Get the free Core Drives in the Wild guide, behavioral design applied to real products: professorgame.com/WildCD

Episode Summary

Rob breaks down why a product can feel stitched together even when every team involved did excellent work, using Conway’s Law as the lens: organizations that design systems are constrained to produce designs that copy their own communication structures. He walks through three real cases, a regional sales dashboard at Procter & Gamble where each region had picked a defensible and incompatible way to count sales, a client engagement where technology decision rights sat with a team that carried none of the consequences, and a project where several functions held contradicting views on what the design should optimize for. The argument is that functionality degrades gracefully under that fragmentation while motivational design does not, because the effect lives in the loop between Core Drives rather than in any single mechanic. Listeners learn where organizational seams show up inside the product and how to test for alignment before any design work starts.

About the Host

Rob Alvarez is Head of Engagement Strategy, Europe at The Octalysis Group (TOG), a leading gamification and behavioral design consultancy. A globally recognized gamification strategist and TEDx speaker, he founded and hosts Professor Game, the #1 gamification podcast, and has interviewed hundreds of global experts. He designs evidence-based engagement systems that drive motivation, loyalty, and results, and teaches LEGO® SERIOUS PLAY® and gamification at top institutions including IE Business School, EFMD, and EBS University across Europe, the Americas, and Asia.

Key Takeaways

• Melvin E. Conway argued in Datamation that organizations which design systems are constrained to produce designs that copy their own communication structures, which is why a product built across four departments reaches the user as four products.
• At Procter & Gamble, every region had chosen a defensible way to count sales, with its own aggregation rules and its own view of the target, so a single regional dashboard failed on reconciliation rather than on SKU volume or technical complexity.
• When the team that picks the technology carries none of the consequences of that choice, the decision optimizes for minimizing spend rather than for the problem the product owner is accountable for delivering.
• Functionality degrades gracefully under organizational fragmentation, because a fragmented dashboard is still a working dashboard. Motivational design does not degrade gracefully, because the effect lives in the loop, so reordering or removing pieces collapses it.
• Contradicting business metrics across functions is the failure that stops a project rather than slowing it. Business metrics are the first input to the design, so changing them after delivery means redesigning a significant part of the experience.
• Before any design starts, ask four people from four different departments what this product should optimize for. Matching answers mean design can begin. Different answers mean the alignment work comes first.

Topics Covered

• 0:00 – Opening hook, everybody did great work
• 0:30 – The single sales dashboard problem
• 1:35 – Where Conway’s Law comes from
• 2:18 – First case, every region counted sales differently
• 5:06 – Second case, decision rights sat elsewhere
• 8:34 – Why functionality degrades gracefully
• 9:05 – Why motivational design does not
• 11:29 – Third case, contradicting business metrics
• 14:53 – Comparing the three failures
• 16:30 – Specialization and owning the whole result

Mentioned in This Episode

• Conway’s Law, from “How Do Committees Invent?” by Melvin E. Conway, Datamation, 1968
• The Octalysis Framework and the 8 Core Drives
• Core Drives in the Wild, the free Professor Game guide

Free Resources and Get in Touch

• Core Drives in the Wild: Professor Game Free Guide
• Get Daily Value on Your Email
• Let’s chat about your gamification project
• YouTube
• LinkedIn
• Instagram
• Facebook
• Start Your Community on Skool for Free
• Ask a question

Looking forward to reading or hearing from you, Rob Full episode transcription (AI Generated)

Opening hook, everybody did great work

Rob Alvarez (0:00): Your product feels disjointed to the people that use it. And here is the strangest part. Everybody involved did a great job at what they did. You have set out to build a single experience and you’re also organized into departments. So what the user gets is stitched together at the borders of each of those departments for a functional product that is annoying. For motivational design, this can be completely fatal.

The single sales dashboard problem

Rob Alvarez (0:30): Many years ago, even before gamification was an intentional thing in my world. I was asked at a very large corporation I worked at to build a single sales dashboard. Across an entire region. Same company, same products, everybody essentially selling the same thing. I assumed the hard part would be the complexity and the thousands upon thousands of different SKUs that were being measured. But that was not it. Today I’ll show you why this happens, Rob Alvarez (1:00): why it damages motivational design far more than it damages any single feature. And what needs to be true before you set out to design any single thing? I’ll show you the three different places where it breaks. the third one is a dead stop. And by the way, if you want to get good at understanding how motivation works and why it is important to the design for single experiences, understand what is involved in all these experiences, I prepared for you a free guide, Core Drives in the Wild. All you have to do is click on the link in the description, Rob Alvarez (1:30): put your email there, and you’ll get all of the guides straight to your email.

Where Conway’s Law comes from

Rob Alvarez (1:35): So let’s start with a small cool fact. And it’s that almost 60 years ago, almost 60, 59 to be precise, somebody already was discussing this and why this is a problem. It was Melvin E. Conway in 1967. He said, Organizations which design systems in the very broad sense are constrained to produce designs which are copies of the communication structures of these. organizations. That’s as you know, products are some form of a system. Rob Alvarez (2:04): I’ll show you this in action, in reality, three separate times for three different reasons and how each of these can kill whatever you’re doing in many different ways. And of course the third one is a complete showstopper.

First case, every region counted sales differently

Rob Alvarez (2:18): Let’s start with a story I was teasing on the first few seconds of this episode. It was a very large organization. Geographically, pretty much all over the world, I think. They had each of them in the region I was working with, they had all made very smooth, sound decisions independently of the other regions on how they were counting things along the lines of sales. Because if you say, How do they count sales?, well. How much do you sell? This much, Rob Alvarez (2:47): that’s it, right? But there’s a lot of nuance when you are counting for sales. There’s especially in large complex organizations, as I said, I it thousands, has thousands of SKUs. There’s a lot of complexity. And how do you measure it? How do you aggregate regions? How are you measuring towards the target? Are you using percentages? Are you using a total? Are you establishing that there was many, many things involved there. Every single one of those methods they were using, Rob Alvarez (3:16): they had a very good reason for actually establishing. Again, they were all very good reasons. They were all defensible. When I was speaking to each one of those leaders who were designing, who had designed or who were working day to day with those measurement systems, they all were very, very reasonable, very sound, very good business decisions. Failure was not making bad decisions. The problem was more of systems design, Rob Alvarez (3:44): how the organization had been up until that point designed. It had to do with how you were able to integrate from one region, how you were able to compare from one region to another, from one country to another. And that was a design problem of the system. It had nothing to do with those individual great. Very well-intended, very smart business people who were involved. The organization, you might ask yourself, none other than Procter & Gamble. I worked there many, many, many years ago. Rob Alvarez (4:19): All of those decisions were very, very defensible. The issue was that it was harder and harder to get to see the whole picture because the the the level of complexity that was happening on the teams was not exported. Than to their managers. And the person who was seeing the whole picture got reviews directly from those people. But in a in an era where data is and can and should be available to everyone, the issue here was that they should be at all times able to access that data, to see that data in the right way and compare it across regions in Rob Alvarez (4:54): a very simplified manner. So who owns that whole measurement? That was a question that they started to answer, in fact, very well. That was literally my job to help answer that question.

Second case, decision rights sat elsewhere

Rob Alvarez (5:06): for part two, new gig, new problems, right? So this time we were designing for a system that had to involve both the content that we were going to be delivering, the stuff, you know, the the the the thing that the system had, but it also had to include somehow the sales system. It had to have ideally it would be something that was fully integrated, or we could have separate solutions that integrate it well together so that this could function in an automated way. This was an organization that was very used to doing a lot of manual intervention Rob Alvarez (5:36): as well. It still does for other for other pieces of of of the product, but this piece required something very different from what was traditionally done. So the technology available was not sufficient, was not necessary, was not prepared for doing that. The problem, the people who owned what was going to be done, which was my my the team I worked at. was not the same team that had the decision rights over what was going to be done technology wise. So the technology team decided to use a certain vendor that had a mediocre solution of an open source software, Rob Alvarez (6:10): which they adapted and personalized, but not really personalized for exactly what we needed. It was a generic general personalization that became restrictive even though it was open source and you could pretty much do anything. What they did was so restrictive that we couldn’t really fit our model into what they had. So we were suffering as the people who were supposed to deliver this product. We were in charge of the results of the products, but we were not in charge of making the choices that involved how Rob Alvarez (6:41): the products could be delivered, amongst other things, with or mainly because of the technology. So again, different departments, there was a place where we couldn’t really meet. the boundary cuts here were authority decision making. different departments, it makes sense. All c companies large larger, the more this happens, have different departments. But here, instead of using the the data to make the decision, Rob Alvarez (7:06): it was a I don’t want to say political because it was not just political. It was literally the functional thing. It’s like, these people make these choices. And the level of influence we could have over that decision was so, so minimal that it made no sense. We had tech people who knew the vendors, who knew the stack, who knew everything that we needed, who knew people who could provide better solutions. In fact, some of them were saying that even if if we took a little bit more of time, Rob Alvarez (7:31): we would have to take a little bit more of time, we could have developed something better using, for example, an open source solution like this vendor had, and it would be ours, we’d be owning it. But this was a Technology team solution solution decision, and we couldn’t make that decision. So, does the person choosing in this case the technology carry the consequences of that choice? Rob Alvarez (7:55): And that was definitely not the case. They were only owning, and that was another one of the issues, they did own the budget, paying for this technological solution. So we had literally no say in that. they were minimizing how much they had to invest in this new technology rather than making sure that the technology was actually fully solving for the problems that needed. They were only checking a box, which is something that also happens when you start thinking of motivational design. You say, oh motivational design, Rob Alvarez (8:23): I use this, I put points, I put badges, I put leaderboards. Motivational design is done. Behavioral design, it’s check, you know. We have gamification, and that is definitely not the case.

Why functionality degrades gracefully

Rob Alvarez (8:34): But before diving fully into that once again, here something that I want to make sure that you that you see very clearly is that the functionality can let’s call it degrade very, very gracefully, right? You know, the dashboards I was talking about before, they’re still dashboards, you know, in their own sense and they’re functioning, they’re working. it’s harder to then integrate, Rob Alvarez (8:57): but they’re still functioning. Like it’s it’s working, it’s doing its daily, weekly, monthly, even yearly job. That is still there.

Why motivational design does not

Rob Alvarez (9:05): With motivational design, which is what I was sharing just a second ago, the the degradation is not is not graceful. This whole concept lives in the whole journey as itself. If you don’t see it as a system, as a full journey, the issue is you go back to what used to be the way things were being done, which is basically linear design. You do this, you do this, you get that, you do this, you do this, you get that. Rob Alvarez (9:32): There is no way in which, if you don’t understand the whole system, the whole loop that is being designed, what we call the game loop, you take something away, and maybe you have a game loop no more. And usually it’s not you take one thing away, it’s you start reshuffling things, the order of things, you put some things away, you introduce some new things, and again, that brings you to a place where you no longer have a functioning, Rob Alvarez (9:59): not functional, a functioning loop. That keeps people coming back. this has to do with the way companies communicate, as Conway said many, many years ago. And it has to do with the way the companies are organized. One team who is able to implement one of the parts goes ahead and implements it. They go and do it alone. When you do that and you separate that into multiple different silos, and you also, you know, as everybody is doing nowadays, Rob Alvarez (10:26): you have a data-based approach. You end up With what we called many episodes back, almost I think it’s almost a year back, we were talking about how this ends up degrading into fully, almost only Black Hat design. You have points, you have urgency, you have things that in the data, in the immediate data, they show up and bring you results, Rob Alvarez (10:48): but they do not think of the journey of the user in the longer term. This is something that if you’re not establishing a relationship, which is very, very, very rarely the case, it might work. For that. And there’s very few ex exceptions where that can work. But typically, when you use an only a data-focused approach, you’re looking at the immediate results of any one of those actions. And the A-B tests that we mentioned there are also a consequence of this. The broader, more structural problem that we’re looking at here is Rob Alvarez (11:17): why this happens in very good meaning, very well thoughtful, smart organizations that are organizing them their them themselves the way they we always have. And they’re not having owners who look at the whole picture,

Third case, contradicting business metrics

Rob Alvarez (11:29): And that’s what brings me to this third case study. This is heavily under NDA, so I cannot give you a lot of details. But this was a a client where we did our whole job, we delivered the design. And you know, we s always stay in contact. I was a I was a team lead for for this project in particular. And talking to some of the members of the team, they started realizing that the company, since again, everything is separated in silos, Rob Alvarez (11:58): as usual. We start by the business metrics, right? That’s literally the first thing that we need to make sure is established because they are the foundation of everything else that we build out there. We are optimizing for business metric number one. And next to that, business metric number two, then number three, then number four. And if we always, if we if and when we have to do trade-offs, we’re always going to optimize for the higher ranking business metrics. Rob Alvarez (12:24): So that is an agreement we need to very heavily cement at the very start. And this organization realized that they, even after months of engagement and working through the project, delivering final designs with battle plan, everything was there. They realized that there were other departments who are saying, you know what, we cannot be optimizing for this. There’s this other thing that we also need to need to optimize for. and then another department say, Rob Alvarez (12:51): Yeah, you know what, you’re right, that’s interesting. But more interesting is this other thing, you know, and that’s the typical You know, I I also teach operations and as I always tell my students that this is typical in every organization you have. You know, finance is optimizing for minimizing risk and optimizing the the financial resources. And then marketing is trying to sell more and get more budget to be able to invest more in marketing. Operations is trying to optimize for being things being sort Rob Alvarez (13:16): of physically as efficient as they can. And that might involve, you know, big investments that finance is not gonna approve. But it also involves maybe some things that marketing cannot do now because you know things are more efficient that way. So there’s a constant fight between this. If you bring that into a project and you’re not able to bring that strong alignment, and that’s oftentimes why also we we want alignment that comes from the bottom down. It says this is what we’re gonna do, Rob Alvarez (13:41): and everybody needs to align down on this, then it it becomes very, very complex because if you have a finished project. And you start changing the business metrics, you don’t just change what you’re going to measure after the project. Because if you do that, you’d realize from the very get-go that you were optimizing for something that is not now what you need to optimize for. When you change that, the design needs to change if you still, you know, if you still care about actually optimizing, not just mentioning those new business metrics. Rob Alvarez (14:10): So if multiple parts or functions hold very contradicting business metrics, that has to be brought together. before the design happens. we give them the design, then they then it’s pretty much theirs to implement most of the time, depending on how how we engage. We typically do recommend an extra stage where we sit with them and guide them along the implementation phase so that we make sure that the motivational design stays very, very potent. But this, you know, Rob Alvarez (14:36): this was not the case. They’re in their right. That’s how every organization works. And now the changing of the business metrics, you know, they still at least haven’t decided to to bring us back. But that would require if they really changed the business metrics, that that would definitely require a redesign of the experience because now it needs to optimize for something else.

Comparing the three failures

Rob Alvarez (14:53): Look at what just happened with these three. At P&G, it was, you know, it was numbers to be reconciled slowly, painfully. In fact, it was my job to do exactly that, reconcile them, and then of course build the system so that this in the future didn’t happen that way. For technology, there’s always a way to recover from it. It’s expensive, you know, it takes time, changing vendors, building around it. These are two functional problems. Rob Alvarez (15:17): There’s something you can just invest into and recover from. This third problem is a very different animal. This is not just one of the inputs of a design. This is the first thing that we talk about. Everything else we build is on top of these business metrics. You cannot just average out contradicting answers as an input for the design. You can’t just sort it out later and see what happens. Everything you had built was optimizing for something else. And when you design that way, which is the right way to design, Rob Alvarez (15:53): and you change the foundations, yes, you can redesign, but you have to actually do it a significant portion of it all over again. The first two I mentioned, they’re terrible. They cost you, they cost effort. This third one has to stop you in your tracks. If you’re working on a project and you’re trying to build something new, and you ask, you know, four people from four different departments, what should this optimize for? Rob Alvarez (16:24): You want to make sure that they are giving you the exact same answer.

Specialization and owning the whole result

Rob Alvarez (16:30): Every again, every one of these decisions we’ve gone through was rational. It was smart. It made sense. Specialization is why and how. Large organizations are able to scale and do things at massive scales globally, right? These global, massive corporations are only able to do that because of specialization. I’m not against specialization. The cost of specialization is that it makes it harder for anybody to be able to hold the whole product, the whole system in a nutshell and be able to optimize for what needs to be done. So that’s that’s a very high risk. Rob Alvarez (17:07): One person Or one group should be owning the results. And when they own the results, they are the ones who are in charge. And this in product management as well is something that product managers should be doing, gathering and aligning everybody on what it is that is being optimized for, right? When you’re going to implement a new feature, which is a feature that you’re going to prioritize, Rob Alvarez (17:29): that’s something that product management, good product management, is constantly doing. It’s a significant and meaningful part of the job. If you don’t have that implemented directly there, when you bring somebody in to do gamification or when you want to do it yourself, you need to make sure that alignment is already there. Contradictory answers to the question, what are we optimizing for? is what blocks or absolutely kills a project. If you try to change the business metric after you are in implementation phase, Rob Alvarez (17:59): you are absolutely going to kill. The project because the motivation you optimize for is no longer right. So if you don’t want this to happen to your product, go and click on the link below and get started. Just put your email, we’ll send you one email per day so that you get a read of each of these Core Drives in real business scenarios, of course, with my read, with my recommendations as well on that. And it’s from past episodes of the podcast. So you will be listening to real other people talking about this and you’ll Rob Alvarez (18:30): get my own read on how and why this works or why this doesn’t work. And of course, as always, at least for now and for today, it is time to say that it’s game over. End of transcription