Imagine that you have the opportunity to reimplement Kinetic (in SaaS) from the ground up - new business processes and no legacy constraints.
What would you do differently compared to your current setup?
What pain points would you make sure not to repeat?
What are some neat tips and tricks you’ve discovered?
Triple check everything related to OUM…
Enforce no testing/dev in live much more strictly
Formalize sensible part numbers…
Implement a sensible menu/security strategy
Pain points…a few but some were self-incurred to be honest (and that’s likely the case with every ERP implementation in history). I’ll pick three.
EDI setup - was sold on OOTB setups being available…and that was far from the case (though once we finally got everything in place, it’s performed well for us)
Smart Forecasting - we’ve got multiple business models so there’s no way for us to do a “one size fits all” forecasting setup.
Spreadsheet Server for reporting - already retired it
Hmmm…Kinetic seems to be working OK for us…the one-time shipment stuff works well for our ecomm/pick-and-pack customers, those doing EDI are pretty much locked down with a one-to-one relationship based on location code.
Having tons of duplicate Ship To Entries for the main location of a customer and you can’t delete any of them as they have shipments against them is annoying. Doesn’t cause a serious problem, its just clutter.
Ohhh…you mean the ones created when someone decides to embed a specific PO number or a customer contact name into the address field. Yeah, I can see that.
They don’t want to create contacts, so you have 10 different versions, one for ATTN Joe and one for ATTN Bob… because creating ship to’s is easier than creating contacts??
Start with Global Suppliers, Customers, and Parts-We utilize suppliers and customers but converting/combining after bringing in multiple acquisitions old codes was not clean
UOM/UOM Class Setup to account for our growth and differences in businesses
True. And good luck with that. Not just the tribal resistance, etc. Just the fact that the company doesn’t even know what they are doing that’s so crazy until they begin to see that there are better ways… years into Kinetic forcing you to deal with your existing “procedures.” (Emphasis on the scare quotes.)
I’ve been meaning to post about this, but a project I am finally underway with is to make “Department Hubs” which are Kinetic apps that are like the homepages that I wish people had. Plus descriptions of why you would use this screen or that screen to do your job. And even links to a PPT guide for that thing.
Point is, even though half of these buttons are for dashboards that I made, I still wish I had started with a basic version of this on day zero.
Of course it really only became possible/practical to do this once App Studio was introduced (…and the bugs were worked out).
Occasionally I add a paragraph or two of a rant, even without a button nearby.
And then there are BAQs at the bottom of stuff that people should be keeping an eye on.
I bet that the 4 BAQs in this pic all existed (in some form) within a few months of our go-live. I hardly knew anything back then, but even at the outset, I knew what stuff was being overlooked or needed babysitting (like new parts). It hasn’t changed.
Ahem, block leading spaces in bin names and other terrible choices for IDs/PKs. (PAY SOMEONE to do this if you don’t have the knowhow. But I guarantee there is code here for it already.)
When a BAQ gets complicated, filter it like this. I’m never going back to the normal way.
Love all of the above. I would also snuff out “smart” part numbering in favor of a simple system like:
000-0000
First three could represent some information but the final four are sequential.
Fast to type
Easy to check with BPM
barcode friendly
web friendly
sorta sortable
And I would build a dev system at the same time with sample data similar to Epicors Education db. It could be used for training and development/testing. Much more secure than working with production data.