My personal setup is to create the app/design as a prototype with Bubble and move out after proven QA & User engagement. Sell the app as a prototype with the goal to scale/adapt.

If the tool is for internal use, then it is easier to build and sustain on Bubble, thus making the sale of a “Bubble app” easier, rather than a flow-blown app that takes a mass-amount of traffic. This will lead to optimization, responsiveness, and reliability issues in the long-run, which will be harder to sell. I do not believe Bubble’s WU pricing model is in any means comparable to pricing compared to a traditional/alternative engine running a “massive application”. Bubble charges about 1000% more for simple actions that use little to no resources in the real-world of coding/server ops. Not trying to bash on Bubble, as they did the work, and you (not you yourself, but Bubble app users as a whole) want an easy way to build an app, so you get to pay for all the fancy shiny digery-doo-dads that make that happen for you.