If you have a CRM, do you own your data or does Salesforce?
Microsoft Teams or Office365 or Google Workspace – where are all the emails and messages stored?
All of your SaaS applications hold your data hostage. How do you get at it? You pay for access to your own data.
It isn’t easy to access your data. You have to do it inside the software. Even with “open” APIs can you pull all of your data into your own SQL database? Doubtful.
These software companies pivoted to platforms and added marketplaces. All that means is that plug-ins and tools you pay for to access or analyze your own data also provide the SaaS provider with additional revenue.
All of this wasn’t too important until AI showed up. Every AI project is actually a Data Project.
The whole idea of an MCP server is like an API: an exchange point for the LLM or RAG to access another software system. This was developed (quickly I might add) because LLMs need access to data. That data is stored in silos.
In many cases, if you want to analyze your customer data, you buy a tool or add-on from the Salesforce marketplace or the CCaaS provider sells you an add-on to see your CX data and conversations.
This will get expensive fast. And this doesn’t really solve the silo issue. Your customer data is still only available from the SaaS platform. And it is in 2 different places. All those conversations need to be stored and transcribed with the CRM data. (You can do this with vCons.) But the conversations are in your call recording silo with your contact center vendor and your other customer data is in your CRM software – way over there. Do you see how inefficient and ridiculous this is?
Most companies pay about $199 per head for CRM plus $150 for a CCaaS seat plus $10 for a Teams license. That’s 3 silos of data. Data that you have to pay to access.
I think SaaS will eventually lose because enterprise will move to a data lake and personalized user interfaces provided by Lovable or some other AI.