In this article, we’ll go more in-depth into the powerful and unique way Ardoq allows you to build your audience options in Broadcasts. Read the Broadcasts introduction article to learn about why and how you can engage beyond your IT teams with personalized messaging that’s automated and scalable.
Build Your Stakeholder Map Right Into Your Enterprise Architecture Models
Automating communications isn’t new. Where existing solutions can fall down is their ability to match the right message to the right recipient. When your stakeholders are themselves a long list of different domain specialists, just spamming everyone with the same message may not be the best idea.
This is where EAs need intelligence in how to target their audience.
Email groups, group mailboxes, channels, and forums are all ways of getting the word out to more targeted audiences. Even these methods may take limited account of individuals’ roles or interests.
For really impactful personalization, we can exploit something Enterprise Architects excel at - Building models. EAs are well used to building stakeholder maps and RACI matrices to track their complex collaborations.
They’re also experts in building process maps, applications landscapes, business capability models, and more to map their organization’s systems and processes.
But even though the people in those maps are often directly related to the components in those models, that relationship is seldom explicitly modeled.
Broadcasts is designed to leverage 'People-in-the-graph' modeling where your stakeholders are modeled as components in themselves and linked to the other elements in your Enterprise Architecture models. People become 'first-class citizens' in your architecture.
Doing this enables EA teams to automatically analyze each individual persons' role and interests and use the structure of their EA models themselves to build rules that route the right message to the right person. True personalization of messaging at scale.
Although it enables powerful audience targeting, you don't have to use 'People-in-the-graph' modeling to use the module. Broadcasts offer four different options for building your audience, from simple to sophisticated.
Depending on the task you need to perform, you can pick from the most appropriate option - or combine them.
Level 1: Audience targeting by group email
Level 1 is the simplest and uses a Group or Individual Email to send a Broadcast. We're basically sending a mass-mail here - and like any mass-mailing, it risks being dismissed as 'spam'.
There's still a good case for this type of audience targeting. You may want to invite responses from the maximum number of your coworkers. For example, you're running an innovation program and want to invite the whole organization to contribute their ideas.
It's also useful when you want to match applications, projects, processes, etc., with people in the first place. For example, you can send out a survey with your full application list asking the community to contribute the applications' owners or subject matter experts' names.
Level 2: Audience targeting using People Workspace
Level 2 is the ability to send a message to one or more specific people who have been populated in an Ardoq People Workspace.
The workspace gives you a current and searchable list of all your people from which you can pick those you want to send a message.
The value-add here is that Ardoq can sync that list with your master source (e.g., Azure Active Directory) and automatically add new people as they arrive, and flag when they've been removed from the source, so you always have a current list to work off.
Currently, we only have support for people components of type Person.
Level 3: Audience targeting based on direct responsibility or interest
Level 3 is where we start to make good use of our Ardoq data to intelligently match the right messages to the right people.
Using the Predefined Graph Query option, you can use rules to pick your audience according to their role with or relationship to the components your Broadcasts is about.
For example, you can target the process owners or technology experts directly related to the processes or technologies you want to collect data or notify somebody on.
Whether sending out surveys or an alert, each person will only be messaged about his processes or her technologies, creating a personalized view.
Should that person change roles or leave the organization and hand off their duties or expertise to another, the Broadcast will still work. Because it uses rules based on roles or responsibilities and not named individuals to direct the message, the Broadcast will continue to function even if the end recipient changes.
Level 4: Audience targeting based on indirect responsibility or interest
Level 4 - building an audience using a Gremlin Query - is the most advanced type of audience targeting.
Here we can target our audience based on their indirect relationships, something only possible when using a graph database like Ardoq.
In organizations, most people already know what's happening in their own domain. They can still struggle to discover what's happening around them in the wider ecosystem.
For example, a process owner knows what's happening with her own process, but doesn't see what's happening in upstream processes that may impact her. Or a cloud architect knows all about the cloud services his organization consumes, but struggles to get visibility of the applications that run on those services.
In both cases, they lack situational awareness.
Indirect audience targeting enables situational awareness by creating a personalized comms ecosystem for each person in the organization.
By notifying people about events that are happening next door, upstream or downstream, you can overcome the organizational barriers that prevent effective collaboration and start to lift the 'fog of war.'
Here are a couple of examples:
You can select your audience based on a path - for example, downstream process or data flow stakeholders. So if you want to alert people about a problem with an upstream process or send out a survey to understand how customer data flows through downstream applications, a graph query can send an alert or a survey request to each person along that path - but only them.
You can also select your audience based on groupings of things.
Let’s say you want to build a list of all your applications by sending out an application survey, but you also want that list to be validated by domain experts to make sure it’s accurate.
Instead of allocating an expert to each new application submitted, you can assign experts to other component types like business or technology capabilities - for example, Customer Payments or Security.
When a new application is linked to one of those capabilities, that expert can then be notified and review the new application because those applications are grouped based on his or her domain. All the applications linked to those capabilities can be routed to those experts.
In the example above, you can see there are sometimes multiple paths to the same person. For example, Ariba realizes five Business Capabilities which Bertram Gilfoyle is an expert in. Does that mean Bertram is going to receive five separate emails?
No, the Audience Builder will de-duplicate the end-points of the path before it sends the Broadcast. You can see this from looking at the Audience View tab.
You also can use multiple groupings on the same set of components. Those same applications can also be routed for review to a cloud expert based on their hosting, and to a business product owner based on the lines of business they support.
In each case, the audiences are only indirectly related to the component in question, but the EA model acts as a set of message routing rules that combine to create highly-relevant personalized views.