New team doesn't think they need help
I have a new team.
They are not using any process but believe they are working in an Agile environment.
Without a Backlog they have 10s of items which are all non-functional e.g. refactoring work, updating libraries etc.
They are picking up and creating items adhoc.
PO is not pushing any direction.
How do I get the PO to create a Backlog that is of value to the user?
How do I get the devs to deliver features per sprint?
How do I get the PO to create a Backlog that is of value to the user?
Who is currently holding the PO accountable for maximizing product value?
How do I get the devs to deliver features per sprint?
Start with what it means for work of any sort to be "Done" and finished. Are the team satisfying a Definition of Done that is of immediate release quality?
No one is holding the PO accountable, who should?
I'll have to look at the PBIs marked Done, I'd bet it's not at DoD, the excuse will be that these are only parts of the larger solution and not demoable until the entire functionality is complete.
Features are too large and the end value to user is unclear
The way of working sounds like a lean try of kanban. And for the work they are currently doing wouldn't kanban fit, until they work on functionality?
I suppose your role is the SM?
Why not try to use Kanban as long as they are working on non-functional and guide them to a set of processes and a DoD, not only to go Kanban, but also preparing them for Scrum. In the meantime work with the PO and as soon as the PBL emerges with functional work, switch to Scrum with a bit of Kanban influence.
I suggest to focus on the causes behind the symptoms you mentioned.
A team focussing on non-functional items might either face a huge technical debt or a lack of phantasy on which features to include into the product.
A PO that doesn't give any guidance either might have resignated (maybe, because a large technical dept significally hinders feature development) or he really doesn't know by himself, which feature to provide next.
For both cases I suggest to support the PO to either craft (if not yet existant) or question (if already existant) the product vision.
This might help in both aspects: Identifying and prioritizing new features as well as to identify the most critical technical debt.
Has the current PO overseen the development of the product "from scratch" or has he "inherited" something he now has to find his own way coping with?