Building Outreach to a $4bn unicorn and helping AI agents get paid with Paid.ai - Manny Medina, Co-Founder & CEO @ Paid.ai

0:00Manny Medina:Welcome back to Riding Unicorns. Today we have Manny Medina, founder of paid.ai and previously founder and CEO at Outreach and I think is now an exec chairman role maybe. Manny, you can tell us what the status is there. But it's great to have you on Manny. Thank you so much for joining us. We're going to start with asking if you could just explain what an agent is because we're going to use the word agent a lot in this episode. And then what paid.ai is building to help advance the agentic era. Yeah, so an agent is a piece of software that does work autonomously, that you're sort of expecting to complete a task that was otherwise done by a human.

0:40And it's done autonomously or with the input of a human being, but the input of a human being is not necessary for the agent to perform his task. that's it so like if you can take a body of work performed by a human and completely replace it without the human intervention with a piece of software we call that an agent so what is paid.ai doing to help advance the agentic era so before i answer the question i need to answer the question like how do we came about pay.ai and why does pay.ai exist so when i was building agents at outreach, I realized that there were two things that were happening. One is that the agents were performing way more tasks than was allowable by the seed monetization that we're in.

1:25Meaning the agent was booking meetings, generating pipeline, running compete playbooks. They were running expansions and all those things were actually replacement, partial or complete replacements of our role. And I remember this call clearly with the CEO of DocuSign. And he asked me, Manny, how many fewer people do I need next year to hit the same number? And I realized that, you know, I was building my own contraction event. And this train is only moving in one direction. And I couldn't capture the value that I was generating with these agents. And when I asked my team to look into all the ways we can monetize, they came back and everything is difficult in terms of the monetization stack that we had built on top of the financial stack that we had built.

2:20So that was insight number one. And then insight number two was the variability of the cost of running those agents. So agents consume tokens, but we charge on a fixed rate. So there was clearly a gap between how customers are used to paying for a recurring revenue service versus how agents actually work. And that framing didn't exist in the world of SaaS. You pre-buy all these credits with LUS, you can buy the machine, you can buy the storage, there's all these things that you can do. But tokens were rather expensive and unfungible. So the third piece of the equation was that they run in the background.

3:04So now you're looking at a piece of software that runs in the background that doesn't need a seed. Who is going to success that? Who is going to tell the customer of all the work that you have done for you? Because there is no user experience for these agents. So how do you run a pilot and show success? And how do you run the business and equip your CSMs or your account organizations to make sure that they are showing their customer that you deliver this value? So there was just so much wrong with the world to that point that was not matching where the business was going that I decided to just do something about it.

3:37I think before we get into the meat of this episode, we should just hear a bit more about your career, Manny, because most of our listeners, if not all of them, would have heard of outreach and the huge success that it was. But, you know, you scaled outreach from nothing to, I think, maybe over 250 million ARR,$4 billion valuation. like tell us about that journey yeah so it has milestones so we went from zero to two in arr 2 million in arr 2 to 10 million in arr 10 to 25 million in arr 25 to 50 50 to 100 and then you know all the way to 250 are they are you talking about each year is a year milestone yeah that's that those are years the year milestones and and every year is sort of like a different chapter, a completely different chapter.

4:26So zero to two, you know, I was selling a lot of early days of outreach door to door and then hire, you know, my first VP of sales. And he came in on commission only because I didn't have money to pay him a regular salary. So imagine the conversation that he has to go to have with his wife that like, hey, babe, I'm about to take this job that doesn't pay me any salary. Meaning if I don't make a sale, I don't get paid. So it It was early days and you have to believe in what we build and you have to believe in the potential. And eventually he became our VP of sales, but it was exciting. Selling something that didn't exist, that was zero to two.

5:06And then two to 10 was all about building the category. So what do you call this thing? It's a workflow for generating pipeline. What is it called? And there were other workflows already that were sort of like different parts of the elephant, if you will. Like there was a workflow for email. There was a workflow for calling, for dialing. That was inside sales.com. And then there was, you know, ToutApp and Yesware. But they're all sort of were touching different parts of the workflow. And our thesis was that generating pipelines, a numbers game. You see what I mean? And that there is so many people you're going to talk to.

5:37And then there is a conversion point. And then you have to continuously A-B test your language and your touch points and your timing and all these other things. And there was no engine for that. Like all the engines were built for marketing. so you know the journey from zero to ten was about building the category and building the awareness while you hit the 10 million dollar revenue 10 to 25 and 25 to 50 like at that point the upward draft of the category creation just took us with it and like 25 to 50 was mostly making sure that wheels didn't fall off the track because we were growing so fast what year was that when are we talking roughly ah shit i don't even remember when did it start 2013 is when we pivoted.

6:152014 is when we built it. 2015 is when we launched it. So I guess 15 to 16 was the first 2 million. 16 to 17 was the 2 to 10. So yeah, that was the journey. And then after 100, the complications are different because it's a scale game. I remember when we hit I think it was 25 or 15 million that we got a valuation of half a billion dollars. We did a at a$500 million valuation, Series C, I think it was. And we're like, yeah, we're half a unicorn. I mean, and I used to joke that we're like, we're the back half of the unicorn, the one that looks like an ass. We need to complete the picture. You know what I mean?

7:00So, yeah, it was, you know, the journey of outreach is the journey of category creation. I don't recommend it, by the way. Like, I think it's far easier to build something that already exists and just build something easier, better to use. like I don't know something like light for for ERPs or like linear for CRM first for for project management or like they are at you for CRM I think it's easier to replace something that exists but because when you have to look at the market that this thing doesn't exist yet you need it I mean you're you're you're doing hard miles yeah not not for everybody you say that the scale-up journey was about making sure the wheels don't fall off and and I'm sure that's true I mean from the outside, the numbers were amazing.

7:42The product, I assume, was delivering. What were the wheels that could have fallen off in that stage? I hate this. I hate this. And I still see startups doing the same thing. So if any startup is listening to this, write this shit down so it doesn't happen to you. So when the upward draft of the market or whatever segment you're in starts taking you to high growth, you have to do the honest work. and pull apart. What is you and what is the market? And that shit needs to keep you up at night. Why? Because the market will change. And if you don't have control over that portion of the revenue, you don't understand clearly how is the diet is coming in.

8:25The moment it changes, you're screwed. And it's going to hit you in a bad way. So why was the wheels falling off the truck is because you know i fought against it you know constantly that's how we got to 50 before i left is that every time that revenue would show up just show up like we call it we used to call it intra-quarter revenue and i start batting the name then we called it we call it uh open and close i forget exactly what the name was was creating close we had a term called created and close revenue which meant created and close in the quarter we i i was maniacal about minimizing and they're creating close revenue.

9:04Why? Not because I don't like money. It's because I don't like inefficient money. That money will show up and as easily would leave. And you have to understand clearly why did it come in? What problem are you solving? So if somebody's coming in and screaming, take my money and shut up, you don't take it. You actually, you have to put some friction in the deal to understand why is it they're buying it? Who are they buying it for? What problem are we solving to make sure that we equip that customer to be successful over the long term. You know, before we started recording, you opened up by saying, you know, we're going to talk about fast growth companies, right?

9:40Like fast growth is everything about growth. And I sort of disagree with that coming from the other side, being in my first war and now moving into my next one. I can tell you this, that fast growth comes and goes. What really matters to make people rich is durable growth. And durable growth looks completely different than fast growth. Durable growth means that you are a workflow that learns, that you are part of some kind of operation that is existential to the business, that you're a system of record, that you're something. You see what I mean? That allows you to be durable over a long period of time.

10:15That's what everyone should be building. You see what I mean? Because that's what investors pay for. Not fast, durable. You must have all sorts of thoughts on the lovables, the bolts, You know, all of these AI agents, they get rapid adoption. But, you know, we've probably all seen the sort of retention or heard the rumors of low retention, low gross margin. What are your thoughts on that? You know, when there are these potential winner takes all markets with huge prizes, investor cash flowing in and growth at all costs is the mentality right now. Like, would you avoid that? Look, I'm not qualified to advise people do or not do, because I want to make sure that we don't confuse tactics from strategy.

11:04So if your strategy is to own the coding platforms and your tactic is to price below your cost of operation, that's a fine, because you have access to capital, that's a fine way to play it. You know, matter of fact, Amazon played like that for a very long time in the public market, right? They were losing money hand over fist and the market turned. They couldn't react fast enough. And people forget that they have to get a lifeline from Berkshire Hathaway in the form of a convertible node just to survive the downturn. And it turned out to be, you know, it worked out fine. So does the tactic work?

11:41Of course it works. Is that a sustainable business? Of course it's not. But, you know, I don't know Anton. I don't know what he's doing. I'm sure he's enjoying the ride. But what I would advise the founders is to make sure that you understand that that's a tactic and not a way of living. You know, selling something that costs a dollar for 50 cents is not a business. It will get you notoriety, but it's not a business. Yeah. So before we move on to your journey with paid, what would you say? You know, you said you were maniacal about this sort of quality of revenue. And maybe that's your answer to this question.

12:13But what would you say that you guys did best in that zero to 50 million revenue journey at outreach? Oh, by far was category creation. And by far was being the fastest of shipping innovation as part of that category creation. So we created the category. So the category took a long time. So one of the questions that people ask me, and you haven't asked me this question, I'm going to answer anyway. It's like, what is category creation? So when do you know that category has been created? And there's no real answer, right? It's sort of like, it's kind of like porn. and you know where you see it. So it's really hard to tell what is category creation.

12:46But one of the easy ones is when an analyst calls you category what you've been calling it. Eventually, your customers are calling you the category that you invented. And that's when the category is established, if you would. So what we did right was spend thoughtful time and intention in this category creation journey. So we had a lot of discussion with my co-founders about do we create a new category or do we attach to an existing category? So there was an existing category called Revenue Acceleration that was owned by InsideSales.com. The question is, are we a better InsideSales.com or are we something different?

13:28And I'm a disciple at this point of the Play Bigger guys. And I'm a disciple at this point of a guy called Mike Maples. You've probably heard of him as a partner of Floodgate. And he is all about being different. you know you have to be different different is a point of view different is a different take on the market and i you know a lot of companies are built after like the founders on taste and my taste is not i don't want to go compete for better i want to go compete for a different and different i can own a hundred percent of whatever the size of that differentness and we were the first of the market we were the largest sales engagement platform and there's a bunch of others that came in and in it came and out they went and there's a few that stuck the landing like self-loved but when we established it it was just us so that we got to own all the benefits of being the largest in a category that you invented so we did that really really well and once the category took off because of sdr growth we were the de facto so you either pick us or you have to make a choice that is not us and that's where you want it to be

14:34Manny Medina:and now with paid ai presumably it's kind of a similar feeling you're creating the category of a new way for agents to price and run payments and everything maybe you could explain why the old model doesn't work for agents again i mean you touched on it briefly around almost tapping into labor budgets rather than software budgets but maybe you could go a bit deeper on that? Yeah. So, so let's start with monetization. The, in SaaS, you try to, most monetization is done via, via seed, right? And, and, and you have, you know, the software is deployed in a multi-tenant environment and you, you sort of like rent a seed on that software.

15:20In agency, every, it changes, it changes that dynamic because you don't need a seed, meaning one seed can operate a thousand agents. And the agents are performing the same work that a seed would do. So the seed all of a sudden doesn't become the economic measure by which your economic unit, by which you price the value that you're delivering. That's number one. The second thing is that the cost of running the agent varies based on the complexity of the task to some degree. And if you're not capturing the value correctly, that you're losing money. We're going back to like, you're selling a dollar for 50 cents.

15:58And again, that is not a viable business. So, you know, right now, everyone in the agent world is sort of in land grab mode. So they're all going out there, trying to grab as many new logos as you can. So not that many talk has been had around, you know, running the business profitably. But the tide has turned, you know what I mean? Like that is the one thing you can count for certain, that there will be a high tide and there will be a low tide. If you stand in the same spot by the ocean, you'll get wet. So that's what we're doing. We're picking a spot by the ocean and waiting for the tide to turn when it comes to margin management and when it comes to revenue optimization.

16:38And our twist here is a little different and just to finish the swing, is that I fundamentally don't believe that your back-end systems need to be disparate, that your rev rec and your CPQ and your quote to cash and your billing and your quoting and your margin management and your API management and your vendor management needs to be 20 different things. I am in the rippling camp that this has to be one thing because it's one problem, it's one record. And how do you look at that record and how do you use that record is how you build this company. Yeah, fascinating. Manny, how have you found the adjustment going back to early stage?

17:17Like getting back to being really hacky, moving super fast. You know, it's a very different job as you're leading a scale up to getting a company from zero to one. Has that been a tough phase? I think there's a lot of things to unlearn. There's a lot less people here that are at outreach. You have to make choices more frequently that are explicitly declared. And it's actually helping me undo a lot of my bad CEO habits. And now I see it in other CEOs of scale-up companies. So for instance, if I have an idea or if I want to do something and I run it by somebody else, you know, a person may tell me yes, right?

17:58But can we actually execute on what we have given that we only have, you know, 14 people on a team, 15 people on a team? And the most likely answer is no. That means that we're going to have to execute on the new idea. We're going to have to deprioritize an idea that is already on the plate to execute on the new thing. So keeping the team small as a constraint forces you to make hard choices, forces you to pick a winner, you don't have the luxury of running three bets at the same time to see which ones pay out. You see what I mean? Like, for instance, we are running a couple of bets simultaneously, and that's a choice.

18:30That's why we raise the money that we raise so we can have enough people to run both bets. But I can't run a third. You see what I mean? Like, I can only run those two things. And out of those two things, I have to optimize where we're going to deploy our cash and our people. One of the things that I did, actually, is get an office here in King's Cross, and the The office got only hold, I think, 17 people as a max. That's it. And, you know, I'm sort of like having this conversation with the team that I don't want to move offices because the constraint on this helps us stay sharp. You know, helps us, you know, prioritize things right and work on the maximum reward activity.

19:07So, I mean, a lot of startups and a lot of VCs are talking at the moment about constraining themselves, you know, because they want to force their team to adopt AI and make themselves more efficient. Is that something you're particularly keen for your team to do? And is it working or is it sort of more of a dream at this stage? We're sort of doing it, but I want to dig into that. So, I mean, you guys are investors. So if you give$10 million to a company, what do you expect them to do with that money? Like, where do you find the flex? Is it to hire more engineers? Or is it to, you know, take another bet?

19:41Or is it to hire more marketing? Like, where do you ask your companies to deploy their cash that you just gave? Yeah, I mean, hiring, marketing, sales. That's not an answer. That's not an answer. You're saying everything. Do everything. That's not an answer. What are you asking them to put those on?

19:56Manny Medina:I'd say it's people and products, just at the beginning. Okay. In the stage we're investing in, usually it's people and product. So two things, people and product. And not go to market, not marketing, right? So you're not going to go buy ads? I mean, not initially. I would say the next round would probably. It depends on how quickly. At the beginning, I would say I want to see more investment in product than growth. Because I believe in that durable growth piece, which is like go and build something really meaningful, really understand the customer problem and and like really get close to them and you know don't try and scale too early make sure you get really like you actually have genuine advantage in the market once you start to press on the gas so if i am if i am if i take your money and i blow it on ads on on on times square you're gonna be mad at me right i would say yeah i'd be annoyed about that yeah probably at the stage we're investing it yeah you might have a laugh about it but yeah i mean once everybody's gone it's gone bro like they start coming back uh yeah i mean that would definitely at the stage we're investing in that would definitely be uh a jump too early okay so so yeah i mean that that's that that's kind of it like you know you you have to it's not people is like people where and doing what?

21:21And what do you expect? You know, you're running milestones, right? So what is your next milestone? In your case, you guys are early stage. So presumably your next milestone is a funding round. Is that how you think about it? Or is your, or you want to do the last? Well, I mean, the funding round is a product of the milestone. It's a milestone in itself as well, but you need to achieve milestones to get the funding round. Right. But that's what I'm saying. So what's your hypothesis coming into a company? Is that a funding round? or it or it's now you know it's revenue yeah you know the the thesis ultimately is around revenue you know do we think that this solves a human need that enough people are suffering from

22:01Manny Medina:and can make people money save people time got it got it yeah i mean obviously the next funding round is a a score it's a mark it's marking some of your homework or some of your work but it's it's only as a step to the next phase so it's not really you know the job's not done but it definitely if the company does a massive up round then you're thinking yeah we we did a good job there basically but you know you're still years away from realizing the full potential so got it got it there are stepping stone so you're not coming in that's not your hypothesis coming in You say, I'm going to invest here and I'm not expecting you to raise again.

22:46I'm expecting you to drive revenue. And if you raise again, that's just, you know, high for yourself around.

22:51Manny Medina:Most of the companies we're investing in, and I'm sure Hector feels the same, we would expect them to raise multiple rounds because they're getting on that hamster wheel of venture. And so they're going to, but so they've got to hit some milestones. but really it's you know we're expecting them to build a great product and those milestones kind of naturally come through and then they've got to we've got to believe that they have the right profile and the right ability to attract more capital further down the line uh whether they hit five million arr or one million arr but you know as long as they've got the right products they can still raise that next round so it really and it depends on the category the person there's there's a lot to it the business model we do consumer and b2b so it's it's very variable but yeah people and product is where the most of the capital is spent got it yeah so this is this is my point of view and this is just you know one man's view is that there's only two ways a product grows either in especially in b2b either by self-serve adoption people call it product-led, or by selling.

24:07If you're selling and you need to throw more dollars to hire your reps, ramp them, and get them to sell, it better be an enterprise sale, and you better clear the 4X target of like OTE versus quota. If that criteria is not met, you should not drop a dime into the company until you hit that criteria. Or you're self-serve, in which case, go for a girl and just keep growing. Anything that doesn't land in one company or the other, it's not a durable business. You just stop doing whatever you do until you become a durable business. That's my criteria now for growth, for pay.

24:48Manny Medina:And which one's been more successful? Has it been doing things like having an amazing domain, which gives you great brand recognition? which then drives good you know self-adoption or is it being more hit the phones let's get the word out there we're the we're the dominant force in this space so where we are right now like to like at this point in time meaning you know wednesday you know 3 p.m is that we haven't launched officially out of surf surf we're finished we're building it so in that case we're you know finalizing the product and making sure that the product has a fit to the sales sort of needs of agent builders.

25:32And for the sales, for the part that we're selling, we haven't hit efficiency yet. Meaning my deal cycles are not predictable. My ACV is too low for where it needs to be a successful sold product. We just closed a deal that hit that mark that was within the sales cycle time that will be acceptable and within the ACV band that will be acceptable. Let me pump$2 million into that within those constraints and I'll tell you that I'm winning. But right now, I, you know, now that said, at outreach, it took me, you know, two years to get this dial, right? So I know a bunch of shortcuts now to how to get there more quickly, et cetera.

26:14But in terms of all the efficiency things that I want to see before we scale, we're not quite there yet, but I can see the light at the end of that tunnel on how we get there. Like I now have a path on how to get there. I didn't have a path like almost a week ago. Now I can see it.

26:29Manny Medina:Oh, only in the last week. That's cool. So what really changed? Was it just having a customer hit those numbers or did you change something to help define who that customer should be to then go, okay, that's what we need? No. So in every organization that is doing a considered purchase, meaning this is not a purchase that you have to make, on a recurring basis, but it's a purchase that is something that you're sold because you identified a pain, et cetera. You have to find what the persona is that allows you the quickest sell cycle, right? And for us, I finally figured out, which is hilarious to be frank, that the quickest sell cycle to sell paid into an organization that is going to buy paid as opposed to sell served is the commercial team.