A Product Market Fit Show | Startup Podcast for Founders
00:00:56 · Pablo Srugo, Mark Hughes
[00:00:56] S1: You raised a $25,000,000 series a last month. How many people were you when you raised that? [00:01:01] S0: 12. Honestly, I was lucky and fortunate that we got a term sheet within six days for that first week. Process was actually quite quick, had the calendar completely stacked. Like, I'm a big believer in, rather fundraising or not fundraising. There's a lot of, like, oh, I'm doing these coffee chats and, like, hopefully, we'll get preempted from this coffee chat and all these it's, you're fundraising, you're fundraising. So calendar was probably blocked with, you know, 12 meetings a day. That's product market fit. Product market fit. Product market fit. I called it the product market fit question. Product market fit. Product market fit. Product market fit. Product market fit. I mean, the name of the show is product market fit. [00:01:37] S1: Do you think the product market fit show has product market fit? Because if you do, then there's something you just have to do. You have to take up your phone. You have to leave the show five stars. It lets us reach more founders, and it lets us get better guests. Thank you. [00:01:51] S0: Mark, welcome to the show, man. Thank you very much for having me. [00:01:54] S1: Let's start with the thing that really matters. Let's start with product market fit, and then we'll work our way backwards from there. Tell us, for you, like, when was the moment when you felt like you'd found true product market fit and why?
00:01:54 · Pablo Srugo, Mark Hughes
[00:01:54] S1: Let's start with the thing that really matters. Let's start with product market fit, and then we'll work our way backwards from there. Tell us, for you, like, when was the moment when you felt like you'd found true product market fit and why? [00:02:06] S0: Yeah. I think the moment we found true product market fit, there was four of us. So it was myself, my cofounder, and the two kind of early employees. We were in our very small annex of a a building office in in Dublin in Ireland. And up until then, we hadn't had much usage, no usage in the product at all. And we onboarded a new customer called Partner Hero, their kind of BPO outsourced call center who have a lot of conversation volume. They were using it in their Partner Hero Academy for onboarding new agents into the business. So we had left work at whatever time in Ireland. The company is based in America, and we came into work the next day, and we checked the post hoc metrics and the the usage metrics in the tool, and we were running around the room celebrating because it was was definitely a pinch me moment where when you see the graph flat for so long and then a massive spike where we had kind of thousands of simulations, as we call them, used overnight and hundreds of users in the platform, lots of errors and the the tool breaking, but they still kinda push through and use this. And, yeah, I remember my cofounder and CTO, Patrick, being very emotional, tears in his eyes, all the work he put in, that someone was eventually using something that he'd built.
00:03:19 · Pablo Srugo
[00:03:19] S1: Well, that's the moment when, you know, the thing that you've thought about and and you thought people get value a certain way, then you see people actually, like, in reality, get that value. It can be transformative, like a very zero to one event. Maybe tell us a bit more now that we understand that moment, what it is that your company does and what your company did for that customer.
00:03:35 · Mark Hughes, Pablo Srugo
[00:03:35] S0: So Solid Road is an AI first training and quality assurance platform for enterprise CX teams, customer experience teams. There's two components to the product. We have a conversation simulator, which allows agents to practice realistic conversations in a risk free environment and get coaching and feedback. So that's across all modalities, live chat, email, video call, etcetera. And then we have our production environment where we give coaching and feedback on live interactions. So we connect to your help desk softwares like Intercom, Zendesk, Salesforce Service Cloud, and then we close the feedback loop. So we will find issues in production and then automatically create these trainings to close those skill gaps. We also score your human agents and your AI agents so you can benchmark them against each other. So if you're using a Sierra or a Decagon or a Finn, we can see where you should actually deploy your conversation volume. The easy analogy I always give is when you bring a call center and you hear this call is being recorded for training and quality purposes, historically, that was very laborious. It would be human auditing transcripts with a rubric only getting to 1% of that conversation volume and then doing follow-up classroom based learning. We built AI agents that completely automate that process. So a 100% coverage of all your conversations so you have full visibility, and then, as I mentioned, closing that that improvement loop for your human and AI agents. [00:04:56] S1: And this customer that you were just talking about, when was that deployment?
00:04:56 · Pablo Srugo, Mark Hughes
[00:04:56] S1: And this customer that you were just talking about, when was that deployment? [00:05:00] S0: That would have been early twenty twenty four. So I think it was about March 2024. [00:05:04] S1: So that's, like, two years ago. What part of the product were they most using back then? [00:05:08] S0: That was just the simulation tool for so actually practicing conversations like AI role play for customer support, and they were using it in interview loops for new hires and then they were using it for onboarding, and that was really the kind of the pinch me moment that it was being used so heavily during this Partner Hero Academy, which was for just new hires. So to reduce that ramp time, time to proficiency or time to competency as it's often used in the the contact center world, they're ramping agents like in 50% of the time and using half the people, and we're like, okay, this is super valuable stuff. [00:05:41] S1: And you mentioned that before I mean, you started, what, a year before that? Like, twenty twenty three? [00:05:46] S0: Yeah. Incorporated the company, I think it was January 2023. [00:05:48] S1: So it'd been, like, a year or so. And you mentioned before that it was flat. Was it just that you were building, or did you change something to have that click, you know, product market fit moment?
00:05:48 · Pablo Srugo, Mark Hughes
[00:05:48] S1: So it'd been, like, a year or so. And you mentioned before that it was flat. Was it just that you were building, or did you change something to have that click, you know, product market fit moment? [00:05:58] S0: For a big portion that I was building, so it was my cofounder Patrick building the tool. And this was, like, very early, like, GPT 3.5, first version of ChatGPT, building our own speech to text before we could use tools like VAPI and 11 Labs even when were very early to even using 11 Labs. So a lot of this stuff was like, yeah, it wasn't fit for purpose. We have some funny examples of people on pilots screaming at the recording because the AI kept interrupting them, so lots of swear words in there. So we have them, in the the archives to look back on, but it really just the product really wasn't there yet, so that was probably the first six to nine months. And then we went after the sales vertical because my background was sales. I was an early sales hire at Intercom, and we applied the tool for, like, AI role play for sales. Got some early customers there, then end up actually firing those customers because what we found was even in the customers we did close, they were like startups that really didn't have the problem. And then when we went more upmarket to larger companies, it was actually the support org that was using it more than the sales org. Our background being ex Intercom, and that's where I met my cofounder Patrick. Gotta have domain expertise in customer support and CX. Kinda ran hard and fast there after that partner here on moment.