A Place called Something
There is a very simple scenario that hampers getting data into an operations management system. In this short post I want to describe that scenario, and some factors that often drive it.
Intro
One of the three critical phases of any data ecosystem is the “input” phase. (I wrote about the three phases in a previous article explaining the input, storage and reporting phases.) But one concept that especially hinders the input process is a phenomenon I call “A Place called Something”.
Let me explain.
Reluctance to Divulge
When you have created a data system, it’s critical to continually enter new data on projects and workload to keep it up-to-date. There are several challenges to gathering data, but one issue that most often obstructs the collection of this data is that people often haven’t nailed down ALL the details of the projects that they are working on.
As a result, because they don't know everything, they're uncomfortable to give you any information so that they're not wrong and don't want to give you inaccurate information. (Somehow giving inaccurate information is seen as more harmful than giving no information, reminiscent of the famous TED talk by Sir Ken Robinson about ‘how schools kill creativity’.) But I digress.
For any project operations leader building their database, this phenomenon where people are reluctant to give partial information, is what I call “a place called something”, and the conversation usually goes something like below. I know that manager X is pursuing a great opportunity with a client, and that a new big project is likely or possible:
Me: “Hi Rob, can you give me an idea of the fee, and start and end dates for your project so I can get them in the database?”
Rob: “Actually I’m not sure I can. I’m not certain when that project is going to start; they may need to meet with some of the stakeholders before they can even know.”
Me: “Can you give me a guess?”
Rob: “No, I don't want to. I don’t want to give you the wrong info”
This is where I then kick in the “something” strategy.
Me: “OK, I just need an order of magnitude. Like is it a 10,000, 100,000 or a million dollar fee?” And is it going to start late this year, or early next year, late next year, for example?”
If it drags on past this point, I may have to say “OK, I’m going to put in placeholder for $600K starting in May, and lasting 3 years.” What’s critical is that my need to get “something” entered eclipses everything else. Once I have even some ballpark figures, I can start to get some semblance of information entered. That way I can at least be aware of and track the project. But I need to get something in the system.
Finding “Something”
The key concept here is that “somewhere between nothing and everything is a place called something”., and Even with inaccurate information, we can create a budget projection. we can always shift dates as they change, and watch the model shift.
The key concept here is that “somewhere between nothing and everything is a place called something”.
Also, it’s worth remembering that all plans are inherently inaccurate, not because they aren’t right but because plans change all the time. As soon as dates are settled, certain forces usually arise, and the dates will change. There is a popular saying in data analytics that “all [data] models are wrong, but some models are useful”. This is a similar concept to the “something” idea - the plan may not be dead on, but it will be a useful vehicle to track workload and revenue.
I have used various quick checklists to get the information that I need to outline a project. The list below shows a very simple set of critical items to get started:
Project Name
Estimated Fee:
Start Date/ End Date:
Disciplines Used:
Consultants:
The Need for Input
While the concept above may seem blindingly obvious, this idea about being OK with “a place called something” is really important if you're going to start building a robust data system. It is often the case that managers are reluctant to guess, but in my experience having no idea was a lot worse than having an inaccurate idea. With the former, the firm was essentially flying blind, while with the latter the workload is at least on the radar, can be tracked, and the assumptions reviewed.
Analytics is only as good as the data that feeds it. Getting that data to start with is often not as straightforward as it would first appear. But driven operations leaders need to be aware of the “something” phenomenon, and how to push through to get required information for operations. With data, something is definitely better than nothing.