Daniel Horning from Linkflow sits down with Mike Adams, founder of Grain. Let’s explore the fascinating journey of Grain, a company that’s making waves in the AI-driven note-taking and sales analytics space. Mike’s journey is a classic tale of tech serendipity and relentless innovation. Growing up in Utah and later moving to the Bay Area, Mike was initially an econ grad who stumbled upon the tech world and never looked back. His entrepreneurial spirit led him to co-found a couple of education companies before diving into the AI realm with Grain.
The company’s name, intriguingly, was inspired by an episode of Black Mirror, where a product called “Harvest Grain” allowed users to playback their entire life history. Mike found the ‘grain’ part captivating and voilà—Grain was born. This name now stands as a testament to the futuristic vision that drives the company. What sets Grain apart is its ability to turn ordinary Zoom calls into gold mines of data. Initially, Grain captivated users with a novel feature that timestamped notes on Zoom recordings, a feature that, believe it or not, was a first in the market when it launched in 2019.
But Mike didn’t stop there. Recognizing the untapped potential in sales analytics, Grain has evolved to offer deep insights, making sales teams smarter and more efficient with every call. It’s a beautiful blend of tech and human-centric design, showing that even the most futuristic ideas can have humble, if not slightly “dystopian, beginnings. So next time you’re on a Zoom call, remember that your conversation could be the next step in a revolutionary data journey—thanks to a little inspiration from the dark side of tech fiction.
Mike Adams: From Econ Grad to Tech Innovator
Thanks for joining the call today. To kick things off, why don’t you tell me a little bit about your background, kind of where you came from, how you came to become. Is it the founder and you were initially the CEO at Grain, right? Tell me a little bit about what that looked like at the beginning. 01:37
Yeah, sure. So a little background on me is I grew up in Utah, moved to the Bay area in 2009 and had an econ degree and was like, what’s this like, tech? Thing, I could do this and watch the Facebook movie and was like, if he can do it, I can do it. And I met a person who was starting a company. That company ended up, you know, we admit to becoming co-founders. That company is still going and is, you know, reached unicorn status many years ago and, that was kind of my intro into tech. But it really did come through education is like I just love teaching and instructing and helping people to understand something they didn’t understand before. So the first two companies were education companies, and in my second one is an online university where we were like, skip college, graduate high school, skip college, go to school on zoom for a year, and get a job at Spotify as a data scientist.
It actually worked. Believe it or not. I was like in charge of the, my co-founder was in charge of the marketing and he was incredible at it. One of the best, uh, PR related people, like our first cohort was on the Today Show. Like it was crazy. And I always took that for granted as a non marketer for sure. I was like, oh yeah, you know, the hard stuff is the is the product and the curriculum and the instructional design and the technology that’s powering it all. They’re both hard is what it comes down to. So but we sold to WeWork in 2018. Things were working. The goal was to try to like, you know, get one of our program and have it was a hybrid instructional model to where we would, like, have students come in and do, you know, in-person training at a WeWork. And so there was no like, actual campus. The rest would be virtual online. In the end, the deal ended up not being really about that.
We got acquired anyway, and I was kind of free to do my next thing. And what I thought was going to be a huge opportunity was all of the conversational data that we were collecting in these lectures, admissions interviews, like I recognized because I’d done some programs before, like you let someone into your school and you’re like, why did we do that? Or like, let’s get more of these people. But our interview process is almost entirely qualitative over an interview conversation. And so we record all of these meetings and then have people turn them into, we would have have humans go through and evaluate against a rubric to make a determination. So yeah, the whole premise is that conversations, so much of what we do in terms of learning or selling or training, it happens over this in-person meeting, medium of conversation that’s moved onto what is now mostly zoom and these virtual platforms. And when you record them all, that’s all data. Lucky for us, large language models arrived and are like the perfect, you know, thing to work with.
For the first three years of Grain, it was all unstructured data, like no one has time to go back and rewatch a whole call. So we were trying to create like interfaces on top of it. That made it easier to work with, like recording data. Now all of the transcript information is like gold for understanding what happened in a meeting, being able to enrich records outside of a meeting, across meetings, contributing to a larger process, like whether it’s the sales process or marketing process or whatever else. And that’s kind of what we’re working on now, is being able to have these, you know, assistants that jump in calls and, you know, take the transcripts and make us smarter and faster and more efficient by not having to do all the manual stuff that we had to do before. There was really an easy way of making sense of all of that data. So that’s kind of the backstory.
Very nice, very nice. So then your your background then kind of throughout that process was build a product and they will come. Right. So talk to me about then the shift of gears into because that works sometimes. I mean sometimes you land you know on like a talk show and great everybody hears about it. But if it doesn’t like how did you shift? How did you start learning about marketing, driving some demand generation? What would what was that like? 05:24
Yeah for sure. So there’s been these multiple phases of Grain and like there’s definitely been a bit of if you build it, they will come wishful thinking. Across the history of Grain and in particular, our focus was on like a really novel approach that like you build it, people think it’s really cool and, you know, or useful, cool and useful, ideally. And then it starts to kind of spread and, to a certain extent that’s been true. To another extent it has not been right. And ultimately, especially if you’re in the venture backed world, you the key to success is repeatability. Repeatability is the key to scale, assuming that there’s a market that wants it, and the challenge with like being overly horizontal or open or kind of like let’s see where the market takes us, is that it’s almost by definition, not really that scalable. And I think there’s rare exceptions to that. So as over time we’ve had to like increase the depth of value. Well, in order to do that, you kind of have to move from adding horizontal functionality to adding, I would say more verticalized value.
Yeah for sure. So there’s been these multiple phases of Grain and like there’s definitely been a bit of if you build it, they will come wishful thinking. Across the history of Grain and in particular, our focus was on like a really novel approach that like you build it, people think it’s really cool and, you know, or useful, cool and useful, ideally. And then it starts to kind of spread and, to a certain extent that’s been true. To another extent it has not been right. And ultimately, especially if you’re in the venture backed world, you the key to success is repeatability. Repeatability is the key to scale, assuming that there’s a market that wants it, and the challenge with like being overly horizontal or open or kind of like let’s see where the market takes us, is that it’s almost by definition, not really that scalable. And I think there’s rare exceptions to that. So as over time we’ve had to like increase the depth of value. Well, in order to do that, you kind of have to move from adding horizontal functionality to adding, I would say more verticalized value.
And so we’ve we’ve really gone deep in the last year on, I would say, deeply understanding small and medium sales teams and how to help them better serve their customer, better train their, you know, their reps better, um, you know, ultimately close and and perform better and close more revenue and deals. And so that’s been for us the last year. I actually tried to avoid that for a long time, because there’s an 800 pound gorilla in our space is gong. And I’m like, man, they’re huge. But they fundamentally like, don’t serve the mid-market and they do serve the mid-market, but they don’t serve SMB. And so there’s like kind of a crossover there where there’s just a huge amount of, you know, market that’s not being served. And so we looked at our user base and despite not deliberately serving salespeople, the vast majority of the users, we’re using it for sales calls. And so the other half was like product research or marketing research or whatever, but generally customer facing calls.
And so we decided to make that kind of like the the sharp edge of making sure that we are really meeting those needs and, as we do so, we still are able to help, you know, the other use cases kind of around it. But it really does start with sales because everything starts with the customer.
Tackling Sales Analytics: Grain’s Evolution
Nice. And that’s something you guys can use and iterate on with your own sales team using your own product. Right. So you’re able to learn a lot about the product as you develop it and sell it with your sales team. 07:59
Yeah, absolutely. Totally. There’s like absolutely those, you know, positive reinforcement feedback loops and you know, to tie it to marketing. Like what’s been interesting to me as a founder is it’s been about the product and product launches product. You build product, you launch it, you build product, you launch it, and it works. As long as you’ve got like a novel new thing and you’re the first to market. So, for example, we had built a timestamped note taker on zoom recordings that would allow you to, like, take your notes and then go back to the exact moment where you took your note. And that seems like kind of obvious now. But we were building that in 2019. And then when Covid hit in 2020, it was like, oh my gosh, like, what are the tools to like make my life easier now that I’m on zoom all day? Like I’ve been living on zoom for five years before that. So like kind of where we came up with these ideas, but, we launched it really positive reception, did it on Product Hunt and did a bunch, but it’s all been like product launches, and it’s been about the novelty of the functionality.
And when you don’t have an ICP and you’re just like kind of putting a, you know, a thing out there, it’s not repeatable and, so then you’re just constantly having to be on the edge of like innovation and, people quickly copied what we did and put it into their products. And now it’s in probably 15 different note taking products. But we were absolutely the first ones to do it. Yeah and then even that has now been less important as, as why people even buy the product now. Like it’s a total minor use case, even though that was what we like, came to the market and innovated with, we raised venture capital around that momentum. And as you mature, you ultimately have to like, really understand who your ICP is and what they care about, what language they use, and I wish we would have done that sooner.
I was going to say that was one of my questions I wanted to ask was, um, if there was anything that you learned or that, you know, now that you wish you would have known at the beginning. Sounds like some of that really targeting the ICP. Were there any misconceptions you had about marketing that you’ve since overcome? 09:52
Certainly our product is freemium, like freemium is a marketing, you know, it’s not a business strategy, it’s a marketing channel. And so that has worked really well for us, for sure. I don’t think we lean into it as much as we could have or should have, like we don’t have. We have like a fixed number of recordings you get for free as opposed to like a monthly allocation. And so we actually have full freemium instead of uh, and I think if I could go back I would have done full freemium. Our cost, you know, isn’t that high on providing it. And I think ultimately people are adopting tools and they want to be able to have confidence that they can use it recurringly without necessarily running out of an allocation and being forced to to switch over. And so that’s definitely like one strategy that is a marketing, but it’s super overlapped with product, of course. And then I would say the other one has been around like it’s really about product marketing. Like I hired a VP of growth three years ago and ultimately, like made a huge mistake by letting the person go when it turns out they were actually really good at the thing I needed them to be good at, which was product marketing, to be able to take what we did and put it into the market and help people to understand it.
But they weren’t doing all the other like, growth related, you know, growth hacks, you know, really big three years ago and then a growth at all costs culture. But I don’t think that’s as as big of a thing now. And I had to learn that hard way of like, man, I wish I would have had that person there the whole time, you know? And you never know how things really worked out. But I certainly made a, decision to move on on, you know, probably a belief that it’s all about this, like activity and just doing all the growth hacks. But I feel like the core of what I’ve learned is like, who’s your customer? Does your product really deliver it to them? And how are you differentiated? Like, what’s that message that helps them to identify and say, you guys, I want to I want to work with you all. And if you’re just pushing innovation into the market all the time, then you don’t really need to do that as much because it’s just about the cool novel thing.
But that like, wears off and you ultimately need to have like a repeatable message through repeatable channels to, you know, a similar group of people. And that’s the key to scale. Yeah. So I would say I wish I would have probably like it’s not that we don’t want to keep doing product launches, but you make those product launches increasingly more ICP focused, and you’re distributing those product launches into channels where you want, where you know your customers, where their use case is relevant.
Okay, I see what you’re saying. So less of a shiny and new kind of product launch cycle and more of like a satisfying the users that you have or want to have because your ICP is so tight. You know, these are the features they want. This is what’s going to make the tool better. This is what’s going to make their life better. It’s like a long play, like a longevity game for the tool. Then it becomes really strong. 12:30
You have to have that deep understanding of who you’re building for because otherwise, like everything else is off. And if you just build something that’s cool and see what flocks to it, there’s so much noise amongst what little signal there is. Whereas if you have a really clear and deliberate strategy hypothesis like, you know, it’s a hard enough process itself, you can, I would say, make more sense of that noise. And so if I were to go back, I will I will start another company after this at some point. And any new company that I would start, I would be like I would say, super focused on a very, very, very clear ICP. Everybody told me that I didn’t listen to it enough. I just was like, oh, look at this cool thing we can do. And venture capitalists were, you know, giving us money for the cool thing that we were doing. And it was unique and novel and frankly, quite ahead of where the market was. And we were right about where it was going.
But like, I would absolutely do that. And then the other is I think it’s it’s related. But like freemium especially like horizontal and freemium is a really brutal combo because you’re you’re making progress, you’re adding value. But it’s all happening like outside of your purview and your understanding. So then you’re kind of like stuck in the product analytics and the what is there, the product analytics. But a lot of times the why isn’t. And so then you have to like add on like research and interviews. But then it’s kind of there’s a lot of noise mixed in with that signal. And if I could go back I would or if for my next company I will absolutely won’t sell anything below a certain average contract value, I won’t build a product that can’t be sold above, I won’t say at least five figures, and then I would sell it directly myself. Build a direct sales force, in particular a content engine that’s bringing inbound for what we’re doing. And then maybe I would add like a freemium offering, like a stripped down version of the product that can kind of drive word of mouth, but it would be on a much stronger foundation.
That is really hard to build when you just put something in the market and like 5000 people start using it for 5000 different.
Grain’s Tools for Customer Insights and Feedback
So what tools do you guys use or is it just Grain? Maybe it’s just Grain, but what tools do you guys use to gather kind of customer insights as you’re iterating as you’re building it out? Is it just Grain or are you guys using some other like software and stuff? 14:41
Yeah, so I mean, we use Amplitude for all the quantitative stuff and then we use Sprig for like I would say survey based qualitative. And then we use Grain plus Zoom for interviews. I’m actually kicking off a research project this week using Grain around kind of a new strategic direction. We’re trying to go on and so on. The Grain bit with that, what we what we do, and I’m trying to make easier in the product is I’ll have an interview guide, I’ll go talk to ten people and ask those questions, and then I’ll have I created a Grain template that will use AI to pull out the relevant parts of the information of the meeting that relate to the question that was asked. And so it like completely not only takes my notes for me, but organizes them person by person question that was being asked. And it’s freaking insane. It’s really, really cool. I have to, like, use a combination of grain and a spreadsheet to do it, but I’ll like take all those summary responses from Grain and then, you know, have ChatGPT take a summary of the summaries and it’s like, okay, for this question, here’s the summarized outcome of more or less across the ten people.
And it’s like it’s a superpower. It’s it’s incredible. In fact, I get like excited going into interview projects because I like it saves me so much time. And not to mention it’s a little less biased too, because it’s like I’m remembering certain things. Whereas like the LLM processes all of the information at the same time, and it’s just kind of surfacing the patterns ultimately that is that are, surfaced that are, I would say the most prevalent and not just what’s, you know, attached in my brain.
Well, yeah, the idea of ten interviews to like go process some sort of information or insight traditionally like we’ll send shudders down anybody’s spine, like. 16:20
Like trying to pull out any sort of patterns and themes. It’s not really a human task. It’s great for, large language models, though, and it.
Would take a human dozens of hours. And then even then, you probably wouldn’t. You would probably miss some of the insights that overlapped. Thing that’s really cool. 16:36
But I assume like kind of customer feedback based like forms data is it’s the same idea, right? It’s all text based. It’s all structured. You can kind of do a summary of, of the answers in a way that is really helpful to, to understand the reality underneath the situation. But um, yeah. Good stuff. Cool.
Rapid Fire: Mike Adams on Must-Have Marketing Tools
Well, that’s really smart. Well, I’ll close really quick with our rapid fire round of questions. Just kind of shoot from the hip with your first answer. But what’s one marketing tool you can’t live without? 17:00
I mean, we use.
Besides Grain, right? 17:12
Yeah, we use HubSpot for everything. So like I would probably say WebFlow, like I remember here’s a story with this one is my previous company. We used a really fancy branding agency, and they built us a really beautiful website in React. And it was like the worst because we couldn’t update it, we couldn’t change it, we couldn’t do anything with it. And so then when I started Grain, I was like, we are not doing that. And so we were going to use Webflow and we’ve been on Webflow for five years and it’s just the best because I never have to allocate any engineering resources whatsoever because we like we couldn’t iterate on the messaging, we couldn’t change a bunch of stuff on the website because it gets stuck behind, like we didn’t have a very big engineering team putting that power into the hands of the, I would say, marketers was was really powerful. And we have no desire to to move off of it or do something else. Although I generating website stuff seems pretty cool, but I’m sure Webflow is working on that too, so
Probably.It’ll be out soon. I’ve heard I’ve that’s you’re the third or fourth person that’s told me Webflow is their go to and I’ve even played inside of there before, but it’s interesting to hear people stick with it for years. What’s the biggest marketing mistake you’ve made? I think you already actually kind of touched on this. What did you learn from it about that product person. Right, product marketer? 18:06
Yeah, I would say, like if you’re building a product first business nail product marketing, it’s a different type of marketing than the others. Like if you have a clearly established ICP and a clearly established product and a clearly established market like that’s, I would say product marketing is still very important for that because you’ve got to differentiate it and do it well. But like definitely, I guess the some of my like marketing learning is the um, is novelty is can only get you so far. Right. You’ve got to like have first principle understanding and then built build playbooks and processes and systems around it as opposed to like put something novel out there and hope it Figma, you know.
Right on. Okay. Last question. What’s a surprising story or fun fact about you your team hasn’t heard until right now? You can feel free to plead the fifth. Always an option. But what would be shocking to your team to hear? 19:13
I mean, the early days of Grain, I don’t know. This is one that I. I was looking for a name for the company and I couldn’t think of anything. And I remember there was a Black Mirror episode about like, every the entire history of you. You could, like playback. And I was like, man, there’s something there. Listening to the whole I watched the whole episode and the product itself in that Black Mirror episode is called Harvest Grain, and I thought the Harvest part was creepy, but I was like, oh my gosh, that’s pretty cool. And then, you know, bought several like used Grain.com and then we ended up buying grain.com and it’s been a great name. We love it. And it means so much more than, you know, some Black Mirror episode. But that’s actually where it came from. So I was just like trying to come up with some sort of idea. And I was like, what’s that thing that was kind of like where you record all of your stuff and it ends up kind of having a dystopian storyline in the, in the, in the episode, but it ended up having a pretty good, like, product name.I think the writers of the show had been really thoughtful around, like a good name for that product.
Nice origin story. There you have it. Everybody’s heard it. Well, cool. Thanks, Mike. I really appreciate it. Definitely some good insights and I’m sure we’ll catch up later.
20:16
Please. Let’s do. And I got a 3:30 so I gotta jump. But let’s let’s absolutely catch up. Keep me posted.
Awesome. Thanks, Mike. Bye. 20:20
