Has anyone successfully implemented a combined Customer & Product margin cube? I’m hoping to build out ours over the next couple of months and before I get too far in, I’m wondering if you hit any pitfalls I should keep an eye out for.
Hi Jessica.
One of the first decisions is if to do it as DPM or Other Cube. This might be ideal for DPM if you want to make Customer or Project the object of the DPM. Where you’ll run into trouble with DPM is if you want to have multiple products associated with a customer. Manageable with a few products, but near impossible if you want to have many products. Speaking from experience. If you want to do that, avoid DPM and go with Other Cube.
In Other Cube, Customer and Product can each be a dimension. That way, you can get profitability by customer, and by product.
The next pitfall is level of detail. Planning by with hundreds and thousands of customers/products is going to be challenging. You may want to do some aggregation (Small customer, medium customers, large customers). You also want to avoid replicating the data granularity from the host system. At a Prophix conference several years ago, they had Revlon speaking about how they use Prophix with the thousands of SKUs they have. They leave the detail in the host systems, and bring in summarized actuals into Prophix for planning purposes.
Watch out for percentage ratios with leaf level calculations. Use non-leaf calculations.
If recordings are still available from the Virtual Prophix User Conference, I touch on several of these topics in my presentation.
Finally, don’t be afraid to fail. That’s a good thing with Prophix. You can try one way, find it doesn’t work, and do it another way. Again, speaking from experience.
If you’re on CSP, chat with your account manager to walk through your plans.
1 Like
We have not used this in our organization.
Interested to see what other clients use on this subject
Thanks! - that’s a solid place to start. My current plan is to use non-leaf level products (which should decrease the product list considerably), but all customers. Because this is for reporting and not planning, I do have the upside of not getting bogged down on the sales allocation piece of the puzzle (I’ve been burned there before too!)
I couldn’t agree more with Paul about the Customer / Product depth of dimension.
Our original cube has over 7000 leaf elements - which is far too onerous to plan with. Keep that in mind when you’re designing your cube.
Our same 7000 leaves will be promoted to just 118 branches in the cubes next iteration.