The Bootstrapped Founder

Hey. It's Arvid, and this is the Bootstrap founder. When I was running FeedbackPanda back in 2017, '20 '19, I developed what can only be described as a Pavlovian anxiety reflex to error notifications. I still very vividly feel this and it's something I wanna talk about, not necessarily about what happened then, but how I'm dealing with it now. Because every ping, every email notification, every Slack thing, whatever could mean disaster back in the day because I was the only technical person in the business and I was responsible for maintaining many integrations and stability of the platform, but mostly integrations with a third party platform that we had no formal agreement with.

So they would just change their website. We were integrating directly into a web application. So, you know, how quickly this has changed. I deploy 20 times a day and they did too. So whenever they push new code, I had to react and to be able to learn when new code was pushed usually I would have to wait for an error message coming in.

So this feeling of an error message landing and the ding I got, the notification I got, that kind of did something to me. It was this constant cat and mouse game of fixing our integration whenever they pushed updates. And I had Sentry tracking both in the front end and back end errors for that platform. It was an Elixir JavaScript thing, so most errors that were happening related to this particular issue or JavaScript errors and, you know, how quickly JavaScript errors can happen. I would get notified for every single error.

Each notification was potentially catastrophic because back then being the only person running the technical side of the business, it could signal a change that would make it impossible for our customers to use the product. And that is a problem, right? If they have an integration into some kind of web application and the integration breaks, they cannot use the product. So I would have to react very quickly. Notifications had to stay on.

But over time, this vigilance and being able to listen to notifications at all times took its toll on me. I I had physical implications from this. I felt, like, physically bad. I got ill from this and it took me years to get out of this. And even couple years after selling FeedbackPanda in 2019, that anxiety still lingered.

Maybe not as strongly, but it was still there. Do you know this distinctive sound of the intercom chat widget that just opens and does this little thing? That would and still does send a jolt through my system. I still feel it viscerally when that ding happens. Even though now, it just opens on a page and somebody's trying to sell me something.

But, you know, it still is there. And starting a new SaaS business certainly didn't help. When I started PodScan and integrated a new customer chat service tool, I used Krisp this time around. That didn't really help. The anxiety isn't as intense as it once was, but I still feel that urgent pull to respond immediately to notifications.

And as PodScan has grown and more customers reach out through all the various channels, email and that thing, I've had to make this conscious effort to not let notifications overwhelm me. The reality is that notification management is attention management. Every notification interrupts my focus and pulls me from what I know I should be working on into what somebody else wants me to work on. Their problem, their issue. When the customer email or a chat message comes in their problem takes priority.

Even if what I'm building might have a much higher impact on the business. And the conventional wisdom out there is to outsource customer service or implement AI systems. But in my case, the volume of inquiries that I get is too low to Right? I get, like, maybe

10 messages a day. That's not what a person is gonna sit

in front of a computer for. But most questions come from users with specific technical needs about our API integrations or advanced features. And those conversations are not really helped by an AI just yet. As long as I still build the software and AI doesn't build it all by itself, I can respond to these questions better. And these conversations are helpful and valuable.

They help me build deeper relationships with my customers and improve the product at the same time. It's a cyclical dialogue between me, the owner of the business, and the users shaping the product and its use. But still every notification, even the good ones, is an interruption. So over the past couple weeks I've developed a new approach here. Whenever I handle a notification, I then examine the underlying rule and the medium in which it comes.

If immediate action is required, like the server is down, the domain unresponsive or the database failing, then yes, I'll keep the notification and I want it to notify me instantly. But if no immediate action is needed, I just log it and I turn off notifications and then I set regular times to review these logs. This could mean messages that go into Slack without notifications or emails that don't trigger phone alerts now exist where before every single thing would notify me. The key here, what I'm trying to do is to protect my focus time and only allow truly urgent matters to actually pierce that veil, that protective shell around me and my attention. Customer issues, critical customer issues, system failures, or bugs that break core functionality, yes.

These warrant immediate action. I wanna know about this. But a slightly slow database on a big day with a lot of customers or a customer responding to a week old marketing email, that can wait. That doesn't need to pull me out of what I'm currently doing. And I've applied this thinking to PodScan's various integrations that I have just to do all kinds of things related to customer acquisition.

We use Koala, which is a tool for tracking high intent prospects. And while that's super valuable and always interesting to look into none of the insights of that tool require my immediate attention. They can wait for my designated review time where I can properly process whatever happened into actionable tasks. I don't need to be constantly pulled out of what I'm doing by this tool telling me that somebody from company x y z is on the platform. Great.

And I can strike up a chat with them, but should this interrupt the three hour debugging session that I was in? Probably not. So the challenge is not in muting notifications. I think we can all do this. Like everybody knows how to turn off a notification and how to not be bothered by it.

It's very simple. But the real test is to consistently review the non urgent logs and the messages without letting them pile up. I don't know if you're like me, but I'm like inbox 400,000, not inbox zero. Like, if things are not urgent or important, I just ignore them. And then I have a hard time go through all of these things at any given time because I'm easily drawn into the exciting work of product development or talking to customers and all that, but maintaining discipline with email management is an ongoing struggle for me.

I haven't yet found the perfect accountability system to keep me on track. So if you have any ideas, let me know. Some founders, and here's the thing that this whole attention management thing is about, really excel at establishing routines like this checking in on stuff every couple days while others like myself need more structure, we need more incentives. So for those who might forget to check you should think about staggered notifications. Things that arrive a little bit later or at a certain time like a daily summary similar to Apple's notification digest that kind of stuff.

That can provide helpful middle ground. You don't get stuff immediately because it's not critical, but you still get a notification for things that happened. And I've also learned to be strategic with the threshold of what I actually monitor. Because if you run a software business or any business that has internal metrics, which is every business, you need to look at certain thresholds and see if they're reached or if you're still within the good kind of numbers or if you're having problems. So if churn is over 30% you have a problem, so you might want to monitor churn.

Or if your database has outages, you have had outages before, maybe you wanna see what the latency is. Right? And you wanna keep within like two hundred milliseconds or something and you monitor that. There's always stuff to monitor. So instead of getting alerted for every minor fluctuation in my system, I have set up observers that notify me when critical metrics cross important boundaries over a floating amount of time.

On AWS, just to give you an example of how these things work, you have budgets alerts, right, for how much you pay in a month. You can set it up to be, like, I wanna spend at max a thousand dollars and tell me at 90% that I'm getting close to my monthly budget. That you can do. And you do the same thing for application monitoring. You track average values and ninetieth percentile metrics for things like latency or ensuring that you know exactly what your computation efforts are.

There are many many things that I'm informed about before small issues become major problems and that I wanna be informed about before they actually escalate. And recently, another thing that I track as internal things for PodScan, I made a very significant change because I had a Slack channel where all the things that happened on the platform would cause a Slack message and that would always notify me. Right? Like, a new user signs up or I get a new subscription or somebody cancels their subscription or somebody creates an alert or somebody does an extensive search, like all of these things would pile in all the time. And while it's exciting to see things like new users join, we're now at over four and a half thousand users on the platform, getting pulled out of focus every couple minutes, every fifteen minutes or so was not serving the business.

Instead, I've split my notification streams in Slack. General events like sign ups, alert creation, deletion, team changes, they go into this log channel that I review periodically with notifications turned off. But customer actions that might need immediate action like, cancellation, subscription changes, trial extensions, stuff like that, still trigger notifications so I can engage at critical moments. Because as much as I wanna focus on a product, my customers pay for this. So I need to be there with them and for them if they do certain things.

And this balance of staying informed without being interrupted has been quite crucial for me for maintaining both the productivity that I need to build something bigger and better and the peace of mind that I need to not go crazy over this. It's like always about being intentional with our attention and recognizing that not every ping deserves an immediate response. So think about how you can reduce the number of notifications without reducing the insight that you get from those notifications. That will really really impact how you spend your time, how focused you are on the things that you actually want to be spending your time on and what things you actually react to. Because your attention is probably most valuable commodity in your business.

And that's it for today. Thank you so much for listening to the Booster Founder. You can find me on Twitter at AvidKahl, a

r v I d k a h l. If you wanna support me on

the show, please share podscan.fm with your professional peers and those who you think will benefit from tracking mentions of their brands, businesses, and names on all podcasts out there. PodScan is this near real time database with a stellar API or so I'm told by my customers, so please share the word with those who need to stay on top of the podcasting ecosystem. Thank you so much for listening. Have a wonderful day, and bye bye.