Skip to Content

Archived discussions are read-only. Learn more about SAP Q&A

Steps to upload new countries data on top of existing data

Hi experts,

In our demo environment, we have already loaded US data for all( master , transaction and market research data) in DSiM system and we have configured all the settings for US in SPRO. Now we are planning to load UK data for our POC. So the locations details for stores and currency will be varying from US data. We need to know like what are the configurations we need to do for loading UK data( such as master- product and location and transaction- sales and stock and market research data)? do we need to create new contexts ,data delivery agreement and process to load new countries data?

Kindly advice. Thanks in advance.

Regards,

Saranya S.

Former Member
replied

Hi Saranya,

thanks for your question!

In most cases, it makes sense to keep the different data sets (per provider, origin, DDA, country, etc) as separated as possible to avoid future dependencies e.g. if customizing needs to be changed for one country or provider but not for the other. Therefore, we mostly advise to create new entries.

Process could stay the same.... it is easy to later on create a new process for a specific data set (e.g. one with enrichment of POS and or w/o Enrichment of POS) if you are in need of that and assign to the different DDAs respectively.

Also, the objects have dependencies that one needs to be aware of in order to make a proper decision that meets also the future needs. I am going a little bit into detail as this might be interesting also for other users in future:

  • Data Delivery Agreement (DDA): This is a very central object. So here we really advise to create a new DDA for each set of master and / or transactional data periodically (e.g. daily or weekly) provided by data origin, in order to keep the loading of the various data setrs as separated as possible from each other.
    It is linked to the following objects - Data Provider, Data Origin, Logical System and Context. Several DDAs may refer to the same context. This allows you to reuse the mapping of external identifiers to internal identifiers in case several DDAs are using the same external identifiers for the same object.
    Following settings can be done DDA specific: Quality Validation can be set up per DDA, Folder Scanner, File Set (e.g. name pattern), Harmonization Group...

  • Data Origin: This is used to identify the parties that created the data (e.g. the Retailer). It can be used to limit the access to the data for a specific retailer to a selected group of people. So if that is required by the customer during implementation, we advise to create new ones also per country even though it is the same retailer.
    The Data Origin is linked to the DDA, the Data Harmonization, Loction Calender (relevant for Enrichment), Period of Sales Deviation (relevant for Enrichment) and User Authorization. Be aware that the complete Data Harmonization can be set up per Data Origin!

  • Data Provider: It describes the name of the company that provides the demand data (e.g. a third party data provider in case the data is not directly received from the retailer). This is done so that the quality checks can be done per Provider. So in case you have a provider with really good quality, the checks may vary from the ones that have bad quality.
    The Data Provider is linked to the DDA and the User Authorization (the data access can also be managed via the data provider).

  • Context: The context compounds external master data IDs to make them unique. It is linked to the Number Ranges (Location and Product), SPRO General Settings and Partition Values for Master Data. We recommend to use different (non-overlapping) number ranges per context!
    If a new context needs to be created or not needs to be decided in each and every case. Here are 2 examples:
    Example 1: 1 Context for multiple DDAs
    e.g. a Retailer XYZ has 1 central master data system and sends data for RetailerXYZ Canada and RetailerXYZ US. If DSiM receives POS data from both US and Canada, those splitted data sets are then coming from synchronized systems on RetailerXYZ side. So the Canadian Product 123 is the same as the US product 123. In this case, the same context could be used for both US and Canada from Retailer XYZ.
    Example 2: 1 Context per DDA,
    e.g. Retailer XYZ has no central master data system and US product 123 is not the same as the Canadian Product 123. In this case, we would recommend to create 2 contexts.

Those were the main objects. But be sure to also think about

  • Harmonization Group: Create a new one in every case where you think that you need a specific derivation instruction set in the AVD for single or groups of DDAs
  • Attribute Set: Create a new one to show different/additional/less attributes on your harmonization UI for e.g. different products.
  • Attribute Bundle: Create a new one if you want to group different attributes on the harmonization UI (mainly usability)
  • Attribute Group: Create a new one in case you have the requirement ot limit the visibility of sensitive information in harmonization to specific users.
  • Mapping Instruction Set:They control how attributes are imported, converted and exported before and after harmonization. Create a new one whenever the mapping of a new data origin differs from the ones already created.
  • Derivation Instruction Set: They control how attribute values are derived from selected attributes of source objects. Create a new one whenever you think that it is useful to automatically derive attribute values for a harmonized product / location.
  • Mapping Function Sequence: Defines which mapping functions are executed in which sequence. One sequence can work for multiple origins. Create a new one when you think the sequence should differ for a specific origin (e.g. if no GTIN is being provided, it makes sense to use other steps instead, or if the description is really good for one origin, but all other information lacks, then the description could be the first step for the mapping).
  • QV Mapping Sequence: The sequence is highly dependent on the type of data (POS or Market Research) and the quality expectations. we expect that the sequence stays the same in most cases, but the thresholds differ per DDA.

Hope this helps! Please let us know in case you have further questions.

Best Regards,

Anna

1 View this answer in context
Not what you were looking for? View more on this topic or Ask a question