The email subject line said “URGENT: Production down.”
It was 2:34 am on a Saturday. Our entire checkout system had crashed during peak shopping hours. Customer orders backing up. Revenue bleeding. Support tickets are flooding in.
I called our CTO. “What happened?”
“The payment gateway integration failed. Again.”
“Again? We fixed this last month.”
“Yeah. And the month before that. And four months before that.” He sounded exhausted. “Rupesh, we’re not fixing anything. We’re just resuscitating a corpse over and over.”
We got the system back up by 4 am. Cost us about $40,000 in lost sales, plus another $15,000 in emergency developer hours, plus everyone’s Saturday night.
Monday morning, someone suggested we invest in “better monitoring” so we could catch these failures faster next time.
I stared at them. “Or we could stop having a ‘next time.'”
The Conversation That Made It Click
My son brought home a school project about recycling last week. Very proud of himself, he explained the whole thing over dinner.
“Dad, did you know people throw away things that still work just because they’re old?”
“Yeah, that’s wasteful, right?”
“But…” he paused, thinking it through. “What if keeping the old thing costs more than getting a new thing that works better?”
My wife laughed. I just sat there holding my fork because my kid had just described our entire tech strategy.
See, most companies aren’t even keeping things that work. They’re spending a fortune maintaining things that barely function, telling themselves they’re being responsible, when really they’re just scared of change.
What I Learned From Watching Companies Die
I’ve watched three e-commerce companies go under in the last five years. Good brands. Smart founders. Decent products.
Want to know what killed them? Not competition. Not market conditions. Their refusal to let go of systems that stopped working years ago.
One founder spent two years and nearly $500,000 trying to “optimize” a warehouse system from 2016. While he was busy keeping that alive, his competitors launched same-day delivery. By the time he finally rebuilt, customers had already moved on.
The pattern is always the same: they confuse motion with progress. They were doing things, spending money, having meetings. But they were solving problems from three years ago instead of building for three years ahead.
The Uncomfortable Numbers
We’ve been running on a system we built in 2019. Back when we were processing 200 orders a day. Now we’re doing 3,000+.
We’ve spent over $200,000 in the last two years just keeping this thing from catching fire. Companies waste 20% of their software budgets on tools nobody uses. E-commerce brands pay thousands monthly for subscriptions their teams forgot they had.
I audited our spending last month. Found subscriptions to tools we’d replaced but never canceled. $28,000 a year, just gone. Found three different analytics platforms because teams couldn’t agree on one. Found a “critical integration” that we’d moved in-house eight months ago but still paid for.
What Changed
I asked the team: “If we were starting today, would we build any of this?”
The answer was a hard no.
So we stopped funding fixes for systems we wouldn’t build today. Canceled every forgotten subscription. Forced teams to pick one tool instead of maintaining three.
The money we freed up? Let me tell you. This saved up money is now going to the thing we should have built two years ago.
Here’s what nobody tells you: every month you spend maintaining something that shouldn’t exist is a month your competitors aren’t wasting. While you’re debugging code from 2019, someone else is shipping features for 2026.
Your 2026 budget is probably sitting in draft. Before you finalize it, ask yourself: how much of this money is funding the past instead of building the future?
Regards,
Rupesh
Cloverbites By Rupesh Sanghavi 
Leave a Reply