People keep asking how I can be stupid enough to found a SaaS in 2026. Here's my answer.

You definitely know how to bait us into reading. Still, I agree a 100 percent with you

Love seeing the progress on this. The UI looks clean, and the value proposition is super clear. Scaling organic traffic is a marathon, but tools like this definitely make it feel more like a sprint. Keep up the great work!

I agree with you on this! I was nervous at first to dive into building my ecsystem of 6 apps, but as you stated the numbers dont lie & there is plenty of room and need in the market for saas apps.

My new project Emusic Tools, search it on Google if you want

i agree a 100 percent

Well as someone who only recently got into building my own saas, I can definitely say that this post is reassuring. Thanks for sharing

I'd push back on the framing a bit. the 1999-2018 examples were all software replacing software. AI agents replacing apps is reasoning replacing logic - different category. the survivors might look pretty different from traditional SaaS.

Interesting and thought provoking. The value proposition is indeed clear.

So, should you have built it in 2025? Loll just kidding.

I don't believe SaaS is dead. I think this is actually the time they shine brighter with AI.

The moat shift is the real story here. Used to be feature complexity bought you years of runway. Now? You've got maybe months before AI can replicate the feature set. What actually sticks is workflow lock-in and proprietary data quality. IbexAI's real defensibility isn't "we scrape LinkedIn" — it's the signal quality and downstream integrations. Same reason court recording tools (my domain) will never be commoditized by agents. The value is in the specialized data + the workflow, not the UI.

The definitional point deserves more emphasis than it's getting. SaaS is a delivery model and billing structure, not a feature set. Cursor, ChatGPT, Claude — they're all SaaS. The "AI kills SaaS" argument is actually an argument about interface evolution, not business model extinction.

What does shift from an investor perspective: the moat calculus. Pre-AI, you could build a defensible product on feature complexity alone — it took time to copy. Now a well-funded competitor can replicate your feature set in months. So the moat has moved. It's distribution, proprietary data, and workflow lock-in. If you're building IbexAI on top of LinkedIn signal, the moat is the signal quality and the downstream workflow integrations — not the scraping layer itself.

That's actually a more interesting investment thesis than "we built X feature first." Founders who understand that are building differently than they were 3 years ago. The ones still optimizing for feature completeness are the ones who should be worried.

This hits different because most people saying SaaS is dead are not actually building anything. They are just commenting on it from the sidelines. The market is more crowded, sure, but the bar for quality has also never been lower. Most SaaS products still have terrible onboarding, confusing pricing pages, and zero personality. If you actually care about the customer experience you can still win. I run a marketing agency and the SaaS clients who are growing right now all have one thing in common: they talked to their customers before building. That is the whole secret.

The definitional point is the strongest one and it's underused. People conflating ' AI is replacing Saas with AI products are not SaaS' are making a categroy error. Cursor is Saas and the argument defeats itself . The more interesting question is whether the competitive dynamics inside SAAS change - and they do. Switching costs drop when AI can replicate features faster, commoditization pressure increases and the moat shifts from functionality to distribution and data.

This cycle happens with every wave new tech doesn’t replace the model, it reshapes it. SaaS isn’t dying, it’s evolving. The real gap is still in how founders build and launch these products efficiently. We've been seeing this firsthand while structuring product workflows at FoundersBar.

HI, i'm Connell, a Bug Bounty Hunter. I run PentraSec, a Web Application Security boutique, similar to platforms like HackerOne and Bugcroud but more personalised, meaning more focus on in-scope applications to test.

I'd be happy to partner with you to test your applications before attackers do; leaving you to focus on the development while i cover your unintentionally open paths and footprints attackers could crawl into, and making sure you stay trusted and guaranteed by your clients.

I was thinking about this recently. With AI tools making it so much easier to ship a product, the market is changing. It's much less about finding the "best" tool and much more about finding the right tool for your specific situation. We aren't just stuck with massive enterprise options that cost a fortune and have hundreds of features we don't need. If you're a hairdresser, you don't have to try and make Salesforce work for you. There's almost certainly a hairdresser specific CRM that gives you exactly what you want without everything you don't. Everything is more niche and more focused. Tools that solve 1 problem for 1 market will always be valuable.

I’m building a small AI tool myself and keep hearing the same thing: “AI agents will replace apps.”

But when I talk to real users, what they want isn’t an agent — they want a clear workflow and outcome.

• Improve my resume

• Find better leads

• Prepare for an interview

Agents might power the backend, but the product layer still matters a lot.

Feels like SaaS is just evolving, not disappearing.

I've literally spent the last 4 weeks building a new SaaS product (Homeggo - helping home buyers better organize their search) and have been asking myself daily is it worth it. But the amount of time I've been working with AI to build it tells me that more goes into the creation of a quality SaaS then just the code. Wild how I switch the hats I wear (product designer, developer, tester, sys admin) minute by minute as I've built it.

"The average enterprise uses 275 SaaS apps" WHATT??? That's an insane number hahaha

The definition collapse is the sharpest point here. Cursor is SaaS, ChatGPT is SaaS, Claude Code is SaaS. The 'AI kills SaaS' argument is mostly people confusing the interface layer with the business model layer. The interface is changing. The subscription model powering it is identical. What actually changes is the surface area of what's worth building, not whether building is worth doing. The 85% of the world that has never opened ChatGPT stat is the one that reframes the whole conversation. Most SaaS death narratives are written by people who live in a tech bubble where everyone has already adopted everything. The actual market is mostly people who haven't started yet.

Good progress