Walkthrough Ardoq, a recent innovative Enterprise Architecture tool

William El Kaim
11 min readMay 26, 2021


Ardoq was founded in 2013 in Oslo, Norway. The company was built by Magnulf Pilskog (Co-Founder and Board Member) and Erik Bakstad (CEO and Co-Founder).

Ardoq was created to :

  • Offer ready to use and fully customizable information model (named “Metamodel” in the rest of this article)
  • Offer a simple way to import data through excel, that was greatly improved over time (more here)
  • Facilitate data crowdsourcing, though a great user experience, but also through surveys, intended to make data collection via excel obsolete
  • Provides off-the-shelf “views”, that can be organized in stories
  • Enable easy and powerful impact and dependency analysis, leveraging Ardoq internal graph database
  • Offers standard APIs to import and export data, and ready-to-use connectors

Let’s start … Ardoq landing page

When logging in Ardoq for the first time, this is what you will see.

ArdoQ landing page

This is pretty unusual for an enterprise architecture tool! In fact Ardoq is more a meta-tool, that will enable you to build “your” tool, based on your needs. It could be nevertheless be seen as an advantage, since it does not look like a standard enterprise architecture tool, and could be of interest to increase business users curiosity and then adoption.

Ardoq key concepts

  • Metamodels define the structure of workspaces, including their components, references and fields.
  • Workspace contains Components and References between components. A workspace contains a single metamodel, however, you are able to link dependencies across workspaces.
  • A component represents any business object or entity documented in Ardoq. The component type is dependent on the metamodel being used.
  • References are the glue that connects and links components together.
  • Fields are permanent attributes that can be set per component and reference type(s) or globally to all component and reference types. Fields are also searchable and filterable.
  • Tags can be used to give your information additional analytical dimensions, so you can filter information based on those tags. Tags are great ways to quickly mark references or components. #winning
  • Survey: To collect data outside of the user base, you can create Surveys that people with relevant information can fill in as easy as 1–2–3.


Essentially, a workspace is a logical construct or collection that allows you to group all your Components into one place within metamodels

You administer security and permissions based on who is allowed to edit and work with your data in a workspace. It might be that you would like to group all your components by project, by department, or company wide. That’s up to you.

A new Workspace is made by clicking the ‘Create new’ button in your home screen and selecting ‘Workspace’.

Create an Ardoq Workspace

This takes you to the workspace templates available. If you do not know what a metamodel is about, by just positioning your mouse cursor on top of it, you can see the components of the data metamodel involved. For example for the Hosting and infrastructure, you see four components: cloud, hosted, on-premise, server and Network.

It is to be noted that Ardoq supports natively ArchiMate …

You can either choose to use one of the metamodel available in Ardoq or build your own. If you want to build your own metamodel, the ‘Blank Workspace’ is available under the ‘Business’ tab.

Let’s create a Data Mgt. workspace

Imagine that I want to create a Data Mgt. dedicated workspace where I can define my application data in a hierarchical way. For example a “consumer”, is described by its personal information, social profile and customer segment.

You have to create a workspace that uses a blank metamodel, and start structuring your information. There is a one-to-one relationship between metamodels and workspaces. Every metamodel is associated with one workspace and vice versa.

You can think of a metamodel as a database schema that defines the hierarchy of pages and valid relationships between the pages inside a workspace. The metamodel allows Ardoq to automatically create visualizations, as well as validate input. As an example, I will create a metamodel for Application Data management. I selected the blank metamodel and added the name, while deciding to stick on ArchiMate standard for the shape.

A data object is composed of a name and a category

The workspace is now called Data Mgt., and the first available view when accessing it will be a capability map

Once created I can immediately start entering data through the integrated Grid Editor

So I started to describe a consumer, which is described by its personal information, social profile and customer segment. As you can see, hierarchy between objects are natively offered.

You can also select the existing Ardoq provided views you want to associate to your workspace.

Not let’s see the data we just filled using some of the native Ardoq visualization modes (component tree, followed by Block Diagram).


My workspace is now ready for Data Mgt.

I can also add attributes to the created Data Objects. I wan to create a data domain map, with “data domains” characterized at a lower level by “data families.” In the workspace menu, I will add a managed field type.

Then enter all required information, and apply it to the metamodel

I can now add the right classification to my data objects

Ardoq Metamodel

The metamodel can be created by clicking on the metamodel sign, which can be located at the left-hand sidebar. By choosing to include the workspaces located in your Ardoq organizations, it is possible to display the metamodel of your entire organization. You can also create metamodels out of a given list of workspaces or folder:

This is a good way of limiting what to show in the metamodel, but be aware that references that point to components outside of the selected workspaces will not be shown.

Let’s now create a new Metamodel with all the pre-existing Ardoq workspace, plus the one we just created

The metamodel generated is shown below. You can all components of our metamodel described, including their references.

When clicking on a component- or reference type, you can see a list of fields that are added to the specific component or reference type you selected and the number of entities that are missing a value for the selected field. Let’s see what is available for Data.

One great functionality is to save your personal metamodels as a template, to be re-used and standardized across your organization. The new template will then appear in the same create workspace overview.

Creating references

As you can see,. Data object is still not connected to any other objects. Let’s connect my “customer” Data and Applications (for example Ariba and “customer payment portal”).

Then, once associated, you must give some information to characterize the reference. The type field enable to use pre-existing references and icon (here a simple arrow) created in the workspace.

The result is presented below.

To know more about references, read here. Let’s now regenerate the metamodel, and we can see now, that references were taken into account.

Scenario — from current (AS-IS) to Target (TO-BE) architecture

Let’s now deep dive to a new feature releases recently by Ardoq, Scenario Mgt. You must note that is a paid add-on.

Scenarios in Ardoq lets you select a subset of your components from across your workspaces, where you can make changes in isolation from the mainline. You can then compare, share, and present the suggested changes to your team and other stakeholders to get buy-in and alignment, and to find the best way forward.

Ardoq Scenarios is like a versioning system, enabling to create branch of possibilities and merge

Let’s imagine that we want to add a data lake to store our customers on AWS S3. I created an application Data Lake, with a reference towards AWS S3

You can either create a scenario from the components you have selected in the navigator, or from what you see in the view (Block Diagram is the only view supported currently).

Once created, we have access to a new branch.

My scenario to build a data lake for “customer”, does impact only two applications: Ariba and Customer Payment portal.

So this is my target vision, Ariba and Customer Payment portal will access directly the customer data in the data lake.

Ardoq visual Diff feature enables me to see what as changed. Below, I’m showing the “mainline”. Selecting Diff will enable you to see all changes, then “build a data lake” will show the results. Pretty amazing, isn’t it?

In the mainline window I can see mainline (or as-is) assets impacted:

The merge workflow consists of two main sections: metamodel and data, with sub-sections for each type of change:

  • Metamodel merge: The metamodel merge section handles differences in the workspace metamodel structure (rigid vs. flexible metamodels), component types, reference types, and field definitions. Read more about the details here.
  • Data merge: The data merge section handles differences in your components, references, and tags. Read more about this part of the merge workflow here.
  • Progress bar: On the left side, there is a progress bar that highlights the steps with issues to address. The purple step is the one you’re currently at, steps with filled circles have nothing to address, and steps with empty circles need attention.

Only steps that require your attention will be triggered. This should normally only be a few steps, and it depends on how many changes have happened in either the scenario or in the mainline since the last merge.

To know more about scenario, read here or watch the short video Managing transformation at scale with Ardoq’s Scenarios by Ian Stendera

Ardoq Advanced features

Ardoq is offering some other interesting features:

  • When you are putting data into Ardoq you are essentially building up an Enterprise Intelligence Graph. This provides an always up-to-date Digital Twin of all data, processes, and their relationships, meaning you can ask the system about all parts of your documentation. Gremlin is the graph traversal language that Ardoq’s TinkerPop-enabled Enterprise Intelligence graph system is built around. Gremlin is a functional, data-flow language that enables users to express complex graph traversals with relatively little code (more here). You must note that Graph Search is part of the Analytics and Reporting module which is a paid add-on.
  • Calculated fields in Ardoq make it possible to collect and process information across everything you have stored in Ardoq, and store the result as a field value on components and references. The calculated value is read-only, but apart from that, it behaves like any other Ardoq field. Calculated field values can be used for filtering, grouping, and conditional formatting as you can with regular ArdoQ field values (here for more). Calculated fields offers the capability to traverse dependencies and to create calculated attributes that could reflect the metamodel complexity or facilitate risk and governance analysis. For example, you can calculate residual risk by automatically identifying connected and aggregate risks and evaluate the impact of the mitigations implemented. Or to roll-up cost to an abstract level based on deep underlying and evolving dependencies (capabilities down to infrastructure for example).

Calculated fields are one of the most useful feature of Ardoq enabling you to start creating analysis and insights from the structure of the Ardoq graph.

  • Ardoq Presentations Mode allows you to present live data from Ardoq in a simple way, and also to embed your interactive presentations outside of the tool. Think Powerpoint but with live data. As you can see at the bottom of the screenshot below, each view is presented as a bullet on a timeline.

To open an existing Presentation, just click on the Presentation button in the upper right corner:

To add the current view to a presentation, just click the “Add View as Slide to Presentation”

Best Practice Module

Ardoq’s Best Practice Modules are meant to give you everything you need to get started with a specific use case. These best practices are designed to build off of one another so that you can grow your scope as your adoption increases.

Best Practice Modules include pre-configured:

  • Metamodels
  • Surveys
  • Dashboards
  • Calculated Fields
  • Presentations
  • Excel Import Configurations
  • Reports

Today 3 Best practices modules, all created by Ardoq are available:

  • Business Capability mapping
  • Application portfolio Mgt.
  • Strategic planning and execution

To know more


I like to have free access to the tool I explore. So I really want to thank Ian Stendera (VP of product) and Gard Havelin (Head of Partnerships) for our open discussions, and the full support provided to access the tool and especially to the scenario module (needed for As-IS and To-Be analysis).