Analysts, EAs and our users see the fidelity of the Essential Meta Model as a huge benefit, however, for those new to it the scale can be quite daunting at first, and we often see some initial concern that it is too complex. The reality is quite different, and it does not take long for its broad and unrivalled fidelity, built up over 20 years of Enterprise Architecture consulting and addressing client problems, to be appreciated.
Key to this is that the Essential meta model reflects the real world, not abstract concepts, which means that business users can quickly understand it. As an example, one of the best jobs of the initial population of Essential was completed by a user with no Enterprise Architecture experience at all – a huge benefit when looking for collaboration across the organisation. That being said, organisations are complex with lots of interconnected parts that we need to understand to do a proper job as EAs. The Essential meta model reflects this complexity where it is needed, but there are no concepts created purely for the sake of it.
We often see users moving to Essential from other tools with a more simplistic meta model once they have got passed the initial capture, as they find that this simplified approach cannot support their decision making – rather it is ok as a high-level pointer, but the fidelity is not there to enable proper enterprise architecture decision making and support.
An example of this is the request to map applications directly to business capabilities, something that pretty much every other EA tool does. This very quickly gives some nice, high-level dashboards showing, for example, how many applications support a business capability, the overall cost of that support and a view of the fit of the applications to the business needs. These high-level views are, in our opinion, interesting, but not very useful for analysis. To take the knowledge forward and make decisions requires further detail that is not available in the tool, requiring users to go away and do further analysis– taking resource time, effort and money.
A good way to understand this is to anticipate the follow-on questions you will get asked once people see the high-level mapping, e.g.
- Which applications can we actually rationalise?
- Where is the highest cost saving?
- Why is that application fit showing as poor against that capability?
These questions require a much deeper analysis.
As an example, the initial high-level view might show Oracle HR, Workday and several bespoke applications against a Payroll Management business capability. Management will say ‘that’s a lot of applications, can we get rid of some and reduce cost?’. At this stage all we can say is, ‘Possibly, we need to gather some more information’. The question cannot be answered because the EA using this tool does not actually know what the applications are used for.
In the Essential Meta Model, we map applications to business capabilities via the processes that use them and, importantly, the application service, or business functionality, that the application provides to it. This allows us to understand that, for example, Workday does provide payroll management functionality, but the Singapore payroll process has a specific legal requirement that it doesn’t currently support, and which is provided by the local payroll solution. Having the applications mapped to processes isn’t enough, and the application services are critical to giving us the fidelity we need to make an informed decision. So, when Management asks, ‘Can we reduce the number of applications and save cost?’ we can answer fully, ‘We can make Workday our standard application and remove a number of other applications such as Oracle HR, but we will need to retain the local Singapore solution until Workday can meet their legal requirement’. Mapping applications directly to capability, or even process, without understanding the services, or functionality, they provide, would not give this level of detail.
Crucially, you can still see the same high-level dashboards that Senior Management like, but the answer to detailed questions is immediately available and, ideally, you would have presented the option rather than waiting for the question.
We are, of course, very aware that not all organisations have all this data up-front, and it is possible to model a simple relationship, using placeholders, that allow you to complete your initial capture very quickly but then expand the level of detail as and when it is required. This allows you to take an incremental approach to data capture without losing the benefits of the fidelity in our meta model.
This is just one example of how the Essential meta model provides huge benefits, but there are many, many more.