317: Podscan, 2 months and $1k MRR in

This episode focuses on Arvid Kahl's experiences in building and growing his startup, Podscan, over its initial two months, specifically achieving a milestone of $1,000 in monthly recurring revenue (MRR).

1. Strategic funding usage can accelerate product capabilities and growth.
2. Direct customer interaction is crucial for real-time feedback and product improvement.
3. Understanding customer awareness levels can significantly enhance marketing and sales strategies.
4. Structured communication and proper call management can improve customer relationships and sales outcomes.
5. Setting realistic goals and milestones can reduce stress and increase focus on meaningful business growth.

1: Introduction

Arvid Kahl introduces the episode’s focus on Podscan's early success and the strategic use of funding. Arvid Kahl: "Let's take a look at what happened with Podscan in the last two months."

2: Financial Overview

Discussion on financial strategies, funding details, and expenditure on computational resources. Arvid Kahl: "I spent around $10,000 or $11,000 so far, all on GPU compute."

3: Enhancements and Infrastructure

Details on the upgraded infrastructure to enhance Podscan's transcription and AI capabilities. Arvid Kahl: "Now I'm running a pretty beefy 48 core server for my main application."

4: Customer Interaction and Feedback

Kahl discusses how he engages with customers to refine Podscan and adapt its features based on direct feedback. Arvid Kahl: "Talking to customers has been an eye-opening way to refine our product."

5: Learning and Application from Microconf

Insights gained from Microconf, particularly in customer awareness and structuring sales calls. Arvid Kahl: "Rob Walling's talk on the customer awareness scale was particularly enlightening."

6: Future Plans and Goals

Kahl outlines his future goals for Podscan, including developing new AI features and improving customer service strategies. Arvid Kahl: "I'll be working on new AI features this month."

1. Utilize strategic funding to boost computational power for quicker scalability.
2. Engage directly with customers through structured calls to understand and address their needs effectively.
3. Apply learned sales strategies to enhance customer discovery and retention.
4. Monitor and adapt business goals to maintain motivation and focus on impactful outcomes.
5. Explore educational events and resources to continuously improve business and marketing strategies.

It's been two months of running Podscan at full speed.
Have I spent 6 figures of money yet? Well, you'll see in this week's update :)

This episode is sponsored by Acquire.com

People

Arvid Kahl

Companies

Podscan

Books

None

Guest Name(s):

None

Content Warnings:

None

Arvid Kahl
Welcome to the Bootstrap Founder podcast. It's been two months of running Podscan for real, so let's take a look at what happened, what didn't happen, and if I already spent all that funding money. This episode is sponsored by acquire.com. More on that later. First, because it's quite likely the most interesting topic, let's talk about money.

I never disclosed the full amount of what I raised, and I likely won't for a while, but let's just say it's a six figure number. That's all I can say. It's not a surprise, because it's pretty usual for the Kalm company fund to invest this much into bootstrap companies like mine. But I won't go into the details because I talked to Tyler about this, and I might actually release a podcast episode that I recorded with him prior to me signing the whole thing, the contract about the deal, you know, and the specifics. We had a conversation and we chose to record it.

So I'll probably release that this week or next week. But, you know, that's what it is. Six figures. And before the company fund invested, obviously I was bootstrapping, then invested, and then things changed. So did I spend six figures on the business already?

Well, of course not. I spent around $10,000 or $11,000 so far. And that's all on GPU compute. The plan was to ramp up the computational power behind Podscan's, you know, transcription and AI features much faster than I could have with a purely self funded business that I had before I got that little cash infusion. And that's exactly what I'm doing.

So before the funding, I had a smallish main server at Hetzner or something for the app and a single database, and around five GPU included cloud servers for transcription, all fairly small. Small server for my search engine as well. That was my whole setup. Now I'm running a pretty beefy 48 core server for my main application, which is hilarious because it's just a PHP app running on 48 cores. Is quite funny to watch.

I have a similar 48 core server for the search engine. That one has a lot of storage, like, I think roughly a terabyte or so, and with the amount of data that I have, it's actually quite needed. So also quite beefy. Very fast at indexing, very fast at responding, which also allows me to have full text search, like type ahead, full text search that responds within 20 to 50 milliseconds, which is really, really cool. That's always surprising to my users.

When they start searching, they actually get results in real time from a massive database of millions of records. That's a lot of fun. And paying for a server, I don't know what it is, like $150 or something like a month for such a server to do this. Absolutely worth it. Beyond that, I have 16 cloud GPU servers, just like the five I had before, but more for transcription, inference.

And I have a staging cloud GPU server for experiments with new features if I want to build something that I hadn't done before. And I'm working with diarization, which is detecting the speakers like who's speaking in a podcast, and summarization and all these kind of things, and you have a little experimental server to run these things on and figure things out. Now, I will ramp this up to 24 or so cloud servers over the next month. Like I said, I'm working on summarization and entity detection features that require some intense AI magic, which needs a few more resources. And I'm also really slowly but surely adding more of these resources.

And it will cost me probably around $15 to $20,000 a month in the end, which means that I have a good year or more of Runway just keeping expenses as they are. And mind you, this is all done so I can get through the historical backlog of transcribing all past podcasts, right? That has an end once I'm through with that. Pretty finite number. It's millions.

It's hundreds of millions, but I'll get through that. My daily compute needs to be, I don't know, just ten, one 10th of what it is right now. It's going to be really, really low once I've caught up with all past things. And I only need to go through the currently released I'm on a podcast. Very finite.

I think there's roughly 20 to 30,000 new episodes in any given day, which is not a lot. I can do two 3000 an hour. So within 10 hours of any given day, I've already worked through all the episodes of that day with my current setup, so I can have that even like maybe use one third of it and still be very much fine. And besides compute, I don't really need to spend the money that I have on anything else just yet. Like, I'm not wasting than anything.

I might be hiring a few experts for like, design and user experience work soon, but that will be on a per project base. So I don't expect this to be a recurring expense, at least not for the whole duration of, you know, the next couple months or years. So a year of Runway just for my war chest, that's where I'm at. But what's happening on the revenue side of things because, you know, Runway kind of is affected, at least by how much money is actually coming in. So I've just crossed the $1,000 monthly recurring revenue mark last week, actually.

And on average, I get one or two subscriptions a day, mostly on my lowest tier plan, which is $20 a month. Several people buy the yearly plan, which is nice. So there's a lot of, you know, money coming in for the next year or so. Churn at this point is 0%. I mean, I've been around for two months.

It was funny, but it bodes well for the long term retention cohort graph that I will likely have in a couple months. Month one to to month 2100% retention. That's not too bad. I really like it. Apparently the product, when people commit to it, is not so bad that they immediately stop purchasing it, which is kind of cool.

I mean, I will see churn. It's just a natural thing over time. But right now I don't have any, which is the optimal situation to be in, and I'm quite happy. So one k mrr, obviously that's not ten k, like, or eleven as much as I'm spending. So Podscan isn't profitable on that end, but it's technically halfway there.

Like I said, the post, you know, history catch up pot scan will not cost more than like, $2,000 a month in pure infrastructure costs. So that's what I'm aiming to get to as soon as I can in terms of revenue. Cause then I know that I can just turn off all these servers for, you know, historical catch up and run a profitable business. That's. That's the crazy part, right?

Imagine, like, you know, if there is a dire situation or run out of money, I can just, like, more slowly ingest historical podcasts and have a profitable business. All I need to do is double the amount of customers that I have, and I already have customers. So doubling is really not that complicated. At least I guess so. I'm getting better at acquiring customers, too, maybe that's an important thing to mention here.

And I've noticed that over the last couple months, I've changed quite a bit here. I've been forcing myself to be as available as I can for customer research conversations, conversations that are mostly induced by customers actually reaching out to me. I put a link to my, a calendly link into all my emails. That I sent. And in most of my communications, it's just, hey, if you want to talk to me, book a little 50 minutes slot.

And I've talked about this here on the show already several times, that this has been an eye opening way to talk to customers because it allows them to actually speak to me. It's not just like typing something, like writing an email, writing some kind of draft, whatever. It's just going on a call with me, 15 minutes, explain to me what they need, explain to me what they're doing, and then I get to give them real feedback. I can show them stuff. It's kind of a sales call, even though some of them are already customers.

So it's kind of a nurturing call. Like, it's different every time, but it's extremely valuable to give people the option to talk to me. And I have currently restricted the days on which people can call me because it's been a bit chaotic. I'll get to that in a couple minutes, I guess. But just being open to these conversations and letting people schedule them into my calendar, as interruptive as it is, has been very, very helpful.

I'm now over 20 conversations into learning more about my customers over the last couple of weeks, and that can be a lot, but it's been really useful. It's incredibly powerful too, to chat with people who have a problem, and they know it. And I think this might be a good opportunity to mention how much Microconf. The conference in Atlanta last week has actually helped me with this. Two talks from that conference stand out to me.

I think I talked a little bit about etalk last week, but I'm gonna talk about the other ones now. One was by Rob Walling, organizer of the conference. He was diving deeper into Eugene Schwartz's customer awareness scale, which is really, really useful to understand for your marketing and sales purposes. Like, how much do customers know about the problem? They have the solutions in the space that the fact that there is a solution at all, the fact that there's a problem at all, and how much do they know about your product?

That's kind of what the customer awareness skill is about. Rob gave us a really nice marketing framework for all the different stages that was really helpful. And Steven Steers talk about structuring sales calls was also really cool because I can take all the knowledge about the sales calls that he's talking about and apply that to my customer discovery calls and my kind of sales ish customer discovery calls as well. Because if you talk to a prospect, you're always trying to help them through the trial. You try to sell them, right?

Or you try to get them to go to a point where they get the solution to their problem. So it is salesy or sales ish? I don't mean salesy in the sense that it sounds like a sales call, but the outcome is that they might be sold on the product. So it's really nice to have these experts share their knowledge and I could take it and actually apply it. Rob taught me that it matters what awareness stage prospects are at talking to somebody who is completely unaware that they have a certain problem.

That's complicated. You have to educate them on the problem, on the solution, on your solution. And the fact that they need it is a bit much more than somebody who already knows they have an issue. They know they're looking for a solution, and they just try to figure out which one to use. They're much more easy to communicate with for somebody in my position.

And Stephen taught me that there is a way to structure these calls. You can set an agenda. You can find alignment through a certain couple of steps and then create these situations of genuine connection over somebody's challenges that work out in a positive way. If there is alignment and you can just say stop and no, if there's not. These things, they don't have to force anything.

It's really nice, really helpful. And the video recordings of these talks, they should be out soon, I guess, on the microconf.com website. Highly recommend for you to check these out. They, I think, are $50. You can still buy the ones from last year.

If you do, you'll have a talk with me in it as well. I gave a talk at Microconf in Denver last year, so that might be interesting, too. But, you know, just look at the ones for this year. If you look into. If you want to look into customer awareness and sales, there was a lot of good stuff about sales.

And if you're anything like me, then sales is kind of weird and hard, and talking to people is hard, and getting them to open up is hard. These talks will help you. Certainly help me. So two days worth of lectures and workshops for $50, that's a steal. You don't even have to pay for a flight or hotel.

That's quite useful. Anyhow, I'm really getting the hang of talking to my users, which is nice. And I just ask them flat out what their job to be done is. I don't necessarily use that language because I don't know if they understand the concept, but I asked them what's the outcome of this? What do you need?

Who are you reporting to? Which often is the case in agencies. There's a structure. Who are you doing this for? What do they expect to see?

What do you get when you get into the job, and what do you need to present when you come out of the job? I ask for the workflow that they have, how they accomplish it right now, whether they consider part scan as a solution to it at all. You need to figure out where they are in the scale and what's still standing in the way of getting their dream results. It's always nice. That is one of the things that I've noticed.

Getting them to explore the perfect state is really fun. Like they make up all these things, these little things that work so well. Like they describe to you a state where there's no issue along the way, and this is the perfect outcome. It has everything I want. And then they kind of dial it back a little and you see the points of friction.

In reality, they often say, well, here's where it gets hard, here's where there's friction. Here's where I have to take an hour to sift through the AI generated results and, you know, remove the weird ones that you see the points of friction and what they struggle with, and you can then actively ask questions about why that sucks, what solutions they tried, what the errors were, what other tools they use, how yours compares to theirs and all that. Really, really useful. So that's why I do it. And often these conversations surface some underlying assumption that neither I nor they themselves have ever actively reflected on.

I highly recommend recording these calls, or at least transcribing them as they happen. I use otter AI and grain, two tools for this, for post processing these calls, and both of these tools offer action items that some kind of AI grabs from the conversation and then tells you what to do next. Which is nice because you don't need to take notes during the conversation. The AI kinda pulls it out for you. And that is highly recommended.

And I think I'm glad I made my choice for who my ideal customer profile is a few weeks ago, quite publicly here on this podcast. It's been helping me a lot in figuring out which feature requests and which ideas I say yes and no to and how I prioritize them. I'm doubling down on creating a really spectacular API experience for people who built their tools on podcasts. The Podscan podcast platform, let's just call it that, which then in turn makes it easier for me to build my own features in the alerting and searching sections of Podscan. So that's kind of how this is working.

I built a lot of API stuff, makes it very simple for me to then build more features for my agency clients. And what I've noticed is that my customer service load is quite low at the moment. Every other day, sometimes not for two or three days in a row, I get a message which I can usually quickly respond to. It's never really a troublesome thing. It's just like somebody had a slight issue with the UI, or somebody has a data question or search results didn't come up the way they thought they would, and then, you know, it's just a couple quotation marks away from working.

It's always little stuff, and it's really cool to see that most things actually work out well. But I can do more in that regard. I can start a knowledge base and make it easier, record videos for people to just watch a two minute thing which really clearly shows how things should work. That kind of stuff. That is what I'm going to do.

But customer service load quite low at the moment. It's not a complicated tool, and I think the rest API, the more complicated technology part, quite literally documents itself. At this point I'm using Laravel side framework or a certain kind of library, a plugin that does all the documentation for the API for me, and it's always up to date. It pulls the data straight from the code. It's quite, quite useful.

I may talk about this in the future, but yeah, it's not much in terms of customer service. I talk more to customers in my exploration calls than I talk to them in customer service at this moment, which I want to keep going for as long as I possibly can. Let me finish this update with a look at my expectations and the reality of my experience and kind of how they differ to one KMR within a month from last time I checked in. And that happened. It happened just so.

I think I hit the goal one day before the month was up. And in a way I think that was a bit too stressful, like setting that goal. It was quite, quite close to not getting it and it probably would have sucked a little bit. So if I set my own deadlines and goals, I might just set them in a way that doesn't induce anxiety in me. So I'll skip setting a new MRR goal for the next month, because why would I need to?

And instead make it about something else, make it about getting several, maybe five really amazing customer testimonials. Just people telling me in video or in writing, what they like about the product and that they like it. That's what I want. I want to build a product that's so good that people are willing to publicly commit to using it. And those testimonies I could put on my landing page can print them out, hang on my wall, and use them to fuel my motivation when I need it.

I think it's much better than an arbitrary number to reach because that's also very internal. It's about oh yeah, I got to that number, but d liking my product, talking about it, that's a better goal for the business. One thing that was a bit underwhelming this month was my experiment with cold outreach. I'll have to talk about this in a future update, but let's just say I tried to find podcast booking agencies and send them personal emails, but kind of cold didn't go far or not far enough. I still struggle with attribution at this point, like understanding where my customers come from.

And that's also something from Microconf Asia. Orangio taught me that that that is required and really, really important to really understand, like where your customers come from. Again, get the, get the talks, get the videos, and attribution is on my list of things to fix. I need to know more about where my customers find me, where they find the business, when they make that choice, and then what kind of customers make which kind of choices. So segmentation and attribution arising on my priority list.

And they also expected the business to be a bit calmer than it is right now. To be quite honest. I guess it's unsurprising that spending a week in Atlanta with my SaaS founder nerd friends cost a few things to pile up on the coding side and on the, you know, the product side. And even this week I didn't get to do all the things that I wanted because of the calls and recordings and podcast related stuff that I also didn't really prepare for. It took a whole week to go to a two day conference.

So with calming and cooling down and recovering and all that, as an introvert, those things usually take quite a toll on me. And so I'm kind of out for two or three days after and then as travel. So I should have structured this a little bit better, particularly with the podcast that always requires a lot of my attention that I did not prepare for. So next time I only need to create more time for these things prior to and then after to get back in my flow state, particularly with Podscan I really want to be in my coding flow state, and that is hard to reach if I'm not prepared for it, if other things are piled up and they kind of, you know, gnawing on my attention. But hey, things are going great and I'm excited to tackle month number three.

I'll be working on new AI features this month. There's going to be summarization, entities, extraction, all that stuff. I'll streamline ux, start working with experts in there that regard. I'll update my landing page, do more marketing, create videos, start a knowledge base that's also on the menu for me and so much more. There are many things and I cannot wait to share them all with you, as usual, here on the pod, on Twitter, and all over the place.

So let's get back to work. And that's it for today. I want to briefly thank my sponsor, acquire.com. It's always a pleasure to work with the people at acquire, and they've been so helpful. I talked to so many people at Microconf about acquisition and acquire.com always comes up, not just because they have a really good team that helps you with getting your acquisition figured out.

They're just very present in our community and that is, as somebody who's building in public, I really admire that because there's so many people who are building businesses who they don't really think about the happy path and the sad path. What might happen if I become super successful? What am I going to do in two years? Is it going to be something that I'm going to sell? Is the business something that I want to be acquired with, or might I lose interest at some point and then I have to sell it?

What can I do to prepare? Well, that's what the people over at acquire can help you with. And they have helped hundreds, if not thousands of people at this point set their business up for a successful acquisition. The best part about this is you don't need to want to sell your business right now to get started. Setting it up for a good acquisition, right?

It's all about preparation. It's about documenting the things you need to document, setting up standard operating procedures, getting your affairs in order, having a p and l ready for whenever you might need it. All of these things, and way more than what I just said, they know how to do, they know how to teach you, and then you will have it for whatever contingency might happen along your way. So go to try dot acquire.com arvid and see for yourself if acquired.com and selling your business all in itself is the right option for you either today or in the future. I bet it is, so why not check it out?

Thank you for listening to the Bootstrap founder today you can find me on Twitter at avidkall arvadkahl and you find my books on my Twitter course. That too. If you want to support me and the show, please subscribe to my YouTube channel because that exists. Get a podcast, this particular podcast in your podcast player of choice, and leave a rating and a review by going to ratethispodcast.com founder. It makes a massive difference if you show up there because then the podcast will show up in other people's feeds and that's where I would love for it to be.

Any of this will help the show. Check out partscan fm if you want to. I appreciate that too, obviously. Thank you so much for listening. Have a wonderful day and bye.