Scale Like a CEO

Letting Go & Leading with Gil Allouche | Scale Like a CEO

• Justin Reinert • Season 1 • Episode 55

Share episode

What if the fastest way to scale isn’t doing more, but doing less yourself? We sit down with Gil Allouche, founder and CEO of metadata.io, to unpack how he turned a hacked-together experiment engine into a platform that automates B2B performance marketing end to end—audiences, creative, execution, and optimization to pipeline—so leaders can spend time on strategy and message instead of chasing broken UTM tags.

Gil takes us inside the hardest leadership pivot: moving from hands-on founder to true orchestrator. He shares a simple, repeatable handoff pattern—“I do you watch, you do I watch, then you do”—that builds confidence without dropping standards. We explore how to respond to mistakes without judgment so people keep experimenting, why waiting to hire creates bad choices, and how to recruit leaders who already share your pace and values. Gil makes a strong case for promoting from within by letting rising stars shoulder responsibilities before they get the title, so the promotion feels obvious rather than risky.

Culture shows up most when stakes are high. When metadata.io doubled down on AI, Gil chose radical transparency: one-on-ones with every team member to assess appetite for the journey, clarity on workload and expectations, and respectful exits where needed. That honesty avoided quiet resistance and aligned the group around action. We dive into the operating rituals that sustain psychological safety in a remote company—quarterly off-sites, weekly town halls, monthly pulse surveys, and a confidential advisor who surfaces issues early. Gil closes with two force multipliers for any founder: find a coach who can call your bluffs, and surround yourself with peers who model the CEO you want to become.

If this conversation sparks ideas for your team, share it with a founder who’s ready to grow, then subscribe and leave a review to help more builders find the show.

Speaker:

Welcome to Scale Like a CEO. In this episode, host Justin sits down with Gil Allouche, founder and CEO of metadata.io for a deep dive into the realities of scaling a business. They discuss the critical transition from founder to CEO and examine how leadership and automation drive company growth.

Justin:

Thank you so much for joining me on Scale Like a CEO. Just to get us started, if you wouldn't mind, give us a quick intro to yourself and your business.

Gil:

Right, it's Jesse, I'm happy to be here. My name is Gil Allouche. I'm the founder and CEO of metadata.io. I've been running this company for about eight years. I'm originally from Israel and I'm a software engineer in my background, robotics and AI. Spent almost a decade in the United States post-my graduate school running marketing for B2B companies. Metadata was the impetus of combining the two, my software engineering and AI and marketing.

Justin:

That's great. Tell me a little bit more about metadata. You founded it nearly 10 years ago and have been growing it. Tell me a little bit about what are the problems you're solving at metadata and why is it important?

Gil:

Yeah. As a software engineer running marketing in a few B2B companies, I can tell you I had a very different background than the average marketer. I was no Done Draper sipping whiskeys and doing cool presentations. I was more on the technical side, trying to figure out which campaigns to run and which audience segments to go after and what experiments I'm going to execute. I ended up tailoring some kind of system together using Zapier and spreadsheets and webhooks. I had a system that I tailored together from my job that ran my experiments for me. And then I didn't have to choose. I didn't have to rely on an agency and I didn't have to rely on some technical team to develop this for me. I was able to do it myself, and it worked out pretty well. Like the three companies I worked for, we were able to generate demand constantly and predictably for the sales team. That was basically my job and the job of every VP of marketing in the B2B setting. That's exactly what we solved. B2B companies today, I think our customers are Zoom or Noche, Cisco, IBM, some mid-market enterprises, they spend tens of millions a year on advertising. They have a large team, usually an agency as well. And they end up with, let's say, 10,000 possible campaigns you can run at any year. They have multiple products, multiple businesses, multiple locations. There are a lot of operations that are involved in running performance marketing function. And our technology essentially takes over the full spectrum of tasks. Anything from setting up the audience targets, creating the creative, the copy, actually creating the ad, the different types of ad on the different channels, Facebook, Instagram, LinkedIn, Reddit, X, AdWords, et cetera. And then executing the campaign and reaching the leads and then optimizing them for pipeline. All of those tasks usually take a lot of time, a lot of people. And we essentially automate the entire thing from start to finish so that people can, instead of waking up at 3 a.m. in the morning, panicking about the UTM tag break or something that is totally technical, repetitive, and mundane, they can focus on the fun things like creative ideas, the type of campaigns they want to run, the messages they want to test, the copy, maybe the creative, but not have to worry about all the small tasks of execution that are constant. Our system does something like 100 different optimizations per compent per day. And that's something that a human would never do. No human is expected to do it. And it's not very interesting to change the bid from 200 to 203 and 40 cents, but the system will do it for you, and it's worthwhile. There's a lot of benefit in doing that automation, and a computer is best fitted to do it.

Justin:

That's incredible. I find modern advertising is so fascinating because there's so much that goes on behind the scenes that people have no idea of how that ad is getting in front of them. So it's amazing the work that you're doing. I want to shift a little bit and talk about when you have been growing your company. So as you started from being the founder or the founding team, um, and you grew, what were some of the shifts that you needed to make personally as you shifted into that kind of scaling CEO?

Gil:

And I could talk about that for days. There's a lot there. Absolutely. Team, and first and foremost, the CEO and the founder have to change and adopt and grow. And when you don't, there is a lag time for the company to grow, right? CEO and the founder, you haven't changed on time. Guess what? You're just now prolonging the change and adoption for the entire company. And so there are a few things. First, I think is moving from doing everything to orchestrating and supporting everyone else and removing roadblocks and friction. It took me some time to get there, let me tell you. It was not an easy one because you can do a task maybe really fast as a founder, right? That's partially our competitive advantage. We're really fast and experiment really quickly. The problem is that you now expect someone else who does have your context, they also don't have the same level of equity, not the same incentive, right? And you expect someone else new to exactly as good as you did or higher. Sometimes that's expectation before you hand over some task to someone else, and that's not the right approach to it because it's not gonna happen. Then you have to flip the switch on that one and let go of this task and let them maybe be a little different in the beginning and kind of take a breather and wait and improve and give feedback, and over time it's like it saves you a lot of time and scale, but not in the beginning. So that's one thing that I remember I struggled with. Another one is for example, when you go through sometimes the graphs on technology is like this, but the actual graph, I can't do it with my hands, but it's like this, and those dips are a shit show. No one talks about, but then suddenly the retention is cut by half, suddenly you have like a month of payroll left on your cash, and you don't forget those things as a founder, and sometimes you remember them too much, to the extent that if you're lucky to overcome them because of fluck and because of a few good decisions, you're now stuck with the memory of that cash running out or the churn increasing. But it's your job to forget about it when you're done with it. When you're done with it, your job is still good, awesome. Now, retention is great. I have plenty of cash in the bank. Daily reminder of these numbers anymore. It's gonna take a lot for my founder CEO, and I think you will have to rely on other people, especially new people who are not around for this and are around for that, to rejuvenate you with energy so that you're ready to go the next step. These are just two examples. I have a hundred more.

Justin:

I have no doubt you have many, many more. You're right. That up and down, that roller coaster can be really challenging. I want to go back to the first role that you talked about, though, because that is, I think, something when I'm talking to founders that comes up so often is shifting from doing it all to getting things done through others. So I'm curious, what was the hardest role for you to let go of and what made it hard to delegate?

Gil:

Probably product for me was hard to share because you're the one who started it. So you think the original vision still holds, but there is other stuff now. There are customers who are paying for this product. So you have to figure out how to balance this with some other ideas you have. That's also one of them. Another one was account management and customer support because you're obsessed. I had the first customer of two people in a garage, but I used to be obsessed about their level of happiness, right? With the service, I think handing them over, especially things that you used to do every day, is difficult. It's so hard to do. So once you give it, you accept it, you almost do it against your will, although you're the one forcing yourself. I think the way that is better is perhaps do I do you watch, you do I watch, and then you do on your own, and I'm gonna help you. Do it much more gradually and not release control completely, but just set them up for success. So not wait for something to happen for you to be unhappy and not be at the bar of expectation, but actually guarantee that as much as you can that they're gonna have their own wins, that they're gonna feel like they are capable and are able to take on this task from you. And that means being very auditing it and being with them and joining them on calls. And also, I remember a call where I gave something, it was the demand plan, which I've done many times. And I had a head of marketing who was doing the plan, and he had a mistake in the plan. And I that I did it, I did it okay. I pointed the mistake, but without judgment, and just told him, hey, this is how I would fix it on the spot. I remember after the call, he told me, it was really nice. I did I was worried, but then I wasn't worried when it wasn't a big deal. We just fixed it on the spot, how to do it better next time, and we just moved on. And if you can do that in this kind of demeanor, then you're gonna get the trust of the team and they're not gonna be afraid to take on and tell you when things are not working and when they need your help. They don't feel like you're gonna judge them and be able to do that.

Justin:

Yeah, I love hearing that that's how you dealt with that mistake, because too often we punish people for their mistakes. And what we've learned from brain science is that when we punish people for their mistakes, it actually makes them less likely to try again because they're afraid of failing. And so when we can teach that mistake or that failure as learning, no judgment, as you say, and move forward, then you're really setting your people up for success in the future. So that's really great.

Gil:

I'm I felt it both ways. When I did make them that mistake and judge, and you can see there is no risk of earnest to trying, then it's on you as a founder, on the manager, if you caused that. So that's something that is really critical. Um, I think.

Justin:

Yeah. And now, um, Gil, you've grown the team beyond a hundred, right?

Gil:

We grew beyond a hundred. We're smaller now. We went through a process where we implemented a lot of AI. So we're a smaller team now. Okay.

Justin:

Okay. Great. But yeah, so you've grown quite a bit. I'm curious, I'd love to talk about by the time you get to that size, you've got to get some good leaders in place. So, how did you go about identifying and then developing your leadership team either internally or did you hire from the outside? Tell me about how you approached that.

Gil:

We did both. I think I've seen two things that worked for me. First of all, I tried to have a coach at any point of time to help me with some of these hiring. Did help. Just first principles about hiring. For example, I used to be very reactive where I would wait, I needed someone immediately. And then really depends who I have in my pipeline at that moment. I would start interviewing them for the role. Sometimes I like to answer your question with a mistake first. So this isn't how I'm not to do things. The way, one of the first principles I adopted is the moment even before I need someone, start interviewing for them and start defining what is it that I'm looking for. Hopefully, by the time you need them, you already have the pipeline, you already have people, even if they were not in the market, etc. So that that proved to be very helpful for me. I was able to bring some really good people. Another one is bringing people that that other A players already worked with before, relying a lot on people that already know the way of the company, the culture, the pace, the values, etc. Bringing someone who already fits that in the company, bringing people that they worked with before. I am a huge believer in bringing in people from the existing team. We just promoted two people from within. I think what I like to do there is one of the risks of why white people sometimes don't hire from an internal team is that they're afraid that that person is doing such a good job at their current job. If you hire them, if you promote them, they're not gonna do as good of a job in their new job. And now you have to hire someone else to do the previous job. And that is the fear of many managers. The way we try to deal with that is first of all, set it set off the stage to say, hey, you're an amazing developer or you're an amazing marketer or you're an amazing account manager. I don't know if you're gonna be a great manager of account managers. So you just set the stage because it's true for everyone, it's true for me, right? Like we could go without within a size that I am actually not a good fit to the CEO. I need to bring someone else. So that's one. That's I think the expectation. The second one is let them occupy their role before you promote them. So essentially tell them, hey, you're doing an exceptional job. I think you're gonna do an exceptional job here, and there's more for you, but I don't know yet. So can you start taking the responsibilities? I'm gonna observe, and if we're good, then you're gonna achieve the role. I've done it a few times, even this year, and that has proved to be very, very successful. Also, by the time you give them the promotion, everyone is wondering. I thought that they were already in this position. That's a good sign. That means, first of all, it's super easy for the culture to accept, but also it means that they already did all the due diligence on themselves and kind of got there on their own. Well-deserved hire.

Justin:

Yeah, it's great when you can get people to kind of tiptoe into that new role and get comfortable there. Then there's no surprises when they're promoted and they can ease into that successfully. What are some of the other things you do, other than kind of getting them to try on that new role? What are some other things that you do to prepare your leaders to lead their people?

Gil:

We're a remote company. I encourage them to do off-sites often, at least once a quarter. Everyone meets together in an Airbnb and just work together from the same room. Some human connection is very helpful there. That's one. Second, we have a person who is like the shrink from billions for lack of better description. And he's basically guaranteeing he has a level of intimacy and connection with all the staff to make sure that if there is something, then we know about it, even if it's anonymous, and we can deal with it and we can float it all the way up. And all the way up is usually one or two people maximum. So it should be very, very, very flat. We've done those things and they've helped the leaders. I think so. Making sure that if someone is in a risk of burnout or there's something they don't agree with, instead of keeping it inside, we do a monthly poll survey also and questionnaire. We do a weekly town hall for people to speak up. I really encourage there is no meeting that I don't ask people and name them and ask them for feedback and ask them for the questions, the comments, disagreements. And last but not least, let the examples where people are brave to speak up, let it be an example. So instead of shutting it down, thanking for the feedback, addressing the feedback, all the most of the cliche that you can imagine, they are very effective. Then you make the work of the leader much easier. Okay, he has a team that he's not afraid to speak up. They know that he can always reach to me or to someone else if there is a problem. They don't feel like they're isolated by their boss only. And then they get a chance, they get many opportunities in person in real life to talk to their boss and to their team. Those three things, plus the down halls and the surveys, they get you to a pretty good level, I would say, of trust.

Justin:

That's great. I love the things that you're talking about are directly tied to developing psychological safety within a team or within an organization. So that's just incredible. And according to Google's research, psych safety is the number one determinant of success for a team. So it sounds like you're making really great steps at building that. So how have you defined the culture that you want to build for the future of the organization?

Gil:

I think a lot of it is I've learned not in the about a stage company statement and culture, but in the people that you hire and in the way that you deal with the problems. When there are issues, how you deal with those issues, how do you address them? And between that and the people that you hire, I think that sets the tone for the culture. For example, when we made one of the moves of integrating AI in it meant people had to work harder, more time, more wear multiple hats, and as well as automate a lot of the work. And some were done for it, but not everyone. Before I made that transition, I had that dilemma. Do I just do it and hope that I selected the right people to do this work among the team? Or do I go and interview the team and ask them if they want to do this and assess whether they really want to get into this journey with me? Of course, a risk-averse approach tells you absolutely don't do it. If you do it, half the people are gonna just resign. They're gonna hear that you want to do this. Doesn't matter what they say, they're gonna look for a job immediately. Basically, that's the advice I got from everyone besides a founder who did it. And he told me, actually, I recommend that you talk to all of them and honestly tell them exactly what you're planning on doing. Then you're gonna have a conversation around the reality. I did it. I was scary. I remember it was a lot of meetings, one week, 15 minutes or 20 minutes each. The cool thing about it is that there was no surprises. I knew who's in, who's not, and the people who are not could get a server and something nice and you know, taper off, and maybe do a contract work, or maybe even stay on a different kind of okay, this is defined role. And the rest were super hungry to do it, were really interested in using AI for everything and just going crazy about all the innovation, were in for the right reasons and were able to make that change. So I think many times the people that you choose is wrong, and I would have a mixed group of people who are interested, uh super risk-averse and don't want to do this, but are doing it because they were forced to, and some were interested in doing it, it will create a mismatch in the culture right there by keeping very different types of people, but with similar mentality of not making decisions out of fear, seeing a bigger opportunity ahead, super critical of everyone, including myself and the organization, and speaking up, but not trying to do anything perfect. Those kind of values that took us a long way, maintaining it by hiring similar mindset and constantly talking about it. And if you're about to make a change, you have to be aggressively direct about and honest and super transparent about the way you're doing it.

Justin:

Yeah, that transparency is key. And it sounds like you had a really great approach to the way that you had those conversations and got folks ready for the future. Um, so one last thing. What's one piece of advice that you'd give to another founder about scaling themselves as a leader?

Gil:

Find a coach. I always need to find my first coach was someone that I really looked up to, and I borderline stalked them until they became my coach. That changed me a lot because then you have someone you look up to, you build trust with, and they can call you out on your bullshit. They'll know this is they tell you, right? You listen. So that's one that in a very meaningful way got me to a better place as a CEO. The second one is surround yourself with the type of founders and CEOs you want to be. So surround yourself, go to the meetings, go to offside, go hang out with your peers, and put you in the right mindset and also take you up to the next level as a founder.

Justin:

Well, that's some great advice. And thank you so much for that. So, Gil, if folks want to get in touch with you, what's the best way to do so?

Gil:

They can shoot me an email. It's gil at metadata.io. I'm always happy to talk to founders.

Justin:

Great. Well, Gil, thank you so much for the time today. I really enjoyed the conversation.

Gil:

Thank you, Justin.