If you only have time to read one deep-dive this week on modern B2B growth, make it this one. In our latest Ignite Podcast episode, we sat down with Max Greenwald, co-founder and CEO of Warmly.ai—a platform building AI agents that warm your total addressable market (TAM) and route only the hottest leads to sales. A former Google PM (yes, the “Where’s Waldo” Maps prank that ~100M people played) who has steered six pivots before breaking through, Max brings a rare, field-tested view of how demand is actually generated in 2025.
Below is a full written breakdown for readers who may not catch the episode—complete with frameworks, playbooks, and practical takeaways you can put to work immediately.
TL;DR (Skimmable Takeaways)
Warm demand beats cold outreach. Use AI agents to initialize relationship and intent before your team ever emails or calls.
Multi-signal > single-signal. Blend website behavior, firmographics, tech stack, channel touchpoints, and recency to score buying readiness, not just MQLs.
Open/Close for PMF. Deliberately open to explore new segments/solutions; close to double down on a narrow wedge once you see real pull (inbound, referrals, fast sales cycles).
Dogfood relentlessly. Warmly sources a big chunk of its own pipeline by running its product on itself—shortening feedback loops and sharpening the scoring logic.
Omnichannel or bust. The modern buyer pinballs across channels. Orchestrate consistent, helpful moments wherever they surface.
The Origin Story (and Why It Matters)
Max’s path: Princeton CS → Google PM → founder. The early venture years weren’t linear: rejections, co-founder turbulence, and six pivots—without changing the company name. The lesson isn’t “pivot more”; it’s pivot intentionally. Max’s team learned to separate idea vanity from signal quality, and to treat PMF as a sequence of experiments rather than a single epiphany.
“Signals, not vibes. Inbound interest, shorter sales cycles, and word-of-mouth are the scoreboard—not the story you tell yourself.” — Max Greenwald
From Anonymous Traffic to Booked Meetings
Warmly’s evolution mirrors how demand gen has changed:
De-anonymize & observe: Who’s on your site, which pages matter, and how often do they return?
Turn conversation on: Live chat → AI chat to capture and qualify in real time.
Retarget with brains: Not just pixel-chasing—use context and timing to surface relevant prompts.
Score with multiple signals: Stitch behavior, firmographics, timing, and engagement into a single readiness score.
Route instantly: Push only the best leads to AEs with context, while nurturing the rest automatically.
Outcome: Less spam, more meetings. Sales trusts marketing because they feel the difference in their calendars.
The Open/Close Framework for PMF (and Go-To-Market Fit)
Most founders oscillate between wandering and tunnel vision. Max’s answer:
Open Phases: Widen inputs (segments, use cases, price points). Look for surprising pull: unsolicited demos, referral chains, faster closes.
Close Phases: Narrow everything (ICP, message, channels, packaging). Write “not-for” rules. Instrument the funnel and operationalize what works.
Cadence: Time-box each phase and define the exit criteria (e.g., 3 straight weeks of >30% demo-to-opportunity conversion in a target segment).
This rhythm protects teams from infinite exploration and from premature locking.
The Demand Gen Playbook (2025 Edition)
1) Warm the Market Before You Pitch
Use AI agents to greet, guide, and qualify visitors—on site and across channels.
Offer contextual value (micro-audits, fast answers, tailored content) before asking for time.
Quick win: Deploy an AI concierge on your high-intent pages (pricing, integrations, case studies) to offer a 90-second “fit check” and one-click meeting booking.
2) Build a Multi-Signal Lead Score
Single-event triggers (e.g., one ebook download) are noisy. Combine:
Firmographics: Size, industry, territory
Technographics: Key tools in their stack
Behavioral: Pages viewed, depth, recency, return visits
Channel: Email replies, social touches, events/webinars
Context: Problem-specific pages, integration interest, pricing views
Rule of thumb: If your “hot lead” definition fits in one sentence, it’s too shallow.
3) Orchestrate Omnichannel Moments
Buyers hop between website, LinkedIn, G2, webinars, and email. Meet them with consistent narrative and next step wherever they show up.
Awareness: Short problem clips, proof snippets
Consideration: Integration demos, ROI calculators, teardown posts
Decision: Customer stories that mirror ICP pain, live Q&A, fast trials
Guardrail: Every touch should make the next step painfully obvious.
4) Route with Context, Not Just Priority
AEs need more than “hot lead.” Deliver a one-sheet: why now, what they looked at, similar customers, suggested opener, and a 3-bullet discovery plan. The handoff is part of demand gen.
5) Dogfood to Learn Faster
Warmly runs Warmly on Warmly—capturing real conversations, false positives, and timing cues. Internal usage produces the sharpest tuning for scoring, prompts, and routing.
Try this: Weekly “signal review” between RevOps and PM. Pick five wins and five misses; update rules and prompts that week.
Inside the Stack: Where AI Adds Lift
Code generation: Faster iteration on experiments and internal tools.
Sales assistance: Summaries, email drafting, call prep—grounded in the multi-signal profile.
Support enablement: Faster, more relevant help so prospects don’t stall pre-demo.
This isn’t “AI replaces SDRs.” It’s AI reduces waste and raises the floor on every touch.
Metrics That Actually Matter
Meeting rate from high-intent pages (and time to book)
Demo-to-opportunity conversion (per ICP)
Opportunity velocity (days stage-to-stage)
Pipeline sourced by owned demand (vs. rented channels)
AE acceptance of routed leads (qualitative trust + quantitative accept rate)
If these move, the MQL count can be… background noise.
What’s Changing in MarTech (and How to Adapt)
Lean in-house teams, more agencies: Keep strategy and data close; outsource execution sprints where needed.
Content goes atomic: Short, high-signal pieces repurposed across channels with AI assists.
Agentic workflows: Helpful, brand-safe agents that learn over time and feel like teammates.
“In a few years, agents won’t just answer—they’ll own outcomes with the same playbooks your best reps use.” — Max
90-Day Action Plan
Weeks 1–2: Instrument
Identify high-intent pages and set up session recording + basic de-anonymization.
Define current “hot lead” criteria (even if imperfect).
Weeks 3–6: Pilot
Launch an AI concierge on top pages with a short “fit check”.
Create the AE one-sheet template and automate the handoff.
Weeks 7–10: Score & Route
Roll out multi-signal scoring; test 2–3 thresholds for “instant route” vs. nurture.
Start weekly signal reviews; adjust rules promptly.
Weeks 11–12: Scale
Layer in retargeting with context (use the pages they touched).
Publish 3 proof assets that match your top ICPs; wire them into agent prompts.
Founder Notes: Leading Through the Messy Middle
Narrative discipline: Keep a running doc of “not-for” segments to avoid silent scope creep.
Team rituals: Monthly gratitudes, weekly “what surprised us” reviews—low-friction ways to stay human and curious.
Capital efficiency: Brute-forcing channels is out; precision and sequencing are in.
If You Only Do One Thing After Reading This…
Add an AI concierge to your top two high-intent pages with a 90-second fit check and one-click booking. Measure meeting rate and time to book for two weeks. The clarity you gain on true demand will change how you score, route, and plan content.
👂🎧 Watch, listen, and follow on your favorite platform: https://tr.ee/S2ayrbx_fL
🙏 Join the conversation on your favorite social network: https://linktr.ee/theignitepodcast
Chapters:
00:01 Intro and guest setup
01:12 Princeton to Google path
02:45 Leaving Big Tech to found Warmly
04:10 Early rejections and resilience
05:30 Co-founder dynamics and roles
06:50 The Open/Close framework
08:05 Defining ICP and not-for list
09:18 Six pivots, one mission
10:40 From de-anonymization to conversations
12:00 AI concierge on high-intent pages
13:22 Multi-signal lead scoring
14:45 Routing with context to AEs
16:10 Dogfooding and feedback loops
17:35 GTM systems and repeatability
18:50 Omnichannel orchestration
20:15 Content atoms and proof assets
22:00 Metrics that actually matter
23:30 Pipeline quality over volume
25:05 Hiring the first great AE
27:20 RevOps cadence and signal reviews
29:10 Buyer intent vs interest
31:00 Demand gen without waste
33:05 Agents as always-on SDRs
34:40 Near-term roadmap and vision
36:06 Closing takeaways
Transcript
Max Greenwald (00:00:00): It’s called the death triangle. It’s called the Bermuda triangle of you’re trying to convince investors to give you money. You’re trying to convince customers to buy a product or to find an idea. And you’re trying to convince co-founders to come join you. So I think in the very earliest score, it’s like co-founders, investors, you know, idea. And each leg of that triangle wants the other two to be perfect, right? It’s like a VC wants to know that you have an idea and you have co-founders. A co-founder wants to know you have funding and you have an idea. And for your idea to work, you’re going to need, you know, co-founders and VC funding. So it’s like just that brutal triangle.
Brian Bell (00:00:53): Hey everyone, welcome back to the Ignite Podcast. Today we’re thrilled to have Max Greenwald on the mic. He’s the co-founder and CEO of Wormly.ai, former Google PM and founder who’s navigated multiple pivots in pursuit of product market fit and someone deeply invested in how AI is reshaping GTM motions. Thanks for coming on the program, Max. Great to be here, Brian. Well, I’d love to get your origin story. What’s your background?
Max Greenwald (00:01:15): Yeah, I grew up in Denver, Colorado, where I was a bit of a mountain kid, loved to ski. Found myself out at Princeton for university where I studied computer science. And then I heard about this company called Google. And I was like, man, that’d be a cool place to work. So I joined Google a little too late, 2017, there for about three years, essentially just doing corporate infighting and not getting anything done. Free food, great perks, feel like I was learning at all. And so then I left in December 31st of 2019 to start Warmly, my startup, where I’ve now had the pleasure of working with my co-founders across four or five different countries through remote work and ups and downs, trials and tribulations. And now we’re five years in to our business. So that’s a bit about my background.
Brian Bell (00:01:58): Amazing. So kind of the standard APM track out of a CS undergrad, go work at the big tech, get a few years experience, realize it’s all politics and pushing paper or virtual paper around. Kind of the same experience I had at Microsoft, right? Just playing Game of Thrones and trying to get, you know, things done by committee and screw that, I’m going to go do a startup. What was the aha moment for you? You were like, let’s do this, let’s take the plunge.
Max Greenwald (00:02:20): Yeah, well, first off, like no knock on anyone that goes to those big tech companies. I actually feel like one of the hardest things when you’re 22 is figuring out how to do life. And so if you have a cushy salary and you don’t have to do much work and you need to like do the hard stuff of like, how do I get a dentist on my own? And how do I work with a roommate? Well, and like, how do I, you know, exist as a 22 year old? Pretty good gig. And I definitely needed a few years to kind of figure that out for myself. But I would say I do coach most new grads to go work at an early stage startup these days. I feel like the learning is just so much faster. Because I think the biggest risk in life really is not learning. The most money you’re going to make in your career tends to come in your 50s and 60s. And so that means learning as much and as fast as possible in your 20s, 30s, and even into your 40s. And so when you go to a big company, you’re just not going to learn as fast as possible. And for me, I think I was just like, yes, two years in, hadn’t really like launched anything. And the most fun that I had was my now co-founder of my startup, Karina and I, she was an engineer at Google. We worked on April Fool’s Day pranks for Google on the side. And so we launched a Where’s Waldo scavenger hunt in Google Maps that 100 million people played in spring of 2019. or 2018. It was just so magical to kind of hack something together and pull in people, work nights and weekends, whatever. And I was like, that’s cool. I want that feeling of like building something unique. And that’s what kind of got me excited about the idea of building my own thing. But I’d be lying if I was saying I was mostly, you know, I was just like excited to go start warmly. The truth is I was excited to run away from Google. And I definitely don’t recommend starting a company just because you’re running away from something. But I had to get the kick in the ass somehow to get out there.
Brian Bell (00:04:00): It’s an interesting take, right? Because a lot of us, when we quit jobs, are running away from something, unless you have this idea that’s burning bright in your head at night and you have to work on it. But would you advise people in their 20s kind of considering making that plunge just to get going or to wait for the idea and just get out of the building and start talking to people and figure out what their problems are?
Max Greenwald (00:04:22): So I would say that the majority of founders in their early 20s are going to end up building companies that sell to other small companies, just statistically, at least in the beginning. And so you get a leg up if you know what it’s like to be in a small company. And I think that you can also understand and appreciate smaller company problems more. And so probably the answer is go leave as soon as you can to go work for a smaller company, ideally an early stage startup, so that you can kind of understand what it’s all about. before you actually have or that’s the best way to find the big idea yeah i mean look there’s different ways to do it the other way to do it is just i’m a fail fast kind of person i’m just going to run through walls and i’m going to just figure it out so i’m just going to leave i have no idea but i just know that i want to build something or i want to like let’s call it self-actualize through like building a business and i’m not sure what that business is going to be and thus i’m just going to jump out the you know the proverbial window and like try my hardest to get it done one downside to that is i think you end up starting a you’ll more likely start a venture-backed company because you probably don’t have the capital to get started and there’s a whole bunch of pitfalls that come from taking venture dollars and great things as well but it means that you’re probably excluding a series of businesses that may be like long-term profitable and you know a good solid kind of lifestyle business by kind of just jumping off and doing it on your own right away and like needing cash to continue So there’s a trade-off. I have a lot of friends who go work for an earlier stage company to try to find ideas, and then they bootstrap something on the side, and maybe they seedstrap, right, where they raise a little bit of money and then go profitable, or they just bootstrap and then build a good lifestyle business on their own, which is generating a million dollars a year in profit, which is Phenomenal. And then I also have friends who, from a personal runway perspective, were like, shit, I need to raise fast. So I got to get out there, pick an idea, maybe go through a YC kind of thing, and then rave venture dollars because I just need to get going and have cash to do something. And then they joined the VC treadmill, which I’m on. And I think parts of it are great and parts of it are not.
Brian Bell (00:06:19): That’s interesting. And I want to unpack that comment for later. So you left Google. Did you go to YC immediately or were you kind of working on an idea and then?
Max Greenwald (00:06:26): No, I got rejected from YC the first time around. Two people that I thought were going to be my lifelong co-founders that I’d love forever agreed to quit Google with me in November of 2019. We had our first version of the idea, which was the dumbest idea in the whole world. It was Tinder for co-founders. So our first idea was, it’s hard to find co-founders. What about a little swipey app where you could like meet co-founders? Now, clearly I was not suitable for that because the two people that I picked to start that business with are not my actual co-founders today. And so two of us, we submitted our resignation for Google. And the third one, upon learning that we didn’t get into YC, was like, oh, oh yeah, like, you know, I’ll submit it tomorrow. And then tomorrow became the next day and the next day and then they never left. And so there’s like still someone sitting at Google now six years later who swore to me they were going to leave their job to start this business and I’m still salty about it. Anyway, so I left with one person. That didn’t work out either. But it was sort of, you know, in the early days of starting a business, you’re stuck with this, like, I’ll call it the Bermuda Triangle. I was going to call it the Death Triangle. It’s called the Bermuda Triangle of you’re trying to convince investors to give you money. You’re trying to convince customers to buy a product or to find an idea. And you’re trying to convince co-founders to come join you. And so- And early employees, but you got to get money first. So I think in the very earliest core, it’s like co-founders, investors, idea. And each leg of that triangle wants the other two to be perfect, right? It’s like a VC wants to know that you have an idea and you have co-founders. A co-founder wants to know you have funding and you have an idea. And for your idea to work, you’re going to need, you know, co-founders and VC funding. So it’s like just that brutal triangle. And so when I didn’t get, we didn’t get into YC, you know, kind of like spiral down. Anyway, a couple months later, we applied to Techstars and Techstars took a chance on us. So we were working on the idea, you know, no pay, no dollars in the bank with my new co-founders for... maybe three months. And then we got into Techstars Boulder, which was great because I got to have a homecoming back to Colorado. And that was the beginning of everything. I mean, I’m so grateful to Techstars for taking a chance on us as many years ago.
Brian Bell (00:08:27): And it’s the same company that became Warmly or is that that was a different company?
Max Greenwald (00:08:31): We’ve always been called Warmly. It depends how you look at it. But there was, I would say, six major pivots in three years until we landed on what we do today. What we do today, I would say, is like a light combination of everything that came before it, you know, kind of compounding learnings time after time. But yeah, six pivots in three years, always named warmly. And it worked out great because now we sell warm leads. So it’s like perfect.
Brian Bell (00:08:51): The name actually ended up working out. What were some of those pivots? And when did you know you needed to pivot versus persevere?
Max Greenwald (00:08:57): Yeah, pivot versus persevere is one of the hardest questions. I have founder friends who have persevered in their space and have been huge. I have founders who have friends who’ve pivoted 15 times and then they’ve made it and they’re huge. And I equally have founder friends who’ve died because they persevered and died because they pivoted until death. Meaning they didn’t literally die. Right. But they’re good point. Sorry. They’re still around and doing just fine. Yeah. No, they didn’t die. I hope so, too. Yeah, they’re doing good. So there’s no magic formula. Pros and cons to both. We fell into the camp of the founders who were going to pivot fast. But I think that one of our lessons was that each pivot got subsequently longer. So our first pivot was like three weeks, then six weeks, then six months, then two years. And then what we’re doing today. And I think that like one of the frameworks and by the way, you know, any founders out there listening, like I didn’t know any of this shit when I started. I mean, I was just a fucking idiot running around trying to like do anything to survive. I’m only retroactively applying these frameworks. So if what I’m about to say sounds smart, which I think it sounds smart. It’s not that I knew it back then. I just sort of. learn it the hard way and apply the framework to it. But I call it the open close framework. And the idea is that you go through periods in your search for product market fit of being open and being closed. And an open period is where you’re open to any new ideas. And basically you and your co-founders talked about it. Like, you know, we can continue on what we’re doing now, or we can try some new stuff. Do you have any ideas? Let’s like kind of figure it out. And that’s usually like a few weeks. And then you pick an idea to go closed on and you say, all right, For this pivot, we’re going to do a closed period of a month. And come hell or high water, we’re going to pretend, we’re going to put our blinders on and pretend that it is this idea or bust. And we’re going to give it everything we have. And any time throughout this month that we come across a new angle or a different idea or wake up in the middle of the night sweating because we have some genius brain blast, we’re not going to chase that because it’s too easy to get distracted. So we’re going to have a piece of paper and we’re going to write down all of the cool ideas that we’ve had that we’re going to revisit when we go back into an open period. And so you finish out your closed period, your month’s in. We’re going to enter into an open period again. We’re going to reopen. We’re going to look at all these other random ideas we had in the meantime. And we’re also going to look at the progress we’ve made on that one idea for a month. What are the good? What are the bad? Let’s recognize the grass is always greener. And you can decide, should we go down to another closed period on the same idea? Or should we pick a new idea from the list? And the only catch in the open-closed framework is that each closed period has to be subsequently longer. Roughly, you can just say maybe you double the timeframe. So one month becomes two, two becomes four, four becomes eight. But the idea is that you know that when you enter into a closed period, it’s going to be longer than the last one. And it forces you to eventually get to the place where you’re building a sustainable business that you’re going to stick with and kind of force yourself into a persevere. Hopefully that’s a helpful framework, but that’s sort of what we retroactively... realized that we did. We had a closed period of two years. And despite that closed period of two years, we still opened back up and pivoted. And now we’ve been working almost three years on this business. And it’s night and day.
Brian Bell (00:11:49): That’s amazing. What a story and great framework for any founders listening out there to apply. I’ve never heard that open-closed framework before. That makes a lot of sense. When did you know that you have some real PMF here and it’s not time to open this back up?
Max Greenwald (00:12:06): My Cedar investor, James Currier at NFX, says that product market fit is like when someone shoves two fingers up your nostrils and yanks your head forward. And if you haven’t had that feeling, then you’re definitely not at product market fit. I would massage that a little bit to say like you’re seeing some inbound coming in. Yeah, some word of mouth virality, et cetera. But I think like some version, I mean, you have to find product market fit for the product. You got to find market fit. And then you also like go to market fit that I think a lot of people leave out, which is like, Are we building, do we see repeatability in our sales process? Like, can I hire an AE and kind of understand how much like, if they’re a good AE, if you’re doing, sorry, if you’re doing B2B, you know, sales with an account with like, you know, people, but they’re going to have repeatable sales. There’s that kind of aspect as well. So a lot of depth to it, but I think that we knew we had it because everything else that we tried, we had to convince someone to buy. And this was the first time where a decision maker who had buying authority and budget to spend was wanting to buy because they knew it was important. I will say that there are some products that you’re inventing in a new category, like that you believe will become mainstream in a few years and you’re ahead of the curve, which is awesome. But it’s quite risky. Like our pivot before the current business that we spent two years on was called Name Tags, Name Tags by Warmly. And we were the first ever app on the Zoom App Store. And you probably never heard of a Zoom App Store because Zoom did a terrible job marketing it and because Zoom’s security settings made it almost impossible to install an app. But back then, we thought it was going to be the next iPhone store. We thought everyone was going to download apps there. And it was basically a solution to help you look more professional on your meetings, targeted towards salespeople, where you could put your name, your title, your pronouns, what the weather was in your location. You could put an agenda over your shoulder. It was kind of like a productivity slash professionalism type tool for your virtual meetings. And we had to basically convince people to buy it. And we told ourselves... This is going to be a huge category. It’s COVID. Everybody’s going to need to look more professional in their Zoom calls. It’s critical. And it turns out that everybody wanted it, but nobody would pay for it. And so getting people to pay was just like pulling teeth to get them to pay. And we do like six meetings for like a 1K deal. And it was like, yay, but also this is not working. And so when we eventually realized that no matter how long we sort of persevere in this space, that category is not going to come around. And by the way, one of the best ways to know that you’re not in a category that people want to buy is that there’s no competitors. And so we couldn’t find a competitor to this product. And we were like, there’s got to be somebody. And we’re like, we’re so lucky. We’re like enjoying world domination. But it’s like, yeah, world domination of a market that doesn’t exist at all. our investors suggested we try to find the red, the red ocean. He called it, which was like, try to find an area where like everybody wants to get a bite of your dollars and invite to the market because there’s a ton of players. And of course, you know, being an extraordinarily competitive space has its pros and cons as well. But that’s where we, we meandered into the red ocean. And that’s what all of a sudden we’re like, wow, like people want to buy this thing. People are coming inbound and you got that feeling of somebody sticking two fingers up your nostrils and just yanking.
Brian Bell (00:15:10): I love that. So what was the aha moment for Warmly? What does Warmly do? And how did you guys come up with that idea?
Max Greenwald (00:15:17): Yeah, well, I’ll talk to you about how our tagline has changed over time. But today, my answer is Warmly builds AI agents for marketing teams to warm up their total addressable markets and deliver hot leads to sales. The word AI agents, of course, didn’t exist a year ago. So that’s why I’m putting a little asterisk there. But we’ve always set out to generate warm leads for our customers. And the final form of Warmly was a combination of several of the earlier pivots we did in the three years prior to us landing on this idea. But where we started, the wedge that we started with was helping you generate more leads from your website traffic because we de-anonymized your website traffic and told you the names and emails of people who came by your site but hadn’t actually converted yet. And then we alerted your salespeople in Slack to say, hey, you should go after these people. And then from there, we evolved to add a chatbot on the website where the salespeople could talk to those leads on the site. Then we added an AI chatbot where the chatbot could talk to those leads on your behalf. Then we added a retargeting solution where if the chatbot misses the people who you talk to on the website, you can automatically follow up with them by email and LinkedIn by logging into those sales reps accounts to do the work on their behalf. And then we added additional signals. So more ways to find warm leads where we’d find leads on your website. We’d find leads by scraping the internet. We’d find leads by monitoring social media sites and as well as from your CRM if you hadn’t followed up with them in a while. So now we have lots of leads, lots of ways to reach out to them. And we have an AI lead scoring product or agent that basically helps you make sense of this big mess that is go to market. So we can help demand gen marketers say, okay, of all the leads we could chase, this is the best damn list. These are the best leads that we go after. Here you go, sales team, you go work them. And then my second best list, the list of leads that my sales team can’t quite work. I can automate reach outs to them across email, LinkedIn ads and chat to basically warm them up and get them ready. So that when they pop into and show up in the hot list, I can pass those over to sales.
Brian Bell (00:17:14): Yeah, it makes a lot of sense. And as we were discussing before the call or before the recording, I built a lot of MarTech, you know, used Marketo and brand demand gen team. So I’m very familiar with MQLs and SQLs and trying to warm up those leads and score them. And I was there back when you had to basically devise your own scoring mechanisms in Marketo. Yeah, that sucked. Design these like little recursive algorithms like, oh, add a point, you know, point one on this and add a point on that. And very, very familiar with this.
Max Greenwald (00:17:41): What a lot of people don’t realize is that one of the common problems between your sales team and your marketing team is your marketing team’s job is to get leads ready for your sales team to work. And so marketing will, let’s say once a quarter, send over a list to sales and be like, these are the good ones to go after. And then sales will reach out to those leads. Some of them won’t reply. These leads suck. What are you talking about? Yeah. And then sales will upset. And so then marketing said, okay, well, fine. We’ll add a score at each of them to show you the hottest ones, the least hot ones. And then you’d be like, well, why is the score 100? And you’d be like, well, it’s really complicated. You wouldn’t understand. It’s a crazy formula. And then you’d reach out to the 100 lead and it wouldn’t reply. And you’d reach out to a lead that says 50 and you’d get a sale and you’d be like, well, I don’t trust this lead scoring. So that left room for software like Wormley to use AI to add value.
Brian Bell (00:18:23): It’s very cool. So what were some of the biggest operational challenges as you started to get in this product market fit and you’re starting to scale revenue? So now you’re kind of in, you know, this closed open seek kind of mode trying to find product market fit for years and years and years. And all of a sudden, OK, we have something. How did that, what are some operational challenges for you as you kind of started scaling the team, the hiring, the structure, the culture, all that stuff?
Max Greenwald (00:18:46): Yeah, I mean, first off, I’m going to pat myself on the back and just sort of acknowledge the fact that it was fucking hard to get out of the pivot hell. I went through this startup founder program called On Deck, built by Eric Torenberg originally. It’s a phenomenal founder program. And one of the things that Eric talks about is something called the idea maze, which is basically the search for product market fit. And you can spend a long time there. And anyway. This is the same situation that happened to us. And so the pat on the back I’m giving myself, I think is well-deserved because my God was it. But it’s just so sucky to be in this constant phase of getting started on something and then taking two steps back and then trying something new and taking two steps back and just being like, how is everybody else able to figure it out but me? And obviously it’s not true. People are not figuring out all the time. People are not, they’re not, they’re not dying with their startups. We’re dying left and right. And you know, you still can’t figure it out. So when you finally figure it out and turn the corner, you’re almost starting from scratch, but it’s almost a whole new set of skills you have to learn. I mean, I got so good at building a landing page and spending $4 on an Instagram ad to drive eight users to the landing page to measure the conversion rate of a tagline. That is a skill that is really not that useful when you’re doing a real go-to-market motion, but it’s critical in the early phase kind of cohort analysis, figure it out kind of vibe for your early ideas. Anyway, so we figured it out. Now we got to be able to go to Market Motion. I think we’re at like 250K in ARR. And we’d experimented over the years once or twice with hiring a salesperson. Never worked out. And I basically got a reach out from a marketing advisor of ours. And she said, I have this great sales guy that I know. He has an offer from one of your competitors, but he hasn’t taken it yet. Do you want to talk to him? And I was like, absolutely. Like I want to win. Yeah, I love to win. And so we talked to this guy and he’s phenomenal. His name’s Keegan. He’s our head of revenue. And so we started talking, we get along really well. And I’m like, oh wait, I don’t even have a job opening open for you. Like I think that as a founder, I should wait until we get to like 500K of ARR and then maybe I’ll bring on my first sales hire. And he basically tricked me into hiring him. I was a good sales guy. Came in as sort of like our, let’s call it founding AE, but really we were testing him to see if he could prove out the motion and then he could hire a team underneath him. And we just kept going from there. You know, that was, I want to say, 24 months ago. And he was able to crush quota. I was able to crush quota of what we expected an account executive to do. So then we hired some salespeople to report to him. And yeah, I think each quarter we kind of iterated on a new motion, right? So we tried, and each quarter we were swapping out our tech stack. You know, we went from like DocuSign and spreadsheets to like a real CRM and like a quoting tool. You know, we went from posting on LinkedIn to using a professional firm to record me to then make professional videos. We didn’t do any, we didn’t go to any events. Then we went to some events, we didn’t get any booths. Then we went to some events and we wore a cowboy hat instead of paying for a booth. And now we pay for some booths sometimes. So we really just iterated on maybe like seven or eight different channels. and evolve them quarter over quarter, tinkering with new ones, killing some efforts, getting new ones working. But yeah, it was a new skill set all over again. But I would say from like zero to, let’s call it 2 million in ARR, we just had a sales motion. And then now from like two to like five and a half, we’re at almost 6 million in ARR. We’ve like kickstarted a marketing motion. And really what you realize eventually is that you can grow in B2B sales and B2B go-to-market motion. We saw the SMB, a little bit of mid-market. You can grow exponentially at first because like any number is better than zero, right? Or like basically growing 10% month over month on 100K of ARR only means you have to get 10K of ARR. But eventually you realize that sales is a linear function. And the more AEs you add, the more costs come out and then the more money you hopefully make, but it is kind of linear. where you really can break out. And I think is what our customers appreciate about our software is marketing. And so when we’ve tried to unlock that next growth curve to get from basically, let’s call it 2 million on ARR to 10 million on ARR, it has to do with develop marketing machine. And so we focus a lot of our efforts there.
Brian Bell (00:22:46): Well, and you know, what’s interesting about a MarTech platform like you guys is you can dog food and recursively use your own platform to grow your own platform. So maybe you could speak a little bit about that and how that’s accelerating your PMF and your revenue capture.
Max Greenwald (00:23:01): I can follow up and show you a graph of this later, but like 50% of our sales come from our own product. It’s pretty awesome, right? It’s like we use it, we like it, we get value from it. I was talking to our SDR manager last week. We booked four meetings through our live chat interaction. We booked eight meetings through our AI chat, a part of what we do. So yeah, it’s kind of nice.
Brian Bell (00:23:18): That’s really cool. What about, you know, what role has AI played in your evolution? You know, you started before, you know, AI was, you know, having its second coming, right? And now it’s in-
Max Greenwald (00:23:30): I owe everything to Mark Cuban. And if Mark Cuban’s listening, first, I want to say fuck you. And second, I want to say thank you, which is that Mark Cuban owns warmly.com. And when I first tried to get warmly.com, it resolved to let it was like it didn’t, you know, you type in the domain and it didn’t go anywhere. And then I found out through a domain broker that it was registered to Mark Cuban. And I was like, that’s hilarious. Why does Mark Cuban have my domain name? And it turns out Mark Cuban’s a domain troll and he has 10,000 five and six letter domains that he sells through his, this is one of his subsidiary companies. And I’m like, I guess that tracks. I mean, this dude will just fight you tooth and nail over anything. And so I reached out via three different investors. And each time he gave us the most ridiculous counter offer I’ve ever heard, which is he offered me, he offered to rent me He would rent me warmly.com for 3% equity in the business and 500 grand. And so if we died, he got the domain back. And if we IPO’d, he’d get some upside. I just was like, it’s like, I guess it’s insane. I can’t, I was like, I’m a little, I’m a little fish. I’m just the first time founder. Please give me the domain. You don’t need it. And he was like, nah, final offer. It’s like, whatever. So we ended up buying warmly.ai. And this was maybe four years ago. And so we’re like, AI is kind of weird, but like, whatever. Like, you know, it sounds nice. Who cares? And it’s turned out to be the most incredible domain name ever. AI really took off. And we were one of the early adopters of putting AI into our product, starting with our chatbot or our inbound agent. And now we have four agents. We have our data agent, our inbound agent, our outbound agent, and our marketing operations agent actually launching next week. And it’s a really exciting time to be embedding LMs into your workflows and to be... And what was cool about it was like, we had built a good amount of the tech before AI came out, really. And then it was like pretty easy to just kind of drop in to each of our different modules and make it so that our customers are getting way more value. So...
Brian Bell (00:25:16): You know, it’s funny, I won’t name too many names. I was just talking to a founder in our portfolio this morning, had the same exact situation on a very in demand domain name that he really wanted. And he signed up for one of these deals where it was like a rev share and all this stuff, almost exact same kind of ask. And he did it because he really wanted the domain. And now he’s rebranding the company and dissolving that C Corp to get out of that deal, basically.
Max Greenwald (00:25:40): Yeah, makes sense.
Brian Bell (00:25:43): So back to AI. So like you guys have been integrating AI all over the stack. Would love to get your kind of thoughts on kind of AI in the present, how you’re seeing impact marketing, your boots on the ground. I mean, my smart technology is five or 10 years old now. So I’d love to hear from somebody that’s out there on the forefront of AI and marketing.
Max Greenwald (00:26:00): Yeah, a couple of nuggets. I’ll first start just internal to the business and I’ll focus on broadly how I’m seeing it in the marketing space. But we really tried to push and adopt AI agents into the business itself because I do believe there’s going to be a stepwise cost reduction effort that can go on as well as a productivity increase. I think all this bullshit about the like 10x AI engineer or whatever is a little overblown. I do think there’s like three to four X improvements on how AI can fundamentally impact each of your departments. Which is pretty incredible if you step back and just think about that. Even 2x productivity improvements is incredible. I mean, we’re talking 3 or 4x. Sorry. I mean, maybe you sound like I’m jaded or whatever. It’s phenomenal. I mean, it’s fantastic. It allows me to be able to grow my business faster without needing to spend too much more money. Yeah. So the three main agents that we’ve adopted internally first is called CodeGen, CodeGen.com, our AI engineer assistant. And while our engineers themselves are using a lot of like cursor to help them code live, CodeGen is almost like the non-engineers is free engineer. So, you know, those like little features, like little things you want to change, maybe customer support notice is a bug, but you know, they’re never going to get fixed. CodeGen fixes them to 99% certainty and then puts the finished code in front of an engineer to review and then press submit. So, you know, a PM will usually feature or will gate the engineering time and say, don’t talk to engineer, don’t ask them about the problems here. But if we’re like, actually we have 99% finished seven or eight small tweaks, then the engineer just goes, yes, yes, yes, yes, no, a little tweak. And then yes, yes, yes. So that’s been a big agent and a boon for our product development. The second big agent we use is called Docket. Docketai.com is a sales engineer. So it sucks in every call recording we’ve ever had with any of our prospects. It sucks in our CRM information and it sucks in all of our Slack knowledge and all of our Notion knowledge. And I think it will drive too. And then it allows our salespeople to ask questions quickly that they wouldn’t be able to answer themselves that are like somewhat technical. And so it’s a Slack bot where you’re like, hey, like, you know, does our product do this this way? And it’s like, yes or no. And it’s quite accurate, which is cool once you train it for a bit. So that’s really helped improve sales. And CS used it somewhat as well. And then finally, we use Pylon for our support system. Basically, customer support, pretty vanilla use case. So just respond to customers using AI first, describe the fact that this AI answer could be wrong, and then humans read it when they get around to it, and then they can follow back up and be like, actually, this answer was wrong. Here’s the better one. I think the impact of this, interestingly, has been smaller teams. I think we’re going to see a huge shift toward onshoring, where I think we’re going to have startups who’d rather hire two or three people in per US based startups who would rather hire two or three people in person in their home office. to run a massive department of AI bots rather than have outsourced more affordable labor that’s abroad and remote, probably equally hardworking and equally smart, but just not with you in person. That would be much more cost affordable. So that’s an interesting trend that I think we’re going to see. But that’s a bit about how AI has affected our business. I’ll pause and see if you have any questions, then I can go on to external marketing. How AI is affecting the marketing landscape. Biggest thing probably is a near freeze on marketing headcount. So there’s really no need to hire more in-house marketing folks. I’m seeing a big shift back out to agencies. Used to be that agencies were quite expensive and you weren’t sure the quality that you’re going to get. But now because all the agencies are using AI tools for content generation, A-B testing, et cetera, they’re starting to commit now more to outcomes. So an agency might say, oh, you know, we’re 5K a month, no matter what happens. But now I’m actually seeing a lot of these agencies that are so confident in how they’re using AI to drive results that it’s like, you’ll only pay us contingent upon a certain number of leads we generate or contingent upon whatever certain deliverables. And so that’s brought about a bigger shift more toward agencies. And then also like just content and graphics and visual stuff. It’s, I mean, I’m sorry to say it, but the graphic designers are done. Video creators are done. The stuff you can do with Sora 2 and with VO3, we have one marketer who focuses on content entirely. And she’s able to... I think she’s a year into her marketing career. And she’s able to do the kind of stuff that would take budgets of 100 grand and six marketers like three years ago. So she’s just really... crushing it and building fantastic content. So that’s been another, you know, kind of huge thing is less marketers, more shift agencies and trying to bring in, especially the content agents. And on the demand side, where we focus, we’re obviously biased because we have our own AI agents that we think move the needle. But I think There’s going to be a increased necessity to go omni-channel, which means as the world spams harder, as more emails go out, more LinkedIn messages go out, more cold calls go out, everyone’s going to just ramp it up and spam 10 times harder because reply rates are down by 10x. And so in that world, you have to have marketers who are savvy enough to hit prospects across wherever you might finally catch their attention. So you need people who can do, you kind of got to do it all to be able to break through the noise you got to do. Events, ads, emails, LinkedIn calls, snail mail, B2B gifts, if you’re doing a B2B world, etc. So all of that translates to the need to use AI to automate and engage outreach omnichannel, which is a big piece of what Worldly does.
Brian Bell (00:31:05): Yeah, really fascinating. Let’s talk a little bit about the future. What are you excited for in the coming years, both for your business, but also for AI impacting MarTech generally?
Max Greenwald (00:31:13): I hope we get our agents to a place where you can actually have a conversation with them and they can like actually learn. I think that like context window is early. It’s like you can kind of like take one bit of context and then do the same action over and over again. Or you can take a lot of context and have general knowledge that isn’t like up to date with the latest. But we’re not really at a place yet where you can treat an AI agent like a teammate. Manny Medina, who’s the former CEO and founder of Outreach, I just started a new company called Paid. They announced their seed round last week. His whole thing is that by 2030, half of all employees will be AI agents. I thought that was really both obviously horrifyingly scary, but also really interesting. And so I think we’re kind of waiting for like whatever, like a ChatGPT7 or something or ChatGPT8, where we’ll be in a place where you can start to really grow with your teammate that is an AI that can do complex cross-browser tasks by hooking into everything. And that’ll be a really awesome day. So I think that day’s coming across probably all verticals, but especially in the MarTech world. And then I’m most excited to grow our business and try to offer cool solutions to marketers everywhere that adds value and grows pipeline because all marketers want to do one thing, which is hit their pipeline number.
Brian Bell (00:32:21): Right. Well, let’s wrap up with some rapid fire questions. So looking back, what’s a belief you held early as a founder that you no longer believe? Listen to people who’ve been there and done it before. Good advice. What’s been the hardest tradeoff you’ve made that outsiders don’t see?
Max Greenwald (00:32:34): Admitting that my style of leadership is to do it all half well instead of doing one or two things extremely well and leaning into that as opposed to trying to fight it.
Brian Bell (00:32:44): Interesting. Yeah, somebody recommended a book called Buy Back Your Time, which is like, you know, try to think about how you can get out of the tasks that you’re doing that you’re just not value adding anymore. That’s my problem. It’s like trying to remove myself from little things. If you’re going to start another company today, what would you do differently from day zero?
Max Greenwald (00:33:01): Set up QSBS trusts so that I could essentially have all of my outcome from my business be tax free. Smart. What are the biggest blind spots that you see among founders and B2B SaaS and AI right now?
Max Greenwald (00:33:12): I don’t know. But if you know, tell me, man, because I’d love to find my blind spots.
Brian Bell (00:33:16): What metrics do you think everyone’s over obsessed with and what alternatives should they look at instead?
Max Greenwald (00:33:22): I kind of love the new hot metric, which is revenue per employee. I think that like aligns both business and customer base actually really well. In terms of metrics that shouldn’t be.
Brian Bell (00:33:32): What do you think is good rev par per like in the seed versus A versus B?
Max Greenwald (00:33:37): what number is good to see for investments in those areas or just for people to focus on?
Brian Bell (00:33:42): Yeah, do you have a sense for like what good looks like for seed round, kind of like a million of ARR versus like 10 million of ARR versus 5 million of ARR?
Max Greenwald (00:33:50): Yeah, I think it’s dependent on what kind of business you are. There’s a lot of probably nuance there, but like the real generalist kind of things, I don’t know, it’d be like, 50K per employee at the seed, 100K per employee at the A, and then by the B, everyone’s paying for themselves. So let’s call it 175K per US-based employee.
Brian Bell (00:34:10): Yeah, makes sense. Yeah, I’m definitely seeing this as an investor. I see founders get to a million of ARR faster than ever with less resources than ever, less money raised, less employees, all that stuff. So you’re definitely seeing the rev per employee go up at all rounds. What keeps you up at night?
Brian Bell (00:34:25): Earthquakes in San Francisco. Yeah, because you’re in the marina, right? And that’s built on landfill, right? So yeah, it’s just built on sand. Where do you guys see yourselves warmly in particular in two or three years, both product-wise and company-wise?
Max Greenwald (00:34:37): I think we’ll be the go-to trusted place to get the best leads for our demand-gen marketers in the mid-market in B2B. And I think we’ll be at hopefully $30 million in ARR trying to raise our Series C. We will have in-person offices in a couple different locations.
Brian Bell (00:34:58): How do you balance as you scale being customer-led versus product vision-led?
Max Greenwald (00:35:01): I don’t know if there’s the right answer to this. I guess my take on it is be product-led for as long as a co-founder is in charge of product. And then as soon as you let go of that, then be extremely customer-led.
Brian Bell (00:35:12): Smart. What’s one underrated resource book or ritual that has significantly shaped how you lead or build?
Max Greenwald (00:35:18): We do team gratitudes every month. Now we should do it every week, but now we do it every month where all you can shout out and thank a couple people for doing great things. It’s sort of like a practice to both give and accept gratitude because I think a lot of my employees called Warmsters at Warmly don’t accept the fact that they’re awesome. And I think coming off of mute on a virtual call and saying to the whole company, thank you, it is true. To accept the gratitude that was sent your way, which is required for this ritual, forces them to confront the fact that they’ve done something awesome.
Brian Bell (00:35:49): That’s amazing. Well, thanks so much for coming on, Max. I learned a lot about early stage pre-PMF startups. I’m sure the founders out there listening will be very appreciative.
Max Greenwald (00:35:59): No problem, Brian. Nice chat.
