“Architecting the Obvious or Adding Value”?
A re post from our sister site www.businessarchitecturetraining.co.uk published in February 2018.
Earlier today I read with interest a post on LinkedIn promoting a standard capability framework for the financial services industry. It comes from IRIS business architect, a company that has its thinking very much in line with the Business Architecture Guild. The capability framework it promotes looks to be seemingly a comprehensive well-thought through output. Job well done no doubt.
I have seen this sort of thing before, but mostly in process model terms from people like APQC and Leading Practice.
At first glance a standard model seems sensible; why invent your own when someone else has done the graft?
On reflection I started to question the value of this approach based on the underlying message that it was giving out; that message being “A sector has a standard model”. A standard model of capabilities therefore means standard outcomes
If you are all the same then why do you create a capability model in the first place?
In essence this type of work it is to identify how your value chains add value to the customer linking capabilities, via value stages back to customer value delivery. This informs where to invest and concentrate ones effort in what adds value to the customer – highly sensible.
However if the sector has the same model in what it delivers where is the differentiation? Conforming to the same model forces you into commoditisation and “rail roads” you into economies of scale and cost leadership. It all becomes price and cost based.
If the role of capabilities is to provide a central “Rosetta Stone” or lingua franca to pull together the business architecture then I suspect there is some value had in doing that and a pre-defined standard framework saves a fair bit of work but that as a primary purpose seems quite limiting.
The standard is maybe forcing your thinking into a preconceived mould, inhibiting innovation and creativity. Even if you do it yourself and come out with something similar to the standard model the process you went through to get there is telling you something – that “differentiation is difficult around here” – but having gone through the process you can at least say that “you explored the possibilities”. It has validity because you went through the pain. In many cases the process of doing the capability modelling delivers more value in terms of: learning, reflection and analysis that the final model or product. Just using a standard loses all that experiential learning benefit and I worry that it just ends up with lazy analysis and lazy architecture.
One of the issues justifying a standard model may well be that being “different” in certain sectors, like financial services, is a challenge due to the constraining regulation and external forces. The regulation potentially dictates the operating model to an extent where the whole sector becomes homogenous. Is this the case here and is this why we can have a standard financial services model?
I have argued for some time that many larger financial businesses differentiate themselves only by their brand, reputation and size. One only has to read the strategies of these organisations in their annual reports to realise there is a “cookie cutter” approach going on here. This is often sustained by internal myths and beliefs of being different when in reality each is as boring as the next. It is a bit like convergent evolution in that the same environment causes species to evolve from different physiological origins towards a common form. By accepting that this standard model may represent the reality, in that differentiation is just not to be had, then why go about this type of value analysis at all?
In accepting this “template” possibility, there would be value in having this framework as a benchmark for some kind of maturity pegging. You could measure the effectiveness in terms of efficiency and quality. This as a baselining exercise provides a reference-point for improvement with measurement against those identified Key Performance indicators. Just focus on doing the same as everyone else but with less cost. Cut some costs, do some lean process improvement or zero based costing exercises and “heh presto” the new operating model appears. So in this case I can see benefit of a standard model but really a standard process model would do the job just as well as outcomes are pretty much homogeneous so why bother. This leads to a debate that high level process models look pretty much like capability models with the verb and noun switched round – especially in services – a topic for another day perhaps?
These standard models perhaps provide a check list, but apart from “architecting the obvious” I am not convinced there is enormous value to be had here. Some value, yes in the right circumstances, but if any model is too generic, and applies to anything or everything, then its value is not that great.
You can make the same types of comments for “Value chains”. There are plenty of cases of high level encompassing value chains that are applied the same wherever you look – what is necessary is for them to be: specific, or low enough in focus to a lower enough resolution to identify differences and opportunities.
I am sure these frameworks are not cheap either, but that is fair enough, you have to pay for the time someone spent doing your work for you – right? We must compliment the vendor for the effort applied – and yes this is a good piece of work.
From the perspective of a business architect practitioner if the business can adopt these standard frameworks as their business architecture then do they really need a business architect or the discipline at all? This could be a bit of an “own goal” for the profession; or perhaps it just means in certain sectors where it’s all standard and generic the value that business architecture can add is limited.
I suspect that actually this isn’t the case and it is the striving towards templates and standards that is obscuring the potential value – but you would expect me to say that – wouldn’t you?