I began writing a quick response to Erik’s question in this blog about Architect’s high-order bit is Adoption. Erik asked the question how to get people to adopt ideas and before I knew it I was over a page in my response because it was a lot of fun to answer it so I decided to make the response an individual blog post.
@Erik, fun question and thanks for asking.
Maybe I should respond using the typcial 'carrot' versus the 'stick' approaches. The carrot being an enagement between an Architect and the team that has obvious value to those s/he influences. The 'stick' approach resembles a policy-enforcer.
I've been down both paths and currently think that the carrot is the path of least resistance resulting in greater adoption. So, instead of focusing on the 'stick' approaches, let me talk to you about ideas how to acheive results using the carrot approach.
1. I'm not saying that the stick approach cannot be successful. I think that the stick approach requires a lot of organizational and cultural characteristics that finding such an organization with these prerequisites is very difficult.
2. I’m also not assuming it’s an all-or-nothing situation. Clever Architects appropriately use both approaches. I will say that they often lead with the carrot approach.
Ok, now for some tips to help you gain adoption of your idea in no particular order:
1. Gain trusted advisor with the business(es) that drive the project. If you are able to build strong relationships with the business leads that control the description of the business problem, you will be able to:
a. Help them articulate the problem in such a way that
i. It desciribes the area of the solution architecture you are most interested in influencing. There are always times when the business wish to achieve a goal but focus more on areas of the business problem that meet their concerns and detail these needs. What can happen is that the business gloss over areas that are not focus areas of the business problem, leaving a bit of ambiguity or wiggle-room for interpretation that may result in a poor-quality design in the solution.
ii. May implictly suggest how to solve the business problem that naturally includes a particular idea you have about the Solution Architecture. If you are able to have the business clearly describe Business Requirements that layout the need for a high-quality solution design idea you have, you will have the ability to ensure that the solution delivery team explicitly meet these Business Requirements by adopting your design idea.
b. Ask for their support in driving architectural change in the direction you wish. If you find yourself in the position the design team have built a poor design to solve a loosely described Business Requirement, you must be able to articulate the business ramifications of the poorly designed solution in terms of financial risk, Customer experience, Supplier/Partner experience, etc to raise awareness and appreciation of the design. Then, naturally suggest architecture options you have to help solve the problem and work with the business to help drive them through the project team. This can get quite complicated because of the dynamics of the project team. Normally by this time in a project’s lifecycle much of the scope is locked, estimates are done, resources committed, etc so getting these types of changes pushed through will require a very good explaination as well as a good bit of politicking to make sure that the business leads and project leads support your idea.
2. Contribute to the design of a project’s lifecycle process. I don’t necessarily mean that you are a part of the lifecycle process therefore someone who requires sign-off before a team can continue. Being a Sign-Off Owner is always fraught with pain and I don’t suggest this approach - it normally falls in the ‘stick’ approach and I prefer to stay away from this. What I am saying is to help the team build a project lifecycle process derived from your favourite methodology/technique/framework like Microsoft Solutions Framework (my personal favourite). Ensuring the team adopts the process that enables them to function together in a complimentary fashion is huge value to you. In fact, I believe that getting the team to function together is onf of an Architect’s prime directives. Help the team to be able to easily declare when an individual is out-of-role or is not performing their function. When you have an idea that you wish to be adopted, you can rely on the project's process model and team model to gain adoption. If a team isn’t functioning correctly, any silverback/alpha-dog personality can bypass any process and ignore you and your idea…or anyone elses for that matter. Very Bad.
3. Steward the models. Let’s face it, he who maintains the minutes writes the facts. All projects will require some sort of documentation. The key is to offer up skills and resources to help document it and, coincidentally all the way, you get the opportunity to suggest how documentation might be done to a) have consistency in the documentation and b) communicate it. Offer to take the burden of managing the model repository and be the modeler for the documentation. This is a really powerful technique to gain adoption of your ideas because as model-monkey, you:
a. Are invited to most meetings to document the discussion so you are well-informed and are present to help influence decisions.
b. When documenting via modeling, you gain integrity in the documentation via integrity of the models. This assumes you are using a proper modeling tool not what I call a ‘picture tool’. You get to spot poorly written requirements ad designs and articulate them via the model. This helps prove the value of the models and your role to steward them.
c. Caveats with this technque. If the modeling tool doesn’t allow for the models to be easily accessible and easy to use then you run a very big risk in the models not being used or referenced. When this happens, you also lose credibility and diminish your trusted advisor position. Be very careful when and how you apply modeling tools.
4. The Architects Middle-out Strategy. Ok, maybe mangling the term ‘middle-out’ too much here J. What I mean is to start with the Leads and work out from there. Avoid top-down and bottom-up, that is, Executive-down and Individual Contributer(IC)-up respectively. Gain trusted advisor for the managers with accountability that do the actual work in the area you wish to change and they will help you manage-up and manage-down. They are often the ones making the decisions that inform executives. They are also the ones who delgate the work to ICs. Partnering with these folks is critical.
5. Be selfless NOT selfish. We all have vision and ideas to be adopted. The key for an Architect is to find a way to make your vision and ideas resonate with whomever you want to influence. Don’t presume they care about your vision and ideas. Don’t presume they will explicitly understand them and the value they bring to the enterprise, customer, partner or shareholder. Seriously, this is super important. Always focus on helping them solve their problem. Then, along the way, interlace your ideas with the solution. This goes for efforts that are in planning, execution or maintaining situations. For example, even if your idea is as extreme as stopping an initiative, work with the planning team to understand their motives and goals and then cleverly help them discover your idea and collectively agree that it is in the best interest of the company to stop the initiative. Architecture is, and should be in my opinion, a thankless job. We are about doing the right thing which often means getting others to be successful and having their success recognized. Get used to it J. If, as an architect, you’re keen to be recognized or be ‘in the media’ as often as possible for driving change, you will dissapointed and often.
6. Understand and articulate your value proposition. Have a crystal clear value proposition to every project team lead and every senior manager of those team leads. I have a rule of thumb for describing value of a role. If you can’t explain the value proposition in 10 words or less and the meaning has obvious value to ALL of the project team leads, you are doomed. For example, here are some basic value propositions which resonate with all leads:
a. Project/Program Manager: Ensure the project is on-time and on-budget.
b. Testing Team Lead: Prove the solution works
c. Development Team Lead. Build the solution.
d. Release Manager. Deploy the solution.
e. User Experience. Ensure the solution is usable.
f. Architect: Ensure the quality of the solution to its stakeholders. This is what I often use, and is an extension of MSF’s definition. It is a bit squishy but good enough in my experience. When people question, what I mean by quality I answer based on their role. For example:
i. For Project/Program Managers, it is the traceability of the business problem to the solution to ensure no wasted efforts and to make sure we are solving the business problem and that the solution nicely fits within the enterprise environment in which it will be deployed.
ii. For Testing, it is the explicit focus on the solution architecture to optimize system quality attributes (Performance, Security, Flexibility, etc) that they must also test for in addition to the obvious Functional Testing.
iii. For Development, it is the explicit focus on the solution architect to optimize system quality attributes to meet the needs of the business owners and IT system admins as well as align to the long-term direction of system integration needs of the shared enterprise systems.
7. Daily Build. This tip applies the Daily Build principle from MSF. In this context, I mean publish models and diagrams every day allowing you to get your latest thinking out for others to use (i.e. adopt) and provide feedback. You will gain a number of advantages such as:
a. Heartbeat. Your architecture diagrams will show life through a regular publishing and increase your chances of being seen by other groups used improving expectations that you deliver quickly and regularly.
b. Accuracy. By building every day you iterate on the architecture and include the latest impacts of deliverables from other groups such as scope changes, technology choices and involved user groups.
c. Avoid the ‘Big Bang’. You do not want to fall into the trap of waiting for the all requirements or all technology decisions or all whatever to be made. There will always be changes in the things necessary to build the solution architecture. By the time all of these decisions are made the solution will already be done and your ideas for adoption are moot.
This is a good start to the list. I’m keen to collect other tips to add to this blog. If you want to add one, please comment and I’ll post it to this blog.