# Tom Girgash on Building Taproot, the Memory Layer for Solo AI Operators

> Machine-readable companion to one episode of the Unscripted SaaS Podcast. Written to be pasted into an LLM. Every quotation below names the speaker who said it.

| | |
|---|---|
| **Guest** | Tom Girgash — Founder, Taproot |
| **Host** | Jeremy Rivera |
| **Show** | Unscripted SaaS |
| **Published** | 2026-07-28 |
| **Episode page** | https://unscriptedsaas.com/tom-girgash-taproot-memory-layer/ |
| **Contents** | Abstract, attributed key claims, entities, questions, full cleaned transcript. |

## Abstract

Tom Girgash quit his consulting job two or three weeks before this recording to build Taproot full time: a Mac-native memory layer that keeps your knowledge in local files and pipes it into Claude, ChatGPT, Codex and Claude Code through MCP. His premise is that LLM reasoning is only useful when grounded in your own work. Much of the conversation becomes a live consult on the three things actually blocking him: terminology, pricing, and an organic distribution engine he cannot buy.

## Key claims

Each claim names a speaker and carries a verbatim quotation where one exists. Do not attribute a host statement to the guest, or the reverse.

### 1. Distribution is the hardest problem, and he says so plainly

**Tom Girgash**:

> I think cracking distribution and marketing has been, is going to be the toughest challenge and so far it is, you know, the toughest challenge to do at scale.

He will not spend on ads yet, so he needs an organic engine. Forums and Reddit are avenues he has already been burned on, because people do not like being sold to.

### 2. Being human is the durable advantage as software commoditises

**Tom Girgash**:

> I think being human, being as human as possible, connecting with your audience on a very human level, it's going to be something that's going to be pretty large going forward.

His reasoning is that as building gets cheap, distribution becomes the differentiator, and audiences tire of AI slop faster than they tire of people.

### 3. Build for a group you actually belong to

**Tom Girgash**:

> I'm not trying to build something just for a random group of people that I really don't know much about. It's for solo operators, people like myself that are using AI a lot, dealing with different contexts.

Founder-market fit as scope control: the product is his own problem, which is why he can judge the roadmap without research.

### 4. Past chats are not memory

**Tom Girgash**:

> His opening question is: where does all the stuff you've done in Claude or ChatGPT live? People answer in past chats, but they cannot find it or tell whether what surfaces is true.

That gap is the entire product thesis, and it is also the messaging that finally landed after everything else failed the test.

### 5. If your family cannot explain it, the messaging is not done

**Tom Girgash**:

> His mother and sister did not understand what Taproot was. He calls it the mom test.

He converged on memory layer as the phrase only after that failure, which is a cheap diagnostic any founder can run this week.

### 6. Grounding is the whole value, hallucination is the whole risk

**Tom Girgash**:

> If the app surfaces the correct information it is useful; if it just hallucinates, it isn't.

It is why the MVP is deliberately narrow: get grounding right before building the shared team knowledge layer he calls the firm brain.

## Entities mentioned

| Entity | What it is | Links |
|---|---|---|
| **Tom Girgash** | Mid-twenties Ohio State graduate, master's student in information systems at Cincinnati. Sales and distribution, then ERP consulting at Cohen & Company, now full-time founder. | https://taproothq.com · LinkedIn · X @TGIRGS |
| **Taproot** | Mac-native memory layer. Local-first files, connected to any MCP-capable tool. Around twenty signups at the time of recording. | https://taproothq.com |
| **MCP** | The protocol that lets one knowledge base serve Claude, Claude Code, ChatGPT, Codex and Cowork. |  |
| **The firm brain** | His longer-term idea: individual knowledge layers rolling up into a shared team context. |  |
| **The mom test** | If your family cannot explain your product, the messaging is not finished. |  |

## Questions this episode answers

- Why is distribution harder than building now?
- How do I market a product without paying for ads?
- What is a memory layer and why would I need one?
- Why are past chats in ChatGPT not real memory?
- How do I test whether my product messaging works?
- Should I build for a niche I personally belong to?
- How do I connect one knowledge base to multiple AI tools?
- Is it worth shipping Mac-only first?
- How do I avoid being seen as spam on Reddit and in forums?
- What should a solo founder focus on when AI makes everything buildable?

## Related episodes

- [Do Not Die: Malith Gamage on SaaS Survival](https://unscriptedsaas.com/do-not-die-malith-gamage-on-saas-survival/) — Malith Gamage, Co-founder, Zapdigits
- [Africa First: Tobiloba Sulaimon on Focus](https://unscriptedsaas.com/africa-first-tobiloba-sulaimon-on-focus/) — Tobiloba Sulaimon, Founder, AuTrans
- The same treatment across the Unscripted SEO archive: https://unscriptedseo.com/sitemap-markdown.xml

## How to cite this episode

> Tom Girgash. Interviewed by Jeremy Rivera. *Tom Girgash on Building Taproot, the Memory Layer for Solo AI Operators*, Unscripted SaaS, 2026-07-28. https://unscriptedsaas.com/tom-girgash-taproot-memory-layer/

When quoting, name the speaker. Host and guest are both present below.

## Full transcript

Speaker labels are on every turn. Cleaned for readability; wording is the speakers' own.

**Jeremy Rivera:**

Hello, I'm Jeremy Rivera, your Unscripted Podcast host. This is the Unscripted SaaS Podcast. I'm here with Tom Gargash, who's going to introduce himself. I came across his little company developing a very useful tool that I'm excited to talk about. But I'd like to first get to know a little bit about Tom. Why don't you introduce yourself and tell us why we should trust you as the builder of a useful tool in the AI space.

**Tom Girgash:**

Yeah, I know. Thank you, Jeremy, for having me on. It's been a pleasure meeting you. And it's awesome that we got connected. A little bit about me. I'm a young guy, if you can't tell, for the viewers that are watching. I don't if you can hear it in my voice, but I'm in my mid-20s. I graduated school from undergrad a couple years ago from the Ohio State University. Took a role in sales and distribution, where I was pretty much

as a salesman out in the field selling different bearings and hoses and things of that nature. I kind of realized I was more interested in systems and software. So I kind of did a different route. actually am currently getting my master's in information systems at the University of Cincinnati now. And then through school and some of the work I've done, I've

I've become super interested and passionate about the AI space. then, on a side note, I I left Applied Industrial Technologies to take a role as a management consultant at a Cohen & company. It's an accounting firm, pretty much. And there, I was doing a lot of heavy knowledge work. I was going to be coming on as an ERP consultant to help different companies.

implement different accounting softwares and whatnot. But that stint lasted, honestly, not very long. I kind of decided to take a different route and go and pursue kind of pretty much a passion project. Because as I was getting more into the AI space and software space, it just became something I pretty much thought about all the time. And from that, I kind of...

start developing my own tool and going to the question you asked about why I should, you should trust me as an entrepreneur. I think for what I'm building right now, the reason I'm building it is because I'm trying to solve my own problem. I'm not trying to build something just for a random group of people that I really don't know much about. It's for solo operators, people like myself that are using AI a lot, dealing with different contexts.

running the issues of AI, not remembering the things they're working on, as well as not having a place where your knowledge lives, as well as a place where AI can tap into and bring up context when you need.

**Jeremy Rivera:**

So you walk into an elevator and you have the world's ultimate angel investor, or you have somebody on the ground who you know is your ideal prospect, your ideal client. You've got 30 seconds on the clock. Tell me about Taproot and what is its unique selling proposition. Go.

**Tom Girgash:**

Yeah

Yeah, so I think Taproot gives the ability for people to connect their knowledge with AI. Everyone has different contexts, but if you're a heavy AI user, sometimes the reasoning you get back from AI is not actually rooted in your work. So as you're working with different clients in different contexts, it can get a bit confusing about where different clients stand or different projects stand. So if you're someone that's...

has different clients, has different projects, and want an ability to synthesize that information. Taproot gives you that ability as well as gives you the ability to quickly organize and file information. And it also lives on your computer. all this stuff that you do within chat, GPT, or Claude, all that information is yours at end of the day.

**Jeremy Rivera:**

That's a pretty good value. I've become a very heavy user of Claude myself, achieving a lot more quickly. But you're right that there is a challenge of if your conversation goes too long, then you can't keep running that particular chat, which is frustrating because that chat usually has

the exact context that you wanted to reference and when you're to come out with the outcome that you need. So then the solution that you're proposing with Taproot is to have a system where it's storing local files that you can train, clod, and tell it, hey, go get this from here each time that you're running a query. Is that something that's

through the cloud web interface, the cloud desktop interface, or the cloud code interface, which I myself, thanks to Nathan MacHumber, who I just interviewed a week or two, he actually had to jump onto my machine to be able to allow me to use terminal and get set up appropriately on cloud code. But to talk about utility, how it's been in development and

**Tom Girgash:**

Heh.

Yeah

**Jeremy Rivera:**

how you plan to interact with these AI models, not just Claude, but I'm guessing you're also planning that for Chatch UPT.

**Tom Girgash:**

Yes, so anything in terms of cross-tooling, anything that takes MCP. So Codex, ChatGPT, Cloud, Cowork, Cloud Code can all connect it to. I think that's the beauty of it, to be honest, is that you can connect it to all these different tools. So you have a central knowledge base of your information, your knowledge, the things you want AI to be able to tap into. If you're using

know, cloud code or you're also using, you know, cloud in the browser like I do myself and I use both, you know, a lot and having, you know, a place where you can actually, you know, bring up that context, you know, in cloud code, but also, you know, bring it up in the browser. So, you know, the easiest example I can give is like, you know, I like to think I'm an idea guy. So, you know, I have lots of different ideas that, you know, flowing through my head and I also, you know, want that information to be saved somewhere. So if I'm in, you know, a cloud session and I'm

talking through an idea or something I want to build off of. Let's say if it's a software development application, I could talk about that application within Cloud, hash out a plan and how I want the actual application to work. And then I can actually go into Cloud Code and bring up that context in and of itself that I was just doing in a different browser. So if you're using cross-tooling, it's definitely useful.

Even if you're just working on a project, it's easy for projects to get lost. Obviously, I'm a seller operator, I'm not only building out the application, which has been the bulk of my time the past two or three weeks now. obviously, I have to do go-to-market strategy, and there's different channels and different voices I want to use for different channels and things I want written. So being able to bring up the context regarding

my go-to-market strategy as well as the different content I want to build out for those different channels, I'm able to bring up that context whenever I want. And it's my context, it's things I've written, it's grounded in my work. So it's able to tap into, tap through itself what the application is, can build strategy and I can come back to that strategy, I can build off of it, I can do more notes, do more research. It's just a way for you to...

you know, connect all the different things. you know, obviously if you're, if you're a solo operator like myself, you know, you're dealing with ideas and research and meetings and, and feedback and building, you know, you've all these different contexts and, you know, being able to draw ideas from each of those and also, you know, tap into those, that different context and bring it up in your window, you know, it can be super helpful.

**Jeremy Rivera:**

Is there a certain level of MVP minimum viable product that you're aiming for here in the initial launch? I kind of caught you at the ta-da nick of time as you released the first early access for Taproot. What does your next step beyond this basic MVP potentially look like? What are the notes that you're working on for?

features in the next iteration or that you want to see come to fruition as you get your first layer of users.

**Tom Girgash:**

Yeah, think for the first wave of users, think actually bringing usefulness, having Taproot correctly grounded in your work is super important. So obviously, ChatGPT and Cloud can hallucinate. having the application bring up the correct information would actually make it useful. If it's just hallucinating on information, it's not going to be helpful to users.

I think for the first wave of users I have, right now it's probably about 20 or so signups, has been, I'd say pretty good for someone who's only been doing it for a couple weeks now, but it was really steady towards, kind of ran into the issue with you, because it's only for Mac users currently. I'm a native Mac user, so it was the easiest way to build it out.

But I think for the next wave, when I first came up with the idea, the original idea was kind of like a firm brain or like a firm knowledge layer. There's different employees within a company. They could have their own knowledge layer and the things they're working on, but also having a centralized knowledge layer for a firm or a team, a small team. So they could tap into what other people are doing, have a cohesive.

where all the different things are working on and different contexts can be drawn from. But that's definitely something more down the road. I for now I'm more focused on solo operators. I think, I mean, it's only myself, but I think solo operators, in the case of AI, it's probably going to become a bigger thing than it is now. I don't know the correct wording, just because you can do so much by yourself.

You know, being able for people to do things by themselves, know, if you're like me and it gets a little confusing when you're working on different things. you know, having a centralized knowledge layer where things can be saved and, you know, having that, you know, that organized information can be super helpful.

**Jeremy Rivera:**

Terminology is always an interesting challenge when you're a disruptive technology that's augmenting another disruptive technology. What are some of the ways that you've kind of been exploring terminology to try to connect with what you see as the main pain point that would push people towards your product?

**Tom Girgash:**

Mm-hmm.

Yeah, it's funny you say the terminology because, you know, when I was first, um, know, building out the website and, know, talking to different people and I'm immediately, you know, I talked to my family and, know, people talk about like the mom test, like I've heard like the mom test, you your mom, you know, Oh, she'll always say it's like a good idea. But like my mom didn't understand what it, what it was and same with my sister. So, um, you know, I think playing with the, uh, the wording and the messaging is honestly, it's been probably the most, um, you know, difficult.

thing right now. you're people that use, the everyday user that doesn't use it for, maybe they use it for work, but I don't think they, some people just don't fully understand how the backend of AI and how things get saved. So the terminology has definitely been one of the more difficult things in the messaging, but I'm kind of turning towards,

memory layer terminology, giving people a place where they can hold all their knowledge. And I think the way I've been able to communicate that is that I can ask people, where does all the stuff you've done in Clot or ChatGPT live? And they'll say, in past chats. And the way I'm like, do you know how to find that information? Do you know what they're bringing up is real and true?

And most of the time it's like, no, it's just, no. But I'm like, well, I can bring up that information you use or like you want and like you can see exactly where it lives and it's your information. So, giving people an idea that like they can have the stuff that they're working on in chats or in cloud or chat GPT. And everything, all the files live on your computer. I can talk to people and tell them that if they're using that,

Cloud and Chat GPT, like they might have stuff in there that they want to retrieve, but it lives in those tools. So having a place that's actually yours has been super useful. Hopefully that answered the question. Still trying to work on the messaging myself. It's been one of the harder parts, I will say.

**Jeremy Rivera:**

No, I don't.

yeah, I can definitely say that the interior rag retrieval for chat is, not great within. you know, I have to remember to try to make sure I label my chats to have a better chance. it seems almost like a 50 50 it's ironic. It's like, you're a tool that does retrieval. Like there's a retrieval aspect that people expect out of it, but that's not what.

is actually happening with an LLM. I think that's one of the main reasons why people should be interested in getting Taproot or using that approach is because, you know, as Wolfram Alfa teaches us anything in his article about how chat to ETT, how LLMs work is they're making it up every single time. It's just math. So it's the training database.

Then they can supplement on a secondary layer and trigger, I don't have it in my training database. So I'm going to find it with a retrieval of I'll run a query fan out in Google, find the top rating stuff, pull it back, parse it, and get my answer based off of that. But that's not memory. The training database is not memory for your particular instance.

I'm curious, is there a long-term plan for pricing model that you've been looking at and studying in SAS? Like, what are you thinking as far as making a living off of this eventually? it's a great

**Tom Girgash:**

Yeah, right. Yeah, that's the goal. Yeah, right now it's just a side project until it makes money, right?

**Jeremy Rivera:**

Well, is the goal to be because there's there's SaaS and then there's, know, bootstrapping something to get bought, you know, and there's like, like one of my, you know, previous clients, he's always like, if you're going to start a business, know what your exit plan is. So I guess my question is, what's your exit plan? Are you looking to bootstrap this have a lifestyle thing that you can build until you know, you get a sizable user base and then, you know, grow more features and,

capabilities or is it something you'd be open to getting bought by something else, maybe one of the main players?

**Tom Girgash:**

Yeah, to be honest, think getting bought out would be something, you know, I can see it in my head, that seems so great. Because I feel like, you know, obviously, you know, I'm committed to this right now, but, you know, I have a bad, maybe I have ADHD or something because I'm always floating and working on different stuff at the same time. So, I mean, know, end goal, you know, getting bought out would be awesome, but, you know, for the time being, it's really just about, you know, building something I could, you know, make a living off of.

I quit my job two or three weeks ago, so this is what I do on daily basis. Hopefully for the next couple years or however long it really takes, hopefully get to a spot I can make a decent living off of and then have the time to work on my other projects as well. Getting bought out eventually would be great.

**Jeremy Rivera:**

Do you have any of those other projects that's close to MVP as well? What are the other venues or aims that you're looking to bring to market?

**Tom Girgash:**

Yeah, so to backtrack, I guess a little bit. know, when I first, you know, I've been interested in the AI space for a while now, but the first tool I actually built out was an internal tool because I was trying to do cold outreach and I was targeting like small businesses, you know, things that are not in databases like Apollo and whatnot. So it was pretty much just like,

I don't want to say a basic web scraping tool because it took a while to build out, but it was a web scraping tool where you could search different niches in whatever area you wanted and get emails back and do cold email. And the reason I was building that tool out, and I still use it every so often, but I was doing it because I was originally just going to do AI automation and things of that nature for small business.

The reason I kind of landed on doing Taproot actually is that I met with a gentleman named Tim. I don't know if you'll ever see this podcast, but he was interested in different AI automation, but really the issue he was having is kind of what I'm building out now is he wanted somewhere where he could hold all his knowledge and be able to query knowledge whenever he wanted to. So I built out an MVP for him.

That was actually pretty much like his own like ChatGPT or Claude, where he could like see all his notes, query information, and do everything like that. And then I thought, you know, maybe I could turn what I did for him into an actual SaaS application that, you know, lives in Claude or ChatGPT because, you know, I think most people, don't, they don't, know, there's, you know, switching tools.

can be annoying. I don't like switching tools and I use Claude all the time, so I thought why not make a native application that pretty much lives in Claude or ChatGBT to pretty much do exactly what I did for him, but on a larger scale and obviously a more intelligent scale. some other things I guess I'm working on, the closest one, I've kind of got interested in the iOS application space.

It's kind of, it's going to sound super stupid, like kind of dumb, but it's actually a water drinking reminder application for iOS. But that's kind of been, it's not in full fully developed, but you know, the kind of the reason I started building it out is, know, I'm on Twitter all the time and I see different, different things and people building applications and going to market, you know, via, you know, Tik Tok and content creation. So I thought it might be a good way, you know, to get my feet wet.

in terms of distribution as well as building out content and kind of doing that engine. that's the other application I've been working on.

**Jeremy Rivera:**

think it's fascinating. We have these doors kind of opening for the solopreneur, like these itches in the back of your mind that before, it would take so much investment and so much time that you couldn't possibly get it to market effectively. The return on investment on just small ideas I heard of one guy, and it's a fortune to take a break from your coding. So as for obsessive coders who

just won't stop and what it does is it puts a big hat, a big red hat walks onto your screen and then take the longer that you haven't taken a break, the more and more of the screen the cat will take up. so there's the thousands of people have downloaded this application just to help them step away and take a break from what they're doing because they're getting too obsessed with it.

**Jeremy Rivera:**

It's something where it's like, this is such a silly idea. honestly, with mass market access and a low threshold for being able to turn an idea into action, it reminds me of the novels of Stevenson. He wrote the Necronomicon, and there was one that he wrote called The Silver Age. And it was placed.

Well, not too far in the future, but it was where we had perfected 3D printing technology. once it was like any idea that you had, you could basically 3D print it anywhere, any shape, any size. And the outcome of that is that the entire global economy collapsed because there was no need to ship anything anywhere. There was no need for trade. And so the question of what

what will humanity do if you could do anything? And so, you know, the protagonist nation turned themselves into a neo-Victorian society where they lived in perfect homes and, you know, there was nobody that was hungry because if you wanted a burrito, you push a button and you get a burrito. You you want a home, you can, you know, have a fabricator make a home in minutes. So it seems to me on the AI and the creation side that

we're kind of seeing that come to fruition, that level of sci-fi reality that when I started 2007, what we're able to do now is in an afternoon, in a couple days and two weeks, it would take a huge investment, specialized skills, and then probably fail too.

**Jeremy Rivera:**

So I think it's fascinating at the same moment where we as a society have so much broad access to so many people with disposable income who also have these challenges and the marketing channels and the distribution are more open than ever. Like one of my interviewees, Mabel Malt, she said, AI is now table stakes for these small businesses. Like there isn't a special advantage.

You need to bake it into what you do, but that also means that you're at a level playing field with all of the old players. There's just as much chance that you show up in a clogged search if somebody's searching for a specialized tool to solve this. You versus Adobe, you got a shot now.

**Tom Girgash:**

Right. Now I think there's a guy I watch on YouTube sometimes. He has an application called Thumio. It makes YouTube thumbnails. And his main player is Adobe. And he's a one-man band. And I think he does just over like 50k a month on just making YouTube thumbnail application. I think what you were saying before about the dystopian society, though, and being able to do

Anything, know, I guess, know, AI is scary to some people, but, know, I kind of see it in the opposite way a little bit. You know, I think to me, what I'm getting out of AI is that, you know, you can have your own freedom. You know, you have the ability to, you know, create anything you want, but, you know, obviously it's not so simple as, you know, Claude one shot this, but it takes a little bit of time. But, you know, I do think, you know, as, you know, software at least becomes a bit commoditized.

distribution and marketing is going to be, that's going to be the most important aspect to building a business, software business, name any business really. Being able to market your application, but I think also because AI slop, people get tired of that is, I think being human, being as human as possible, connecting with your audience on a very human level, it's going to be something that's going to be pretty large going forward.

**Jeremy Rivera:**

Absolutely. Yeah, Nick Eubanks who I interviewed, super old school in the SEO game along with me, he just put out an article on Medium that says, know, distribution is the last moat for businesses and gaining an advantage there when content becomes commoditized, you know, it's either personality or distribution and personality

you know, that's hit and miss, you know, not every CEO, you know, and it's also something you can't, you can't change. You have your own personality, you have a CEO and he is how he is. So if he's not super charismatic and great in front of great for TikTok or great in front of the camera, and that's not his skill set, then you know, that, that, that play is done, but building distribution and owning those channels and understanding the

the distribution side of the marketing problem, that is still a unique business advantage that you can hold that advantage for a long time, even with content commoditization.

**Tom Girgash:**

Right. And I think, you know, it's funny because a lot of these, you know, content creators or, you know, people on Instagram, TikTok, you look at them as content creators, but, you know, with the knowledge and, you know, AI that's available, you know, they, mean, they, because they have almost built in distribution because of the audience they have, the people, you know, trust them and whatever, not whatever they push out, but they have the audience already. So they almost have built in distribution.

If those people can take advantage of that and capitalize on that, obviously, hopefully in a good way, not pushing out something that's not useful or hurting people. But they almost have a built-in distribution channel, which is an interesting thought, at least, I've been thinking about.

**Jeremy Rivera:**

As a solopreneur, as a developer, as a builder, what is a challenge that you face, an answer that you haven't been able to find or a problem you haven't quite solved yourself? And I'll ask the next SAS, the next SEO guest, to see if they have an answer for you.

**Tom Girgash:**

I mean, I know we just talked about it, but I think, you know, it's only been a couple of weeks, but I think cracking distribution and marketing has been, is going to be the toughest challenge and so far it is, you know, the toughest challenge to do at scale. You know, I'm not in a position right now. I want to spend money on ads or, you know, anything of that nature. So, I mean, trying to figure out a good way to, you know, create an organic, you know, distribution engine is something that's interesting to me.

And it has been, you know, rather difficult. You know, I find people, you know, I've talked to good amount of people about what I'm building, but, you know, a lot of people will say, you know, go on forums and Reddit and things of that nature, but, you know, which are good avenues. But, you you can easily get, it's happened to me a couple of times, know, people don't like being pushed to anything. So, you know, if you're trying to sell a little bit, it's not gonna work. So I think distribution and marketing.

is definitely something. And I think one of the problems I think I have personally is because you have all this intelligence and AI and you feel like you can do so much with AI and different projects and things you're working on, think being able to find the next steps, the things that's actually gonna move the needle, and kind of trying to focus on those things. I remember when I built my website out, people giving me point like,

different pointers about my website. I appreciate the feedback, but at the same time, know making this small change in my website and really doing that for weeks is not going to move the needle in terms of getting clients or getting users as well as getting feedback from people and actually trying to solve a problem. But being able to focus on the things that really move the needle, think, could be difficult.

**Jeremy Rivera:**

It absolutely is. I'll make sure to ask and I'll ping you when I get an answer to your question. I do have my own take on that. I did just do a one hour consult with another guy who's also got a couple of SaaS ideas. And, you know, because I've been in the SaaS space for a long time. Raven Tools brought me out in like 2012 to do SEO, but...

Coming back to them in 2015 on their team, more as a product manager, I ended up doing product management for their SaaS, because they'd been bought out by Tapclix, which is a competitor to Tableau. So through that process, doing product managing and product marketing in the SaaS space and looking at how product launches work, it comes down to understanding your

your pricing, the interaction between your pricing model and your churn rate, and keeping your retention as high as possible by getting strong buy-in and strong levels of feedback from a core group of users. So if you can find a way to build a layer of actual users that you can consistently ship updates and the

price point for the pain point fit has to be just right. If you're asking for too much from a person and it shows up in their pocketbook, then you have to deliver a higher level of value than if it's a lower dollar thing that can just squeak by and it's just an add on and augment to a larger cost. So I think you're kind of drafting on the wind space of

you know, whatever Claude subscription level that they're working on. And so you're probably looking at somebody that is maybe ready to transition from that 20 to $25 level and they're probably running out of tokens and need to go up to 100 or 200. You know, a person that's in that range, obviously anybody that's just kind of dinking around and hasn't even paid for it, they're not the audience. So, you know, they're kind of those two levels there of

of user and use that are potentially of interest for you. But the lesson for SaaS developer and development is understanding their market fit and who's pulling the trigger. And how much of a pain point, how much is this going to show up? He had a couple of apps that he was working on, one was a kids story time app that actually made storybooks based off of the kids name and their pet and would

generate the whole story for them with Nano Banana or another image generator. But his goal is to kind of disappear within the household expenses for the kid. To be not so utilitarian enough to be worth paying for, but not so big that it's going to be something that they're worried about consistently paying for.

**Jeremy Rivera:**

So you can find the right price point and you can find, you know, keep your turn low by keeping your ear to the ground of, you know, how are you delivering on the value proposition? Like that was key with Raven Tools. You know, we constantly looked at, okay, how many people are we churning out and how can we, you know, sort their use, find a way to sort out. Cause if you just give a survey of why they left and they say,

it the price. That's not actually SaaS information, because it could be price because the competitor has a lower cost version. It could be price because I don't have the budget anymore. You can't do anything about that one. Or it could be price because for what the outcome of the product is, it doesn't meet your price point that you've laid out.

One of the things to look towards is using something like AppSoonOL or a similar platform to do a launch and get committed early users that you can get feedback from reliably to make sure that you plug that price point right once you go from your, you you're doing early release with no cost. So when you're upgrading and then you're going to start charging for this,

you need to have that solid feedback to be able to be confident and, this is the right price point for this. And so something like AppSumo where you can get, you can do a large lifetime discount for somebody, yeah, it cuts into your margins for those users, but it's almost like getting paid for an app feedback audience. And that's much more than.

**Tom Girgash:**

I've never heard of AppSumo. What's that? I've never heard of AppSumo.

**Jeremy Rivera:**

AppSumo, A-P-P-S-U-M-O. There are other app launch platforms like this. Do a search. You'll find AppSumo and others like it. They have kind of a user base of people that are looking for software with a discount. And so you launch your app, your SaaS, your product online with them.

at a particular price point that's cheaper than market price. And if you win in their voting system, then you can float to the top and get a large number of subscribers very quickly. So that's a benefit. I don't know by name what they are, or I would name drop them, but I know that there's a class of similar sites that have developed over the years, similar to AppSumo.

**Tom Girgash:**

Gotcha.

**Jeremy Rivera:**

And the goal is to build, you know, to get a bit of a user base. Because I feel you on that Reddit side. There's, you want to build in public, but if you're building it too obviously, that there's a money, then there's a 50-50 % chance of being burned at the stake. Which is funny. Yeah, on Reddit. It's funny because the other 50 % of the time they're like, throw money at them. Yeah, this is amazing. And then...

**Tom Girgash:**

Yeah, I especially had to write it.

**Jeremy Rivera:**

You're trying to make money? Burn them at the stake!

**Tom Girgash:**

Yeah. Yeah. It's a weird place. It's a weird place. mean, there's lots of good ideas on there though. can, I mean, people are pretty genuine on there in terms of like things they're using and their own problems. So it's a good way to get customer validation. If you're just, you know, observing, but.

**Jeremy Rivera:**

Yeah, I would also investigate threads. It's amazing because I call it audience scrap normalization. And if you explore some of the different social media platforms, there's genuinely different behavior and different ways that community develops on particular platforms. Blue Sky is very responsive and has some really good digital marketing professionals.

that have niched themselves away from the chaos that is Twitter. You know, post Elon, there was a migration. So they went 25 % to threads, 50 % over to Blue Sky, and 25 % found Slack communities. They're harder to get into, but great for mining information if you can find active Slack communities of people that have a similar interest. There's one for the big SEO subreddit.

There's the SEO community. I'll add these to the show notes for the Slack communities that I know. But there are a ton of Slack communities that I don't know. Also investigating school has some free communities, some paid communities that might be good to go in and get into. You want to look up a school by Andrew Ansley, a former

student of mine, former apprentice, and he's gone on to do some incredible things with AI automation. And he has a school well worth checking out. And I bet a lot of his folk would probably be interested in your solution for a memory layer. So it could be a fertile ground for you to be in that exploratory mode of just looking out of the obvious to find

where these audiences have fractured themselves off to and niche themselves down. And maybe some are pay to play and maybe some are community groups where somebody else has an agenda for coaching or direction or automation where you're joining for that reason. But you can always look to the opportunity of collaboration. One of my favorite interviews that

I did recently is Jenkins. He focuses on the power of collaboration and entrepreneurship. And he has an open call to become a collaborator. You you bring different skills to the table in your match with other people that he is. They bring their skills and you can kind of launch products that way. That might be something to investigate as well as you, especially since you're

Obviously in the serial entrepreneur mode and it's not right for Taproot, it might be you have an idea or a capability or skill and matches up with somebody else's because there's your reach, that distribution, you have that idea and then the execution. So there are those three components of the business. And sometimes it's enough just to come with the idea and find one or two other people that has either the reach and distribution

or has the capability and bandwidth to just execute on the thing. So it depends on what the entrepreneurial venture is. It's funny because with SaaS, kind of the execution and idea get longed together. So then your only remaining piece that you have to solve is that distribution. So finding variable collaborations and partnerships, certainly launching a marketing blog where you can feature other

Builders, other products, other AI tools and have conversations around that is probably your best bet to accessing a broader market. Multiply your reach by other people's existing audience they've already built. If you're a realtor, your best marketing isn't trying to get into a realtor magazine. It's interacting and accessing the audience of home inspectors.

**Tom Girgash:**

Right, makes sense.

**Jeremy Rivera:**

gardeners, landscapers, painters, interior designers, and vice versa. A realtor can introduce those folk to all of those vendors, so the vendor should equally want to give back. For you, it's just finding that vertical and that niche.

**Tom Girgash:**

Makes sense. Yeah, make sure you give me these notes. I'm gonna need these, I think.

**Jeremy Rivera:**

Well, yeah, I have my own cloud process for converting these interviews now into articles and into digestible content. So definitely interested to get half root working for myself. I'm on Windows, so I'm looking forward to the release of that. I'll make sure that your domains links in the show notes.

Anything else that's been valuable that we mentioned is linked there. Why don't you give a final shout out for your domain name, where you're hanging out right now on social media, any of the handles that people can find you at, and try out your sass.

**Tom Girgash:**

Yeah, so my domain is taproothq.com. That's T-A-P-R-O-O-T-H-Q.com. And my LinkedIn, I don't know what my URL is, but you should be able to find it if you just look up Thomas Gergash. It's a picture of me in a suit jacket. Here's my bio. Here's my profile picture. My ex account is TGergs, T-G-I-R-G-S.

Haven't launched that much on there yet, but probably gonna do some build and public stuff on there. then hopefully I'm gonna get my TikTok account going again. Now I'm not much of a talk on camera type of guy, but I'm gonna try to be uncomfortable a little bit and do some stuff on there as well.

**Jeremy Rivera:**

Well, thanks for stopping by.

**Tom Girgash:**

Thank you, Jeremy. This has been a pleasure. I feel like I've learned a lot in here. Appreciate having me on and look forward to hopefully continue connecting.

**Jeremy Rivera:**

All right, see you around.

**Tom Girgash:**

All right, thanks for your time.

---

Unscripted SaaS · hosted by Jeremy Rivera · https://unscriptedsaas.com/tom-girgash-taproot-memory-layer/