Why?
People are at the center of any organization, and are often the center of the model. Who should I talk to about this Application? Who is impacted by these changes?
This data is rich context for personalization. As you map out your architecture, Ardoq will get better and better at using this information to drive personalized experiences -- especially with AI.
How?
User accounts and people data are often involved in two synchronization processes:
Integrations with 3rd party systems containing Person data (e.g. Microsoft Graph)
Integrations with SSO and/or SCIM for user account provisioning
Rather than adding even more complexity to this process, we take the simple option: explicitly binding a user account to a person component. This option can be found in the Users table, as the "Linked Component" column:
This linking is stable, so it will remain unchanged if emails change.
To begin with, you can link each account individually:
However, this can be cumbersome.
We highly recommend enabling Auto-linking at the top of the user management page. This allows you to:
Select which workspace your people components live in
Select which type you use to represent a Person
Select which field you want to match email addresses against
By default, it will try and infer your settings and choose sane defaults:
Auto-linking not only ensures that your users are regularly linked correctly (by default it runs nightly), but also that your users are linked the first time their accounts are created. This means that they will benefit from personalization the first time they log in.
What?
So what does this change? How is it used by the platform?
Currently Ardoq uses this in 2 different ways, but we will be continuing to release many features that leverage linked user accounts.
My Applications
Business users are key personas where personalization brings value, and we've started with application owners. Linking a user account tells the platform: "This user is an application owner, and these are the applications they own". You can then leverage this on the Discover Homepage, with the My Applications widget:
The My Applications widget gives your application owners fast access to their data. It will only show if they are an application owner. Currently this is based on our Foundation metamodel of Person -> Owns -> Application. However please see our FAQ questions related to customization.
Important: This widget will not show by default. You will need to enable it in the Discover settings.
Personalized Images
Person components often have images attached to them to aid in visual presentation. In such cases, we use them to enrich the UI with profile photos. There are many places where this will appear, and it bring a bit of richness to the UI:
FAQ
What determines "My" Applications?
We currently support the metamodel from our Solutions, of Person -> Owns -> Application:
However, we realize that there are many possible extensions to this:
Experts
Initiatives & processes
Capability ownership
etc
In addition, you may want slightly more complex definitions which may span multiple hops.
These are all actively being considered, so stay tuned!
One of my users can't see the My Applications widget. Why not?
It's likely one of two reasons:
You're using a different metamodel than Person -> Owns -> Application.
The user isn't an Application owner.
The user doesn't have read access to the Person workspace or the Applications workspace. This widget will not bypass read permissions, and is purely a personalization layer. In order for them to see the widget, they need read access on the Application and Person workspace(s).
Can we rearrange the ordering of the My Applications widget?
Not currently. If the user is an application owner, and they have applications, My Applications will always be shown above Browse Data.
We are considering ways to support rearranging and customizing the Discover homepage, so we don't see the current behaviour as a longterm solution.
Auto-linking is not linking some of my users. Why not?
We try to disambiguate as much as we can. However there are a few cases which we are unable to link:
No matching component: no component of the configured type, in the configured workspace, has that email in the configured field.
You may have two components with the same field value (e.g. both sharing the same
contact_emailvalue). In these cases, we decide not to link. The best course of action is to correct your data using Inventory to remove the duplicate emails.Autolinking is not turned on.
Some of my users seem to have changed their linkages. Why?
The interaction between auto-linking and manual linking can be a bit tricky. Auto-linking always takes priority when a user's email uniquely matches a person component. A manual link is kept only while it doesn't conflict with a match. Changes are applied on the next auto-linking run (nightly, or when the configuration is saved or auto-linking is turned on).
Re-linked: if your email matches a different component, your manual link is replaced with a link to that component. From then on it counts as auto-linked
Unlinked:
Another user's email matches the component you were manually linked to. Only one user can be linked to each component, and the auto-linking match takes priority, so that component moves to them. You're left without a link unless your own email matches another component.
The link was auto-linked and your email no longer matches exactly one component, either because it matches none or because it matches several. The link will be removed.
This is why we generally recommend all links to be maintained through auto-linking, and simply update your data to ensure the links are created appropriately.