Sorry Apps...The End Is Near
The app economy is rapidly heading to an end.
Sensational? Perhaps, but bear with me for a moment.
I’ve spoken before about how most of the enterprise software and SaaS services we use now are going to be deprecated to being data or databases as AI tools and Agents remove abstraction layers and enable a single pane of glass type of use case where we can leverage multi-modal generative tools to build a real time UI for a specific query or use case. (Think search with Gen AI for business)
Now, move your thinking over to the app experience that most of us have today. The way we interact with our mobile devices and tablets. Many of us use dozens if not hundreds of applications for everything from banking to travel to work to game and so on and so forth.
To make this work, we organize our apps in folders and we use FaceID to give us easy login access and we permit these apps unadulterated access to our data to make sure the experiences are good—often with little regard to the way the data is utilized (a topic for another day). The app builders have built apps on making better UI/UX and connecting to data services like ride share, hotel and flights, food, and investment services etc.
Now, imagine in the future on an AI Powered Device with a powerful SoC with the GPU/NPU and of course amazing connectivity that allows digital assistants and agents to sit at the OS level. You can just talk to your digital assistant, IMO this is where Apple Intelligence has a big opportunity, and Siri or whatever it is ultimately named can understand your request and go out to not just the apps you have, but cross reference the apps across all of the available data services to deliver the best experience.
Recommended by LinkedIn
An example. Travel Apps
Right now, if you want to book a trip you may you an aggregator service like Google Flights or an app like Expedia. These apps have access to a broad set of data and a UI that makes it easy to book a flight or a trip. As users we are limited to the data services of that particular app and/or a group of apps that we may use to compare to one another (like Google Flights or lending tree).
In the future, we will be able to use an O.S. Level digital assistant and multi-modal (text, speech) to just tell it what you need. I need a flight to SF, a hotel until Thursday, a driver to pick me up when I arrive, and a reservation for a dinner meeting on Wednesday. Also, can you confirm my meetings and so on and so forth.
The assistant + agent technology will be able to navigate your preferences like which airline you prefer to fly based upon status, budget consciousness, and also based upon seat preference and availability. It will be able to complete the task including pay for you securely. It can choose your favorite ride share or car livery service app and your preferred hotel destination, room type, amenities and services and location based upon the meetings. This will also work for your restaurant booking and of course the meeting confirmations will be easy, but always useful to be sure you are traveling for good reason.
Furthermore, it can leverage/find data and services that you don’t already have an app for. Finding better prices on your shopping or travel, or even suggesting a better dinner location based upon the preferences of the person you are meeting.
So why do we need the apps? At least we know it? The apps are deprecated data services and user interfaces that can be generated in real time? Do the apps go away or do they become more of a data layer like our business apps are likely to become. My preference isn’t the app, it is the experience that the app creates. Hail my ride, prepare my coffee, show me my account balance.
The UI/UX will be an abstraction of the combined open data ecosystem and the ability for us to articulate the need. But we aren’t far away from a multi-modal future where we simply communicate what we need and let the agents do the work—apps become a figment of the past.
Maybe, but businesses have been telling us the Mainframe was dead 20 years ago, but here we are still doing transactions on the mainframe in the biggest banks, insurance and healthcare companies in the world.
Global Director & GM, Strategic Clients Group at Oracle | Global Sales & Go-To-Market Leader, SaaS & Cloud Solutions (AI, ML & Data Analytics)
1moI totally align with your POV, Daniel. As you know companies who have not modernized critical business functions to SaaS (and there are many F500), is mostly due to the data complexities from siloed legacy systems and some SaaS applications. For those that have done so, to your point, the SaaS application unfortunately still serves as a like database or modular app. I’ve been coaching my teams for very long time to not sell their SaaS widgets they are responsible for, but seek to first identify the key business data challenges, use cases and desired outcomes. The approach not only helps separate sellers from the competition (product versus product) by changing the perception from a transactional seller to a trusted advisor BUT also enables the SaaS product desired to be sold to become the bi-product of the data use case/s aligned upon. Holistic customer value sold and bigger strategic deals earned with Cloud/AI = WIN WIN!