top of page
Search

Why does Saas even exist? Will Ai eat Saas?

  • Feb 18
  • 3 min read

Updated: Feb 19


Why does Saas even exist in 2026 and that too for multiple decades? Will Ai eat Saas? 


  • For decades, Saas gave convenience to organisations. You need not build, deploy on cloud, maintain and update/ upgrade, worry about privacy, security and more. Industry best practices included. And in most cases the Saas cost is ok unless you become a complex web of integrations over the years.


    Example - quick commerce in India. I will charge you a 10 Rs handling fee but you need not get off your couch, take your bike, face sun/ rain, pay for bike petrol, take stairs (busy elevators), find aisles,  shop, wait at POS, digest the stares from security personnel at checkout, etc. Saas gave similar convenience to organisations. 


  • Saas nailed key business workflows, transactions etc. and capitalised with features. Let's not look at adoption from a volume perspective. If it is legal or contract Saas, the transaction volume is bound to be low … 🙋 


  • Saas over decades hooked professionals, communities and helped them build careers and life. This is a huge catalyst and glue. It became a system of record, globalization/ localisation and compliance included.  



Now, what would it take for Ai or anyone else to replace Saas? 


  • Most six-figure Saas are sold based on contracts. Not pleasant if you have to tinker with those contracts. I wonder how Ai can help here?  

  • Saas pricing is flexible, seat based. It can be moved up and down. A lot of Saas players also give consumption/ transaction based pricing. It is appropriate to have a base price with a consumption top up. Outcome based pricing is also possible if customer and saas partners are well aligned. Hmm - some of the world’s biggest Saas is offered as a part of bundle/ consolidation. Seriously, I have no clue how to fight with a big boy bundle.      

  • Ai can reduce/ optimise the cost of software development, however the token costs, tool costs, set up time may mount up. Will you train using synthetic data? If data is not there, traditional players will charge for their data/ api access. Vertical Saas is domain focused. Example - a Life sciences Saas. Does any AI company own a database? Do they own data not synthetic data? … 💡 

  • In case of custom build with ai-native players there are additional costs such as change management, training, migration, integrations, setup, onboarding, legal, privacy and so on. Ai may reduce these in coming years, but they will still remain. Additionally, the Ai-native tools have to be enterprise ready - I mean really ready.  

  • Companies can try to build with open source Ai models. If IT development is the core capability of your organisation, sure. However, I do not see why a Finance or Manufacturing company will build a custom software. I hope you are aware that the average mid-size, enterprise organisation has 26 to 100 plus Saas tools across all departments. Ok, good luck with replacing all of those. 


I suggest you try building a Slack alternative with open source chat apps in your company. Slack fits all criteria for replacement. You don’t have sensitive data, integrations in slack, do you? … 🎉 


  • Lastly, can Ai pose a threat to Saas distribution. Saas players have thousands of customers, professionals and decades of knowledge. Even with Ai, can that distribution be hacked in time? On-prem is also a solution. 


One last thing - to contrast the current sentiment regarding Saas stock prices, multiples, etc.; keep in mind that some of the Saas companies are doing the rule of 50. Seriously, the rule of 40 plus 41, 42, 43 and then fifty. It is amazing to be in that saas business. 


So, will Ai kill Saas? I believe software did not eat the world, it became part of the better world which is growing and scaling.  


  • So will Ai kill Saas? 

  • Will Saas plus Ai kill Saas? 

  • Will Ai kill Ai? Think about it - AGI, ASI. 

  • Will Saas + Ai co-exist? It is global democracy, all coalitions are welcome.  


In the last two weeks, it is kind of assumed that Ai may try to kill services; because it is the bigger pie. In this sense, instead of who may kill whom, the right question would be - what can existing software and AI together do for humanity? 


Thanks a bunch for reading. 



Saas pricing is flexible, seat based. It can be moved up and down. A lot of Saas players also give consumption/ transaction based pricing. It is appropriate to have a base price with a consumption top up. Outcome based pricing is also possible if customer and saas partners are well aligned.



……. 


Swagat Irsale is Growth Advocate. He works with startups and scale ups to grow revenue and build products which enterprises love to use. 



 
 
 

Comments


bottom of page