On the era of SaaS, I see many companies hiring a lot of SaaS services. They create a dependency on them for their production operations.

It may seems the company is saving work and being modern when joining e-commerce portals with SaaS ERP’s and other online providers. Event-driven architectures are modern, aren’t them? They can receive data from one provider, easily transform and redirect the information to another one.

The problem is: Where will be the company data?

I see many companies leaving their data in the hands of SaaS services.

Call e-commerce API’s to register your products. Read from e-commerce API’s and send directly to other SaaS ERP’s for management. Read from these 3rd party services and send directly to other 3rd parties responsible for product delivery. It seems modern, but in the end, where is the company’s data?

The company will only be moving data from one 3rd party provider to another and always be subject to the format and the data availability of these 3rd party providers.

These are the companies who look for us asking “I would like to use BI and AI to improve my business.”

Great. But where is your data? Where is your database? They don’t have.

Some of these companies even believe they are being modern by using event driven architecture. Recovering data from one external provider, transforming the format, and directly sending to another provider.

In fact, they are trusting their most important resource, their data, to external providers who use different formats, models, and communications, making the company totally subject to 3rd party providers.

The realization these companies really miss is the difference between a production environment and the data analytics. Building a data platform intelligence over an environment like this is like building a house over sand. No solid foundation.

Conclusion

Convincing these companies about the need to take ownership of their own data to build a solution foundation for a data intelligence platform can be challenging, but fundamental for a solid data strategy.

I would love to hear your own experiences and opinions about scenarios like these.