Because conda is designed for projects with multiple languages and environments for scientific computing, it's overkill for regular Python projects.
It is great for developers since you can easily switch between complete environments with different versions of packages, for testing and development.
Conda can be installed using a shell script that should work on most systems with minimal initial dependencies.
It’s obviously an added feature compared to pip, but it does get significantly slower with a project with lots of packages.
In pipenv, spinning up an virtual environment for Python 2 or 3 is simply just pipenv --two or pipenv --three.
pipenv may or may not become the de facto standard, but no one knows yet because the project is quite new
So for example it keeps track of subdependencies, subdependencies are uninstalled as well when uninstalling a dependency.
pip is not a package manager, it won't handle revisions or rollbacks https://github.com/pypa/pip/issues/3867