AI-native businesses are evolving rapidly. How do you see artificial intelligence transforming the future of SaaS products and customer experiences?
The first thing changing is the interface itself. SaaS for two decades meant forms, dashboards and humans clicking through workflows.
AI is collapsing all of that into something closer to a conversation. You tell the software what outcome you want, and it figures out the steps, runs them, and reports back. The product stops being a tool you operate and starts behaving like a teammate you delegate to.
Manish Choudhary, Co-Founder and CEO, Flexprice
For customers, the experience is moving from waiting in line to being served before they ask. Think of how ChatGPT remembers your last few chats, how Gmail finishes your sentences, how Spotify builds a playlist before you ask for one. Software is starting to behave like that everywhere, not just inside dedicated AI apps.
This changes what people expect from every product they use. Customers used to write to support and wait two days for a reply. Now an AI agent answers in seconds, knows what they bought, and can refund them on the spot. The bar for good service isn’t what a busy support team can do anymore; it’s what a smart assistant can do for one person at a time.
The companies that win this shift will feel less like software you log into, and more like a service that quietly works around you.
Usage-based and flexible pricing models are gaining momentum globally. What is driving this shift, and how should businesses adapt to it?
Three forces hit at the same time, and the combination is what makes it feel sudden.
- Cost is the first. AI products carry real variable COGS behind every action, and you can’t bury that inside a flat subscription without watching your margins disappear. Vendors are moving to usage-based pricing because the math demands it, not because pricing pages got more creative.
- Finance teams spent the last two years cutting back on software nobody was using. They don’t want to pay for 50 seats when only twenty people actually open the product each month. They want to pay for what they actually use, and usage-based pricing is the simplest way to do that.
- Habit is the third. AWS, Snowflake, Twilio and Stripe taught a whole generation of buyers that paying only for what you use is normal. AI products are showing up in the same world, and the old per-seat model just looks out of place next to a bill that moves with actual usage.
As companies scale digitally, what role does cloud-native infrastructure play in ensuring agility, scalability, and operational efficiency?
Cloud-native is just how most new companies build now, and it’s not hard to see why. You spend less time babysitting servers and more time actually shipping products. A small team can now manage the kind of setup that would have needed a big infra group ten years ago, because things like databases, queues and autoscaling are handled by managed services.
The real unlock for people is how fast you can move now. You ship, see what customers do with it, and decide quickly whether to double down or kill it. That loop is what makes a company feel genuinely quick on its feet.
The catch is over-engineering too early. If you have 3 customers, you don’t need Kubernetes and a dozen microservices. Most of the “cloud-native gone wrong” stories come from teams designing for traffic they don’t have yet.
Open-source ecosystems are becoming increasingly influential in enterprise technology. How do you think open-source innovation is reshaping product development and collaboration?
Look at the biggest infra companies of the last ten years. Databricks, Confluent and others all have open source at the core, and all of them took customers from the closed products they were up against. The same thing happened with AI models last year. When Llama, Mistral and DeepSeek released their weights for free, closed providers had to cut prices almost overnight.
A big part of this is vendor lock-in. Ten years ago, buyers worried that going open source meant more risk. Today they worry more about being stuck on a closed platform with opaque pricing and no easy way out. With open source, there’s at least a clear path to run it yourself, switch vendors, or add your own changes if you have to.
What’s really changed is what people expect by default. 10 years ago, open source had to prove it was safe enough for production. Today the closed vendor is the one who has to explain why their code is hidden, why their roadmap is private, and how you won’t be trapped if your usage grows.
With real-time data and automation becoming central to modern businesses, how can organizations build smarter and more responsive revenue and decision-making systems?
Most companies still run on monthly reporting. By the time the numbers land in a deck, they’re already thirty to sixty days old. That worked when contracts were annual and usage barely moved. It doesn’t work in a world where one customer can burn through their commit in a week, or a model price change overnight can wipe out your margin.
Companies need to move from batch to events. The moment a customer signs up, uses a feature, hits a limit, upgrades, or stops paying, that event should flow into the systems that bill, alert, and decide in seconds, not days or weeks. That’s how you get real-time revenue visibility and real control over what’s happening, instead of running the business on stale numbers.
Most companies stop at a dashboard and call it real-time. A dashboard tells you what happened; a responsive system acts on events as they occur, nudging customers, flagging margin drops, or pausing bad overages so you can correct course while the money is still moving.
Enjoyed this interview? Now imagine yours. Write to:
jeevika@thefoundermedia.in
About The Author
