hours a week, every week, feels like crazy. Because hiring will never go away. He’s hiring, my CTO is hiring since Q1 last year. Hiring has never stopped for him. That means all these hours he’s been putting into hiring instead of building a better product. That’s just another real life example from here at the Forge on how we’re thinking about not just sales, but also other areas of the business that you need to implement.

– Frank Sondors

For me, there’s also a personal pain I have as a CEO. For me, something that’s really annoying is collecting invoices and passing over to my accounting team, et cetera. Guess what? I said, We need to build an AI agent that hooks up to my mailbox, scans for all the invoices in my mailbox. Trust me, some people are like, Oh, no, I don’t want anybody to have access to my mailbox. I don’t care. I As long as it goes and find these invoices, instead of me trying to find the invoices across all the emails that need to be forwarded, etc. That’s not on your own individual level that can be done to make your life easier. Because I also sometimes spend 1-2 hours a week trying to chase these invoices. For me, that’s crazy. Why? It doesn’t generate revenue. It’s adding in work and not very enjoyable. Anything that has to do with accounting just generally is not enjoyable. That’s why another example. If something sucks in your company, yes, maybe it’s only one hour or two, but imagine if that pain, I could just get rid of Again, your quality of work, something goes up because you suddenly eliminated this thing that you really hate through an agent.

– Frank Sondors

You will start appreciating agents and automations a lot more. Here are a few examples on how you could definitely kill in order to focus on what matters in our business is to have these 500 plus meetings that we had last month. How do I repeat this every month? How do I make sure it’s growing month for month? How do I talk to more customers? That’s really what we’re all about, I’m building a great product. Folks, in the world of business, I think only two things really matter. Build a great product and sell that product. That’s it. We should be building a better product, more of that better product, and we should be selling that product and looking after customers. Anything else that we do, admin work or something?

– Joran

Everything you do leads up to talking to the client, AI agent working together with a human to get to talk to the client. You mentioned you are going to run a test as an agent versus human, at least outside of office hours. I think maybe one thing people are afraid of, how are your employees going to think about that? What if the experiment actually works out really well in the Each one does it better than humans. Is there any setback within the company or do they embrace it because they know they’re going to have something else to do when it actually works out or how does that work?

– Frank Sondors

So humans should be doing what humans are best at and then somebody that they’re not good at, they should not be doing that. And then there’s also the legal side of things. So let’s talk about the legal side of things. Let’s say in the world of sales, using any form of autonomous systems for call calling is illegal in the US and Europe. So that means you as a human, your job It is safe if you do call calling. It’s safe because the agents cannot intervene in that particular channel. Nothing is allowed. That’s one way to look at it. But the other way is we will see a lot of redistribution of labor over the next five years. Think of it back in the days with Ford and Ford factories, when we had this assembly line, and we had people working on the cars, and there’s a gazillion of people in factories. And today, what happens is the robots are assembling the cars and humans are supervising. We’re We’re going to see something similar. We will have humans that are supervising agents. I’m supervising agents. My N10 guy is supervising agents. We’re already doing that. We’re just going to do more of that.

– Frank Sondors

We’re going through this software revolution where AI agents use softwares and they run stuff, and the humans that supervise. We’re going to have a ton of supervisors out there, prompt engineers, etc. It will happen in sales. In sales in particular, my recommendation is, A, you either specialize in things like call calling, that will never go away, social selling, the agents on Sure, you can have a lot of fake avatars on LinkedIn, but I don’t think people want to follow fake avatars on LinkedIn. Social selling on LinkedIn and other channels will be always a thing, essentially becoming more of a, it’s called ambassador of your company. These jobs, I think more will be out there. Influencers is a thing that’s coming. Authentic voice of a brand or of a space that’s definitely going to stick around. Then the other side of the things is people that actually build all these automations. Essentially in sales, we’re going to see a lot more technical individuals because it’s not good enough that you can type some emails, whatever. It feels like the economy and society and the humans have moved on. Why are we still writing emails that humans are typically horrible at?

– Frank Sondors

Humans are horrible at copywriting most of the time. So humans are horrible at copywriting most of the time. That’s why you need to There are automations agents, et cetera, data layers, whatever. There’s a lot of things that go into that. So that is going away from humans. We’re going to see, I think, the death of what I call email templates moving towards a stage in an email where we’re going to see more and more what I call programmatic emails, emails that will be highly relevant, highly targeted to every single human at scale. Billions of personalized emails to every human computed by AI. That’s where we’re headed. So templates will die. The thing of it like advertising on Facebook, where Facebook knows everything about you, your gender, everything where you’re based, what you look like. That’s how emails will become a lot more relevant because I can find everything about you on the web. How many podcasts you did with Home and on LinkedIn, on your website? And I can I can craft a much better email today than I could two, three years ago. We’re moving towards this phase. But at the same time, the more AI on automation, whatever is going to happen on the seller side, we We’re also going to start seeing more and more AI on automation on the buyer side within the sales.

– Frank Sondors

Ultimately, we’re moving towards a direction where an AI agent on the seller side will be talking to an AI agent on the buyer side because two businesses are ineffective to communicate with each other. The The sellers or the sales reps, don’t crop, maybe, let’s say, good emails or good messages, etc. Sure. Let’s apply AI agent. Ai agent will do a good job. But then on the buyer side, the buyers are potentially overwhelmed by the number of emails that are coming into the mailbox. You need to use an agent to qualify conversations and just see what this email from this vendor aligns with my business challenges and pain that I have. Personally, I want to talk to that vendor. Everybody else, that’s garbage. I think that’s where we’re moving, where agents talk to each other before something happens. I actually already have seen this. Let me give you a couple of examples on that front. On one hand, when we have called up after two weeks of somebody having not bought anything from us, our voice AI agent has talked to another voice AI agent, and they had a bit of a conversation back and forth.

– Frank Sondors

That’s one example. On in The other example where we sent an email to somebody and that somebody uses an agent, so they’re already having a conversation. We’re really seeing this at a tiny scale. This will accelerate. That’s where we headed. That’s what’s coming.

– Joran

In the future, I think this is going to happen. Let’s start wrapping things up. We’re going to dive into two famous questions about growing a B2B SaaS. We already passed this stage, but what advice would you give a SaaS market starting out and growing from 0 to 10K money What’s the value?

– Frank Sondors

Zero to 10K. Definitely read the book by Paul Graham from YC about doing things that don’t scale. What does that mean in practical terms for us? We actually closed four customers before writing a single line of code. That gives you conviction, even though you believe in your business, if you have no paying customers, nothing will happen. The best way to validate whatever is that you’re doing, you’re at zero. The best way to validate, should I actually build something? Should I actually start coding whatever? Is when you have already paid pre-orders. Engineers freak out about pay pre-orders. They are getting motivated. This is how I got my two co-founders super motivated, both our engineers, to build this stuff with me. On one hand, to validate what you’re doing, try to get pre-orders before you start tinkering and doing stuff, if you can. Sometimes you can’t for various reasons. Then the other thing that I definitely want to say is, and it sounds obvious, I think, is you should be speaking to customers. When I ask a lot of founders today in Europe and US and a lot of other places in the world, Hey, to how many customers or potential user are you speaking to a weekly basis?

– Frank Sondors

Somebody says one or two, and that’s not enough. You should be speaking to 10 to 40 customers a weekly before selling. In order to get the product market fit and in order to really start scaling and figuring out how to sell this, how to distribute this, etc. Speak to your customers, but aggressively. Have X number of meetings per day. I used to do up to 20 meetings a day when I was starting off. Every day, up to 20 meetings. My record is 27 meetings in a day last year. If you go at that velocity, and yes, we work Saturdays and Sundays because that’s when I do my admin follow-ups a lot of the time. But if you go at that velocity, you cannot get it wrong in the early days. Where do you get all these? It’s called meetings, whatever. How do you start drumming up? Of course, you want build in public. Go on LinkedIn or X, whatever. Just talk about what is it that you’re building, the pain that you’re solving. Just say you spoke to some customer last week, what happened? Try and resonate with the audience. Next on LinkedIn with 100 people per week, you can connect 100 people per week.

– Frank Sondors

Do it every week. Every week, you connect with 100 ICP, so ideal customers of yours, connect with them, and then feed them, do social selling. Feed them with great content. What have you learned? What you’re working on? Where you are? What are you up to? And so on and so forth. They can follow your journey. Then when you’re going to reach out to them with a request to jump on the call, et cetera, because you fed them with a lot of great content, it’s going to be a lot of warm introduction. It’s much easier to convert people as in getting them on a demo, et cetera. Because a lot of cases, let’s say my sales rep reaches out today, and a lot of the times we get, Oh, do you work with Frank? The rep says, Yes, Frank is a great guy. I’ve been following him, blah, blah, blah. It’s very easy then to get that person on a demo. You definitely want to be connecting early on. Then you want to activate your whole network that you can. Actually, I’ve been in sales for over a decade, so I have some domain expertise, some network.

– Frank Sondors

I spoke to other sales leaders. I will go to them. If they’re not interested in my product, they’re talking to me, whatever, always proactively ask for a referral. I get it. It’s not for you or you don’t have time, but do you know anybody in your close circle that could benefit from this? I think what Europeans really struggle to do is to proactively ask for referrals. It’s something that Americans are very much trained on to proactively ask. Americans, actually, the bias, or also do it a lot of times. They’re always thinking, Who could I introduce you to? It’s just normal for them to do that. In Europe, we’re struggling on the front, massively. Nobody thinks about referral as a thing. Then you have to also access these, let’s call mini communities, where it maybe is much easier to spark a conversation in the early days. For stack communities or general communities you could access, just go to highindex. Net, I believe. I can’t remember. Hiveindex, I think it’s called. Just Google it. There’s a lot of communities that you can search for. The other thing that you can do is to figure out which communities your ICPs are hanging out is ask your ICPs.

– Frank Sondors

You ask them, Hey, which events are you attending? Which communities are you in? Et cetera. And you want to join it. For example, today, I asked another company, Hey, I just wanted to know, which events are you attending? Where are you hanging out? Et cetera. Just do it all the time. I even did it today with somebody. I’m just trying to find new events, cross-pollet my brand, et cetera. And then do things that don’t scale by program. I went and wrote some blog posts about sales, called email, whatever, back in the days. And guess what? I didn’t write it myself. I did it with AI. It was crappy back in days, but it worked. I got some clicks. I got 10 clicks in a day. You get this dopamine hit, the things are compounding bit by bit. Today, we do crazy stuff, like real crazy stuff. We heavily doing programmatic SEO, YouTube programmatic. I wanted to drum up more leads. Community is doing things that don’t scale like SEO. Also on YouTube, I used to do YouTube videos. Hey, guys, this is how I write an email. I’ll do some demos. All of that compounds.

– Frank Sondors

Then the other thing, I think it depends on the audience, but if your target audience is in the US, Reditus is super powerful. You just go on Reditus, go into community, say, Hey, guys, I’ve launched this. I’ve built this, I’m working on. Some people are like, Oh, wow, this is cool. I would love to check this out. Think about this, how do you distribute these? Sure, this channel did not work out this time. Do it again next time because you have to be persistent. Always be where your customers are. In our case, we have the luxury that our ideal customers are in different places, and we’re trying to figure out where things work, where do we get traction. For example, what we dropped because we didn’t think it’s going to scale to any level is we stopped being in Facebook groups.

– Joran

If you were doing everything you just mentioned, you reach 10K. I think this goes then to the next phase where you’re currently in yourself, Chrome towards 10 million ARR. If you did all of this, and you’re probably still doing it after 10K, but how would you go from 10K to 10 million ARR? What advice would you even give yourself?

– Frank Sondors

No, I typically think the one million, typically that step is now two, three million for different reasons, inflations, whatever. But essentially, my belief is that through this market, two, three million, you can do a lot of cool stuff without a big team. We scale nine employees to a million, two employees with just a few engineers in support. Then at two, three, two million, we started to hire go-to-market people because I knew it back then, but it was a point where you’re like, damn, I’m never going to scale it on my own to 10 million. Never. It’s just not going to happen. It forces you, and it doesn’t matter how many AI agents I would apply. Even though we’re super lean and mean back in the days, you need a team. If somebody was watching the Silicon Valley, you need your stallions. You need the other CEOs that they used to work for. They’re called the core 50 employees. You need the team that will get you to 10 million, the key team in the early days of any company. Finding that team is super difficult. I highly recommend nearly every founder, hire fast and fire faster.

– Frank Sondors

That is super important. I think one of the key things that we do once you’re to 3 million and to 10 million, the number one thing you will be doing, it doesn’t matter how else we’re doing, is you’ll be hiring people aggressively and you’ll be firing them also aggressively. Ultimately, the dilemma is that there is a huge problem that you can’t really know whether these people that you’re hiring are actually good for the job. It doesn’t matter how many interviews you’re going to do. Doing more interviews does not de-risk, unfortunately enough. The best thing that you can do for your business, if you’re done one or two interviews and you feel like, I think this could be something here, just get them into your… Put the bums on seats as quick as possible. Give them the tools, the budget, everything. Run it for two, three months and see if you have the commitment Conviction on that employee. Just like he sees you have conviction on founders, have a feeling whether you have conviction, whether you see things going up or not. There’s the CEO of Revolut here in Europe. He says that he understands whether it’s a yes or no in two weeks.

– Frank Sondors

I can actually do that as well, I quite a while in two weeks. I can figure out this person is for us or not. Two weeks, like first time, I need, but we have this period of three months. By default, you will be let go by default in three months. That means we build a business case internally. There’s three of us, three cofounders, another person. We build a conviction that are we keeping this person or not? Everybody has to say yes or has to agree that we are keeping this person and we’re building a business case. Anybody in a team can veto to pass the It has happened in engineering product that engineering product says, We’re keeping this person, and I have vetoed, and that person has been let go. There has to be conviction across go-to-market and product engineering that this is a person that definitely keeping every time up to 50 employees. Because beyond 50, you’ll be forced to have management layers. But right now, we have no managers in the company. Imagine running a 40 people team, three cofounders, no manager. How you’re able to do that is by hiring what I call autonomous humans.

– Frank Sondors

Humans that will feel the ownership in their territory that they’re doing. Another guy that I listened to, actually, he called them entrepreneurs. Essentially, the CEO of your own turf, so that’s how you can think about it. You have big sense of ownership. You want to make it, you want to grow this baby. The vice versa example, when do you know to let them go, is when you feel like you’re babysitting. When you feel like as a CEO or whatever, as a manager, you feel like you’re doing babysitting, hardcore babysitting in a company up to 50 employees, it’s a sign you have to let them go, typically. Because you are spending as a manager or as an individual a lot of time managing these people to an extent where you feel like, damn, all that time spent. But really, the great employees, I think in every company are the ones that are fully autonomous or as much as that can be. The big part is just hiring in people. I think this topic isn’t discussed as much as it should be, especially from the 2, 3 million mark to 10 million mark. That is the topic.

– Frank Sondors

The other topics or areas of importance, naturally, is, let’s call the of how does the business compound and everything that we do. For example, what doesn’t compound is me spending money on Google Ads because I know I’m just throwing thousands of dollars into Google Ads and it’s like a bottomless pit that doesn’t compound. What does compound is content, YouTube, word of mouth, brand. It should be easier for you to scale your business. You need to build a process in your business where unit economics gets better, margins get better, or your top line gets better. Figure how to compound from 3 to 10 million, super important because otherwise, you will not succeed from 10 million then to 50 million if you haven’t built that logic and the process in your business. In our case, the compound, there’s multiple things we do, but from a product standpoint, the way we compound is by building multiple softwares that we stack on top of each other. It’s called the software compound model or startup compound model. Same thing, really. That’s how we’re compounding. The more products we ship, the more good products we ship, horizontally, there’s a lot of products.

– Frank Sondors

We sell as the same user, our business compounds and the users buy multiple products from us. Same as Apple. There’s a logic that you have to build. There’s a lot that goes, I think, into strategy achieve from 3 to 10 million on how to get this business really big. Because we are a team that goes big or goes home. Big for us or for anybody these days, 100 million plus in ARR, that’s the milestone we headed in 10, five to eight years, roughly. How do you get there and what needs to happen in the early days to get you there, to accelerate? You can’t just say the distribution will compound. The product has to somehow compound so people buy more of it or more consumption, etc. Because we don’t believe in humans as being the main lever to success for our customers, we decided that we’re not going to charge per seat for any of our products because most of the company’s legacy wants to charge per seat because they assume more people in sales say more sales that they do the company. We’re charged for consumption. How can we make Our users consume more of our product.

– Frank Sondors

We have to make them more successful. It’s fully aligned. If our customers are not successful, then we’re going to see churn hitting us. It penalizes the product team to do a better job. That’s why consumption-based pricing is super valuable, I think, for any business. Today, businesses have such a conviction that they’re actually moving even from consumption over to outcome-based pricing. Let’s say in the world of sales, you would charge per demo book, I don’t know, $400, $500, for example. If you have a lot of data that suggests that’s possible, no problem. In our case, we still see companies that come to us, they don’t have a product market fit. Sometimes they don’t have a product channel fit. They want to try a product. It costs our business or any business to run anything when it’s outcome-based. You need to have internally data that suggests that when any customer that comes through the doors and we have a certain degree of conviction that’s going to work out, sure, charge for outcome-based pricing. But there’s also a lot of debate there. But again, the pricing model, you have to think about how the product compounds, the hiring strategy, all of those things is what you need to think through to get you to 10 million.

– Frank Sondors

Then I think the other topic that’s not been really spoken about a lot is ensuring your cofounders have the right competence to get you to 10. Internally, you should be honest with yourself, whether you’re a CEO or CTO or whatever, in the business, whether you have the competence to get to 10. A lot of people try to make it a 10, but they’re not being intentional about their weaknesses as an individual and trying to do something about it. In some cases, what makes sense is that you get a sparing partner. Let’s say you are a CTO and you have never been CTO, like our CTO. Maybe it makes sense that you get another CTO, the sparing partner, that has gone through this journey. You have to be intentional about this. Meaning this is super important because if you don’t feel like you’ve got the competence and support in the right head and mindset to get to 10, then you have a risk, essentially, as a company. How do you de-risk yourself? Super important, right?

– Joran

This is really great advice. Let me try to summarize. If we go all the way to at the beginning of the conversation, you can’t get rid of humans. You need that. You make them more powerful by combining them with AI. So learn humans how to work with AI agents. If you talk about all go-to-market, you will hire 60% dead wood, so not real salespeople. It’s not sustainable anymore. When you look at the new go-to-market, you will allow your people to focus more on selling, like talking to actual customers. They will become more efficient, kill the repetitive manual work they don’t like to do anyway. I read GTM engineer doing things at scale by connecting APIs. They’re a bit more hacky people with commercial mindset. So it’s not always your developer you already have in the team. Assemble workflows to add to AI agents, feed them all the information from all the different departments, all the conversations they have. A/b testings. For example, you’re testing now agent versus human really soon after working hours. But do keep humans in the loop because they have a lot of nuances which AI agents don’t have. And do things which might not be that obvious, like being able to then actually reply rates when doing a message via WhatsApp, for example.

– Joran

But be aware, agents need to have the right context to have valuable conversations. If you love this, how to get started, start first of all with ChatGPT. If you haven’t used AI at all, hire a freelancer to set up your first flows. Just down what can be automated takes a lot of time. Go through that list and then give them the task to actually automate it with the goal being, after you must speak to your customers. If you’re going from zero to 10K MR, do things that don’t scale, sell before writing one line of code, speak to your customers, a lot. Frank did set 27 calls on one day. Activate your network, ask for referrals, leverage communities, host 10K going to 10 million ARR. You don’t need a lot of employees until 2 million ARR. Find a team which can get you to 10 million ARR. Hire fast, fire faster. By default, you can even fire everybody after three months, and they would only stay if everyone agrees. Hire autonomous employees, don’t babysit, and then focus on things that actually compound your business. Think about pricing based on consumption, value they’re getting. We did an episode with Kevin Bence on that recently, which was really interesting.

– Joran

And be honest with yourself if the founders are the right people, competent enough to really get you to 10 million AR. Find sparring partners, derisk yourself on getting there. All your words. I might do one point actually leverage Nail to automate it, but for now, it’s all my new work. Thanks, Frank. I know you’re not on LinkedIn, so I probably know the answer. But if people want to get in contact with you, what would be the best way?

– Frank Sondors

Yeah, LinkedIn, Frank Sanders, the guy with the bacon emoji. You can’t miss me. It comes from the saying bringing home the bacon. In Europe, we say bring home the bread, but bring home the bacon, that’s what we do. We have to do that more with the least possible humans. I’m on next. You can check out the website. If you want to talk to humans, you can on the site. Feel free to check out what we do.

– Joran

We had a conversation on LinkedIn and you made the joke, it’s not AI automated, which made me think, is this actually AI automated? Talk to him on LinkedIn and figure out if it’s actually automated or not. Thanks for coming on. For people listening on Spotify, please leave us a review and answer the poll so I can learn if you actually enjoy this episode or not. It’s going to give us data to make the episodes better. And if you We haven’t done so on all the platforms, leave us a beat you so we can burst the algorithm there as well. Thanks again for coming on, Frank. Thanks for having me.

– Frank Sondors

Cheers.

– Joran

Thank you for watching this show of the Grow Your B2B SaaS podcast. You made it till the end, so I think we can assume you like this content. If you did, give us a thumbs up, subscribe to the channel. If you like this content, feel free to reach out if you want to sponsor the show, if you have a specific guest in mind, if you have a specific topic you want us to cover, reach out to me on LinkedIn. More than happy to take a look at it. If you want to know more about Reditus, feel free to reach out as well. But for now, have a great day and good luck growing your B2B SaaS.