Podcast #03 - Decision Analysis Framework - 5 Steps To Better Decisions
James Lewis, our podcast moderator, is a senior Change Management professional with MI-GSO | PCUBED. David Appleyard, is a chartered quality professional (CQP, MCQI) and Prince2 / MSP practitioner with extensive international experience in the oil and gas industry. Ian Mulder, is a senior consultant who has delivered projects within the Financial Industry across many different management approaches including PRINCE2, Agile, Kanban and Lean Six Sigma. None of this mentions why they should be talking about decision analysis which means this is going to be an interesting podcast. We hope you stay tuned.
James Lewis: So good afternoon everybody. James Lewis here. I’m welcoming the MI-GSO | PCUBED community and possibly beyond – to the third episode of Coming Out Stronger, where we’re going to be talking about decision analysis framework.
I’m really interested in this because it’s something I personally know very little about and maybe it’ll be a relatively new subject for a lot of our listeners as well.
So, two colleagues are going to be handling the debate this afternoon. In the interviewer seat is David Appleyard, and Ian Mulder is going to be our SME who’s going to be answering David’s questions and having the conversation with him. Without further ado, David, I’m going to hand you over the baton.
David Appleyard: Thank you, James. Okay, Ian, it’s been a few weeks since we worked together on a client project, and I know you’re relatively new to MI-GSO | PCUBED. Can you please start by introducing yourself and telling us a little bit about yourself?
Ian Mulder: Fantastic. Sure thing, David, and thank you very much for having me on this podcast. I appreciate it.
We’ll start with MI-GSO | PCUBED. I’ve been here for roughly six months and I’ve worked on a couple of client projects. Before that I consulted within some of the biggest banks in South Africa, as well as several different software development companies. Before that, my post grad degree was in engineering and I specialized in decision analysis within that post grad degree.
David Appleyard: I understand you’re a pretty decent shot as well.
Ian Mulder: Ah, yes, that is true. I did shoot nationally within practical pistol shooting in South Africa. So that kept me busy on the weekends when I wasn’t at work.
David Appleyard: Well, let's unpack a decision analysis a little bit. For the first part, what is decision analysis?
Ian Mulder: Well, I think the easiest way that I like to describe it is – do you remember back in the glory days when we used to be able to travel to places? We had to travel to work, we had to travel to a friend’s house.
You used decision analysis techniques subconsciously there without even knowing it. With a little bit of the theory that goes into it, I can demonstrate it.
There’s a goal. You want to get to work, you want to get to your friend’s house. There are several constraints, whether it be time, money, the roots of particular preferences; and then there are certain variables that you get to change whether you wanted a particular thing or not.
Let’s go for an example. David, what were some of your options to get to the London office?
David Appleyard: I suppose, I’m going to have to take a train or drive, partially walk. I can understand this – there are several different ways that it could be done.
Ian Mulder: Sure. If we look at it – you could take a taxi the whole way if you wanted to. If you really wanted to, you could walk the whole way or you could get dropped off. Those are all the options, but they are not ideal. What you’ve basically done is you’ve started knocking off options.
You set a second constraint – perhaps money. I can’t be spending a hundred pounds, 200 pounds, whatever it may be to get to work every morning. It’s too expensive. The buses, they went straight there, though they put in time constraints. Let’s cut buses off.
Eventually you’ll start cutting those less attractive options out and you’ll be left with the car and the train. You have subconsciously made use of a decision tree, and through a process called constraint superiority, you’ve been left with a better option according to your constraints. So, if I remember correctly, you take the train into London?
David Appleyard: Normally if I am going in. Yes.
Ian Mulder: Because it’s, it’s convenient, am I correct? It sits in that balance between time and money.
David Appleyard: Yeah, I’d agree with that.
Ian Mulder: All right, so if you were to look at that monetary and time balance between the two, it leans more towards the train. But now if you were to drive to Milton Keynes, where we were based at before – it now leans towards more of the convenience of having the car.
David Appleyard: Yeah, that makes sense Ian.
Ian Mulder:So, I think you’ve basically done decision analysis without knowing it. One final point is that I think convenience stores lean on this as well. They lean on that equation between money and having to drive an extra five kilometres to save 50 pence. They know they can have their prices slightly higher, because of the convenience.
David Appleyard: So, I see that you can use decision analysis to narrow down your choices and your options. But what is the decision analysis framework and why are we better with this?
Ian Mulder: Within this particular framework, we’re going to take a basic decision-making process and ensure we have the right information.
Technically it’s a formalized approach to making decision analysis. What you want to try and do is get the variables, get the constraints and to make sure you have the right information to do that.
Let’s just say it’s a framework to guide you to get the right information – when you need it – at the right time. We need to understand the problem as well as possible, and then ensure that the right tools and techniques are used, for that particular problem. It becomes very difficult to model a situation if you don’t fully understand it.
David Appleyard: You came onto our project to work on decision making. I've not really heard of this framework. Where did the thinking come from?
Ian Mulder: There’s two main answers to this particular question. Where I received it from and where it was based. I think the answer to the first question is that I gained this knowledge in postgraduate studies. My thesis focused on the difference between subconscious decision making – when meeting a person the first time or, or seeing a particular product, that split second decision making – versus a more mathematical approach using light simulation models.
This framework comes in making sure that you choose the right particular tool. There are a plethora of quantitative, as well as qualitative decision making tools out there such as Monte Carlo simulations, financial modeling, game theory, heuristics, trial and error, pros and cons lists, whatever it may be.
What we are trying to do here is to make sure you use the correct tool. The decision analysis framework was designed in order to best understand which methods would work for each situation – whether you need a hammer or scalpel.
I think though that the main thinking of this started during World War II. Decision analysis was used to spread the limited resources that Britain and the UK had within their powers, through the country where it was needed most.
David Appleyard: Interesting. So, we brought you into the project about six months ago when the client had a particular problem. Could you talk us through that problem, what you did and the outcomes that you achieved?
Ian Mulder: The client had recently procured a company and with that inherited a rather large data center that they needed to figure out what to do with. I was brought in to help them make this decision. Should they move it to their own data center? Should they keep it there? Should they split it amongst their partners? These were all the options that we were trying to investigate.
How I approached this was – everyone loves a five-step process. So, we wanted to investigate, we wanted to develop options, we wanted to evaluate those options, we wanted to then test the particular models and then to iterate – to return it back into the initial understanding of the project.
During the investigation step, you want to understand the environment, you want to understand the viewpoints of people that are closest to this project; get their input, get their understanding of it. We also wanted to understand the constraints that were placed on it, as well as just general thinking of the team about it.
We then wanted to develop options: whether it be workshops, interviews, or brainstorming. We wanted to provide the options that they could eventually put on the table and say, even though this is a terrible idea – it’s going to cost you extra money, or it’s going to take extra-long – it’s an option.
And then test it [the options] against the assumptions that we initially made against the client thinking.
Next is to test the actual model and how it is performing. Sometimes models go wrong and you need to make sure that you’re not just running blindly into an abyss. We want to make sure that it continues to get better and better.
I ended up using a financial model, placed over a decision tree over those particular options that we initially discussed. I was able to show the client the main costs, benefits and drawbacks of each individual decision branch. They were able to then select which one they believed best worked for them; whether it be the high monetary input long-term, or whether it be a big bang transfer option.
The client was able to make a decision going forward, which was really great. That was the success criteria in the end. That they were able to finally make a decision, which I was happy about.
David Appleyard: In this case, there was several high profile stakeholders with some quite radically different views of what they wanted to get out of this. How do you find that using the decision framework enabled them to come to a consensus and make that decision making easier?
I remember from the start of this; the view from the client was the decision should be generally straightforward and easy to deliver. It’s just that all of the fog made it much more complicated.
Ian Mulder: Indeed. It was a lot of fun interacting with the different stakeholders. One stakeholder wants to do this, another stakeholder wants to do this, this stakeholder thinks this is best.
I think the beautiful thing about this process is, by making sure you get everyone’s viewpoints, acknowledging them, and building them into the particular choices; you’re able to sort of remove the noise out of the system. You’re able to see the main variables. You’re able to see the landscape for what it is and it ends in the right decision being made. It becomes very visible then.
By doing this, you’re enabling your client to make the decisions that are best for them. They’re able to see clearer about what the decisions they’re going to make now, will result in in several months or several years.
By placing the options on the table, the client can make the decisions for themselves. Similar to how you were knocking off travel options from getting to work. This process does something similar. You’re able to get rid of the options that aren’t necessarily attractive and bring the good options up to the top.
David Appleyard: Yeah. I suppose also when you’re modeling the, the financials over that decision tree. It gives you a sense of reliance on the facts as well.
Ian Mulder: You’re able to say, if we’re going option one, this is how much it’s going to cost. This is the high-level timeline it’s going to be, and there you go. That’s one option. Option three, we leave it as it is and nothing changes.
So it is this balancing act and you’re able to show them what decisions they’re making now will result in down the line.
David Appleyard: What would you say are your tips for success when using this framework and what would be the pitfalls that you'd want to avoid?
Ian Mulder: I think one of the biggest things that I easily fall into is large analysis paralysis. It’s so easy to fall into a particular pitfall and is a total momentum killer. You’ve got to stay above the current, you’ve got to keep moving.
Every now and then, it’s good to lean on this framework to say, this is what we’re doing next. Let’s keep moving forward. Let it get to the end of the iteration. See what happens, see what the problems were and then go back into it. If you keep analyzing, it just wastes time and eats up resources, so it’s something to be careful of.
I think another lesson that I learnt, especially in this project, is to consider your assumptions very carefully. They are often baked into the financial model at a foundation level which becomes very tricky to adjust if you needed to.
Make sure that the assumptions are thought out very well and everyone’s aware of them. I think that’s a key one.
Third, the information, the insight from the stakeholders and the people on the front lines is essential. If they’re not feeling listened to, it becomes very difficult for them to make the decisions. So, get a view of what the people in the trenches are thinking and try to understand where they are coming from.
David Appleyard: That makes absolute sense. What would you say, as we conclude this podcast would be your takeaway message to the audience concerning decision analysis framework?
Ian Mulder: The takeaway message I would say is decision making hides in plain sight all the time. We are consistently and continually making decisions. The time and attention for these decisions is essential.
There’s a fine line between that instantaneous decision making and that structured decision making that we’re doing. You definitely want to trust your gut, but back it up with some mathematical modeling.
One of my favorite quotes within decision analysis is throughout your life you’re making decisions, but it’s the combination of these decisions that you’re making that end up making you.
David Appleyard: Nice quote, thank you for your time today Ian.
Ian Mulder: Thank you very much. David.
James Lewis: So David, Ian, thank you very much indeed for that fascinating introduction to decision analysis framework. A great example of, I think what these podcasts are trying to achieve: aspects which will help our professional toolkit, which we may have had a fringe consciousness of but not really knowing a lot about.
Hopefully you know quite a bit more about it now. Thanks again to Charles Wilman our sound engineer and editor.
To everybody out in the community, have a good day and stay safe.
More on the same subject