# Business Key treatments: what is allowed and what isn't?

**URL:** <https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379>\
**Category:** Data Vault 2.0\
**Created:** [August 2, 2022, 3:34pm UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379 "2022-08-02T15:34:16Z")\
**Posts on this page:** 12\
**Page:** 1

<div class="post-metadata">

**Author:** ![FrenkLoonen](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/frenkloonen/32/119_2.png) [@FrenkLoonen](https://forum.data-community.org/u/FrenkLoonen)\
**Post date:** [August 2, 2022, 3:34pm UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/1 "2022-08-02T15:34:16Z")

</div>

I have several sources that all deliver references to a local client. There’s even one source that is now considered the golden source wrt client data, in which they’ve brought all client data from all local systems together, and tried to link those to the ‘global’ client. Clients can be identified with the composite business key “local\_id” + “local\_schema”. They even applied some BK treatments already in the sense that they do left padding zeros so that every local\_id is a 12-digit number. I want to perform the same treatment on the fields from other sources.

The only thing is, let’s say I now have these two sources:

golden\_source:  
local\_id | local\_schema  
000000000123 | ABC

other\_source:  
local\_id | local\_schema  
000000000123 | Alpha Beta Charlie

Alpha Beta Charlie means exactly the same as ABC in the golden source. Is it okay to perform a treatment (which is going to end up a CASE statement or maybe a lookup to an external view)? Or would I have to load it as-is and create a same-as link in the BDV?

---

<div class="post-metadata">

**Author:** ![AHenning](https://avatars.discourse-cdn.com/v4/letter/a/96bed5/32.png) [@AHenning](https://forum.data-community.org/u/AHenning)\
**Post date:** [August 2, 2022, 6:59pm UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/2 "2022-08-02T18:59:36Z")

</div>

Both the CASE-ststement and the lookup are two ways to perform soft business rules.  
If you perform a lookup you also create a dependency to another source, this means that the process for loading the dataset stops being autonomous.  
I would suggest that you dont do any of them.

---

<div class="post-metadata">

**Author:** ![jhall\_uk](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/jhall_uk/32/79_2.png) [@jhall\_uk](https://forum.data-community.org/u/jhall_uk)\
**Post date:** [August 4, 2022, 10:07am UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/3 "2022-08-04T10:07:37Z")

</div>

Hi Frenk,  
This looks like a potential use case for a Same As Link to me - you can then load each occurrence and map to your golden business key.

Cheers,  
John

---

<div class="post-metadata">

**Author:** ![patrickcuba](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/patrickcuba/32/19_2.png) [@patrickcuba](https://forum.data-community.org/u/patrickcuba)\
**Post date:** [August 7, 2022, 7:36am UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/4 "2022-08-07T07:36:20Z")

</div>

Same-as link and the link should be provided by the source, do not manufacture this relationship yourself or you will end up managing this technical debt

---

<div class="post-metadata">

**Author:** ![FrenkLoonen](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/frenkloonen/32/119_2.png) [@FrenkLoonen](https://forum.data-community.org/u/FrenkLoonen)\
**Post date:** [August 8, 2022, 8:00am UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/5 "2022-08-08T08:00:05Z")

</div>

Thanks for everyone’s input! I will go with the same-as link.

---

<div class="post-metadata">

**Author:** ![FrenkLoonen](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/frenkloonen/32/119_2.png) [@FrenkLoonen](https://forum.data-community.org/u/FrenkLoonen)\
**Post date:** [August 11, 2022, 2:05pm UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/6 "2022-08-11T14:05:44Z")

</div>

But what now when I get the following source:  
ABC\_local\_id  
000000000123

So here, it is implied by the name of the field that the local\_id belongs to the local\_schema ABC. What would my mapping be in this case?

---

<div class="post-metadata">

**Author:** ![patrickcuba](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/patrickcuba/32/19_2.png) [@patrickcuba](https://forum.data-community.org/u/patrickcuba)\
**Post date:** [August 11, 2022, 11:09pm UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/7 "2022-08-11T23:09:10Z")

</div>

i don’t understand… that would map to your hub table, no?

---

<div class="post-metadata">

**Author:** ![FrenkLoonen](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/frenkloonen/32/119_2.png) [@FrenkLoonen](https://forum.data-community.org/u/FrenkLoonen)\
**Post date:** [August 12, 2022, 7:43am UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/8 "2022-08-12T07:43:50Z")

</div>

Yes it would, but my Hub contains a composite business key (local\_id + local\_schema\_id). This new source only provides the local\_id, the local\_schema\_id is implied by the name of the source column.

---

<div class="post-metadata">

**Author:** ![AHenning](https://avatars.discourse-cdn.com/v4/letter/a/96bed5/32.png) [@AHenning](https://forum.data-community.org/u/AHenning)\
**Post date:** [August 12, 2022, 2:42pm UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/9 "2022-08-12T14:42:14Z")

</div>

Hello Frenk!  
My suggestion(might not be data vault standard) is to never put composite business keys in a hub. It makes implementation and automation a lot easier. If you have to combine two or more keys, do it in a link. I also revommend using multiple hubs instead of bkcc.  
The models wont be as pretty but it is so much easier to handle when you extract data from the RV.

---

<div class="post-metadata">

**Author:** ![FrenkLoonen](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/frenkloonen/32/119_2.png) [@FrenkLoonen](https://forum.data-community.org/u/FrenkLoonen)\
**Post date:** [August 21, 2022, 10:25am UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/10 "2022-08-21T10:25:50Z")

</div>

But what if I get two local\_ids with value 123? They don’t mean the same and they can’t live on their own, they only mean something icw the local\_schema. So I would get two hubs that can’t/won’t have any satellites on them, only on the link I can create a satellite which holds their descriptive attributes… can’t say I’m too keen on that solution.

---

<div class="post-metadata">

**Author:** ![AHenning](https://avatars.discourse-cdn.com/v4/letter/a/96bed5/32.png) [@AHenning](https://forum.data-community.org/u/AHenning)\
**Post date:** [August 21, 2022, 11:12pm UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/11 "2022-08-21T23:12:56Z")

</div>

If local\_id from different sources means diffeent things, i suggest you put them in different hubs.

It is more important that the data can be regenerated (zero data loss) then to create a solution that is keen.

---

<div class="post-metadata">

**Author:** ![RhysHanscombe](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.data-community.org/rhyshanscombe/32/394_2.png) [@RhysHanscombe](https://forum.data-community.org/u/RhysHanscombe)\
**Post date:** [December 1, 2025, 2:41pm UTC](https://forum.data-community.org/t/business-key-treatments-what-is-allowed-and-what-isnt/379/12 "2025-12-01T14:41:36Z")

</div>

Closed on December 1, 2025, 2:41pm UTC


