Game Theory in Management

Recently I took a trip and stopped by a newsstand on my way out of my home town’s airport to purchase The Five Dysfunctions Of A Team, by Patrick Lencioni (Jossey-Bass, 2002), to read on the flight. I was hoping that I would find that this book landed squarely in the genre of management science; alas, it was not to be (I should have paid more attention to the subtitle on the cover, which states that it is “A Leadership Fable,” albeit in a much smaller font). Much like Eliyahu Goldratt’s Critical Chain (North River Press, 1997), or even Walden Two by B.F. Skinner (Hackett Publishing, 1948), The Five Dysfunctions Of A Team is a work of fiction that advances a hypothesis.

Let me say that again: it’s a work of fiction that advances a hypothesis.

I remember reading a book by a martial arts expert (it may have been Bruce Tegner) where he lamented receiving letters from fans and students asking about a certain style of karate that they had seen in a movie or television show, and asking if that particular style was truly as effective as depicted. He would have to remind them that what they were watching was a fictional account. That particular style “won” because the script writers had determined that its practitioner would “win.” In the same manner, Goldratt’s characters attributed their success due to their use of the theory of constraints, Skinner’s characters credited attaining near-utopia to their adaptation of the precepts of Behaviorism, and Lencioni’s characters ascribed their company’s turn-around to the protagonist’s insights on leadership and team dynamics. In fact, all of these “successes” were predetermined. They triumphed because those authors wrote that they should in these works of fiction. In fact, in one of the chapters in The Five Dysfunctions Of A Team, the central technical issue facing our protagonist has to do with the proper balancing of the needs of the Asset, Project (product), and Strategic managers, which is what Corner Cube Theory is all about. But even here, the principals reach their solution through compromise attained in meetings – quantitatively balancing the three competing interests in a three-dimensional model is never discussed.

And yet, as intellectually vacuous as I find an attempt to advance a hypothesis through “fables” or other works of fiction, there’s no denying that, in many instances, it works. The three books cited above were all highly successful. Walden Two, combined with Beyond Freedom And Dignity (Hackett Publishing, 1971) had a profound impact on many university’s psychology departments, and there’s a Theory of Constraints Institute based on Goldratt’s writings. And, of course, The Five Dysfunctions Of A Team made it onto the New York Time’s bestseller list. These facts are leading me towards an unwanted conclusion, essentially, if I can’t beat ‘em, I may as well join ‘em.

“But Michael!” I can hear loooonnnng-time GTIM Nation citizens object, “what about Stanly T. Raspberry?” For relative newcomers to this blog, Stanly T. Raspberry is a PM-centric private detective who first appeared in my long-running Variance Threshold column in PMNetwork, and has made several appearances in this blog. He “solves” business model pathology problems the same way that the fictional characters previously mentioned attain success, by his creator writing his experiences that way. My response to this objection would be that Stanly never advocated for a particular management schema, past some plain vanilla PM and organizational behavior and performance axioms. Rather, clients ask for him to investigate the goings-on within a particular organization (usually the Monolithic Corporation), and Stanly does so. Besides, mocking business model pathologies in an 800-word column or blog using a clearly satirical fictional character is very different from creating a nominally legitimate protagonist that overcomes fearful odds by engaging a contemporary management science theory in what is essentially a novel.

This being the case, it is my intent to spend at least a couple of GTIM blogs on an elongated Stanly T. Raspberry PM adventure, only this time he will be actively promoting a relatively new management science concept, the aforementioned Corner Cube theory. If any GTIM Nation citizens wish to have a character based on them appear in this adventure, let me know in the comment section, and I’ll accommodate as many as I can. Candidates will need to tell me about their character, if they should be considered among Stanly’s allies or opponents, and whether or not you want first dibs on actually playing this character should a Hollywood producer sees these blogs, and wants to make a movie out of it.

Coming up next: Stanly T. Raspberry gets a new case!

Well, maybe not literally a Collie. But after having raised Collies most of my adult life, I can attest that they consistently demonstrate a variety of instinctive behaviors that would serve managers well in a business setting.

Exhibit A for this assertion has to be their unfailing loyalty to their Team. Their highly-developed senses are constantly scanning the environment for potential hazards or dangers and, if discovered, will immediately and directly address the hazard to resolution. Members of a Project Team who perceive that their manager is willing to tolerate a condition or individual within said Team that represents a threat to either accomplishing the Project’s scope on-time, on-budget, or (worse) is a menace to their professional advancement, will almost certainly lose faith in that manager, or even the entire Project.

Collies are also famous for their uncanny ability to figure out the optimal technical approach to problems they face. In 1954, a Collie named Tang planted himself in front of a milk delivery truck, barking and refusing to budge. The driver could not figure out how to get Tang out of the way until he checked the back of the truck, where a two-year-old girl had climbed into the truck, and would have certainly fallen out and spilled onto the street had the truck moved.[i] I’ve always found this particular hero-dog story compelling because of what Tang the Collie didn’t do. He didn’t go to the back of the truck to bark and raise the alarm – he went to the front, seemingly aware that the truck would leave going forward, and that he had to stop that from happening. And, of course, Collies are also famous for being easily trained in the herding behaviors, knowing just when and how to circle, bark, or feign stalking/charging techniques to keep their charges relatively confined. I shudder to think of the number of managers with whom I’ve worked who fail to identify a superior (or even workable) technical approach, opting instead to use some familiar, canned PM strategy when undertaking new scope, even when such an approach is clearly not appropriate.

When my wife was expecting with our boys, our Collie would not let her out of his sight when she was home. It’s as if he just knew that she was in a special, more vulnerable condition, and made it a point to be hyper-vigilant when she was around. This behavior was so acute that, when my wife would get up to go to the restroom, the dog would follow close behind and sit staring at the door until she came out. He would then closely follow her back to the sofa, and resume his place at her feet. Compare and contrast this natural behavior with those PMs who fail to closely monitor the most vulnerable part of the schedule network, the Critical Path, or review the progress of those tasks almost casually, not realizing the potential implications of poor performance.

Collies were originally bred in Scotland, and I remember reading about the sheep pens in those highlands. They were roughly circular, with a single gap in the fence for access, but no gate. Instead, either the shepherd or Collie would sleep across the gap, so that nothing could go in or come out without their knowing about it. Along those lines, I find it maddening when a so-called PM of anything but the smallest of projects refuses to implement an Earned Value Management System, or EVMS. Earned Value systems can be rather simple, but they are absolutely necessary if the Project’s cost and schedule performance is to be measured and relayed to management. Eschewing such systems is essentially broadcasting to the management world that it’s okay for the wolves to come trapsing into the pen the PM doesn’t care about bringing the Project in on-time, on-budget.

I don’t have any first-hand knowledge of this next aspect, but I have read that working Collies who have been trained to herd cattle should never be re-purposed for sheep. I guess the more aggressive techniques needed for bovines don’t work well for the rams and ewes, and might actually harm them. Referencing last week’s blog, where I discuss some of the negative management manifestations stemming from organizations that have moved from great original idea to monetization to keep-it-going-just-to-keep-it-going phase, managers who thrive in this last organization type may not be suitable for organizations that are more closely aligned with their founders’ original vision. The sharp elbows needed to thrive – or even survive – in highly politicized Project Management Offices are usually out-of-place in high-performing Project Teams, aiming at on-time, on-budget scope delivery. Such managers may actually harm high-performing Project Teams.

On the other hand, you should probably cut ties with your new PM if he steals socks from your sock drawer, and defiantly runs around the house with them, or gets into the trash only to regurgitate onto the rug later, or engages in a shedding-blast that overwhelms your vacuum cleaner’s ability to keep your floor clean, or…

[i] Retrieved from https://wellsmerecollies.com/the-hero-collie on December 27, 2024, 13:48 MST.

It’s been my experience that one of the most – if not THE most – intransigent and difficult problems with setting up and maintaining a successful Project Management Office has to do with the implementation phase. Oh, sure, some of the more technical Projects can present quite the challenge when setting up the Scope, Cost, and/or Schedule Baselines, but once those are in-place the actual Cost/Schedule performance measurement systems aren’t that tough. Baseline Change Control / Configuration Management doesn’t require galaxy-brains either. No, the most obstinate barrier for the PMO, to my mind, has to be the implementation phase. After every last organizational element has been made aware of the effort(s) to advance the PM capability, and Earned Value / Critical Path Methodologies software packages have been installed on multiple machines, PMs, Control Account Managers (CAMs), and Work Package Managers have been identified, and even the General Ledger has been prepped to track costs at the reporting level of the WBS, somehow, when it comes time to actually demonstrate an acceptable level of overall PM expertise, things are found to be wanting. And I’m not talking about finding “it’s” instead of “its” in the Variance Analysis Reports, as infuriating as that may be. I’m talking about Organizational Breakdown Structure elements somehow finding their way into the WBS, or trying to claim the percent complete method as Level-of-Effort when it’s obvious that that particular task should be using Direct Units, or filing a Baseline Change Proposal for no other reason than the Control Account encountered a negative Cost Variance. These are not fat-finger-style errors. Rather, they point to a highly problematic lack of PM capability maturity which, in turn, can have rather serious repercussions across the entire project portfolio. What is a PMO Director to do?

Well, the first thing our afflicted PMO Director needs to do is to recognize the type of business model environment where she finds herself. GTIM Nation will recall my previous discussion of the Eric Hoffer quote,

Every great cause begins as a movement, becomes a business, and eventually degenerates into a racket.[i]

If I could be so presumptuous as to paraphrase Hoffer, I think the management version of this would be:

Every successful business endeavor begins with a great insight, which then becomes monetized, and eventually turns into an organization that exists primarily to keep itself going, having largely distanced itself from the founders’ original vision, or insight.

If we assume that this is a workable take on the long-term evolution of many organizations’ business models, the implications for the introduction and/or maintenance of the PMO can be profound. For example, if the target macro-organization is already in the just-keep-it-going phase, then any attempt at advancing a capability maturity in general, and PM in particular, is in for a very difficult time. Unless an advanced PM capability was already in-place, and therefore part of the machine that needs to be kept operating, this type of change to an ossified business model is going to be resisted to Cecil B. DeMille proportions. It will make absolutely no difference if the technical approach to the PM capability advancement is the most excellent ever devised – the implementation strategy that gets it quickly and fully operational doesn’t exist for this type of organization.

Also consider the type of workers that tend to populate the organization in the keep-it-going-to-keep-it-going phase. Drawing from the archetypes presented in Michael Maccoby’s excellent book The Gamesman (Simon and Schuster, 1976), the “great insight” phase is likely to see many Gamesman and Craftsman types. As the movement towards monetizing the great insight occurs, more Company Men will occupy the ranks, and more than a few Jungle Fighters are likely to be present. But by the time the keep-going phase is realized, it’s going to be dominated by Jungle Fighters and Company Men, if for no other reason than genuine Gamesmen and Craftsmen will find this type of organization less appealing, if not out-and-out intolerable, and will move on. I’ve worked for organizations that have placed considerable corporate culture distance between themselves and their founders’ original vision, with a significant number of Jungle Fighters and Company Men in upper management, and can confidently assert that such a work environment is both professionally frustrating and anxiety inducing. Advancement in such organizations – heck, even retaining a current position – has less and less to do with merit or virtue, and more to do with the political machinations among its decision makers.

My recommendation to GTIM Nation would be to maintain your managerial expertise – your PM virtue, if you will – and let the chips fall where they may. If you succeed, great. If not, the organization is likely to be on a downward trajectory anyway, and is doing you a favor by signaling as such.

[i] Retrieved from https://www.goodreads.com/quotes/98215-every-great-cause-begins-as-a-movement-becomes-a-business on December 23, 2024, 17:53 MST.

Worry not, GTIM Nation. I’m not trying to influence our beloved PMI® away from serious academic discussions on the management sciences and towards the tawdry. But I do think that there are some insights into the aforementioned management sciences that can be gleaned if we engage in a specific type of reductionism, stripping away if you will some of the layers of rhetorical fluff that accompany each of the management types from Corner Cube theory fame.[i] These three types of management are:

• Asset Management, with its main objective being “maximize shareholder wealth,”
• Project Management, centered around delivering what the customer wants (Scope) within the constraints of Cost and Schedule, and
• Strategic Management, focused on maximizing the organization’s market share.

I think it’s important to note that these three are different by type, and not just degree. They seek very different outcomes, and use different Management Information Systems to attain those outcomes. I’m looking to demonstrate that both Asset and Strategic Management largely take PM for granted, along the lines of simply assuming that their organizations already produce a good or service for an already-existing customer base. For the sake of this mental exercise, let’s begin our reductionism by removing organizations from the analysis. Our examination will posit individual people interacting with other individuals, so that the nature of the management types will become plain.

So, a single Asset Management-type approaches a potential customer in a whatever setting – shopping mall, office building, sidewalk – it doesn’t matter. What does the Asset Management-type want? To “maximize shareholder wealth,” of course. College-level business courses will maintain that as the ultimate goal of all management, which has always struck me as both classical theoretical overreach and rather arrogant. How does our Asset Management-type accomplish this in such a one-on-one setting? They could rob the potential customer, but that’s against the law. They could try and trick the potential customer into voluntarily providing the shareholder wealth, but that’s frowned upon, and also potentially illegal. Hmmm. What else remains? Well, they could offer some sort of product or service for which the potential customer would voluntarily give them the sought-after shareholder wealth, but that would land them in the Project Management realm, would it not?

Now let’s take a look at our friends, the Strategic Managers. Back in their organizations they are tasked with maximizing market share, typically through advertising or other marketing initiatives, but it can include such extreme measures as conducting a hostile buy-out of a competitor. But here, in the mall/office building/sidewalk setting, they really have nothing to offer without a separate piece of scope, which again, belongs to PM. I mean, sure, they could hover around the potential customer to try and ensure that they don’t seek to obtain the whatever goods or services from someone other than them, but they, themselves, don’t create a stand-alone good or service that would entice the potential customer to voluntarily part with their shareholder wealth money. In those instances where the particular Strategic Manager follows the potential customer to prevent the choice towards another organization with something more than verbal pleadings, the result is known as coercion, which can also be illegal.

And so we arrive at the Project Manager. What would the PM-type do, interacting with a potential customer, one-on-one? The answer to this question is the same as to the first principle of PM: discover the Scope. What does the customer want (Scope Baseline)? How much are they willing to pay (Cost Baseline)? When do they want it (Schedule Baseline)? Such proffered service is what drives a free-market economy – not “maximizing shareholder wealth,” and not dominating a specific market, though those outcomes can be achieved once the PM objective is attained.

Also consider the results of the stripped-down management types. As noted above,

• Asset Management cannot attain its goals without providing a good or service.
• The same is true of Strategic Management.
• Project Management without Asset nor Strategic Management, turns into … volunteer work!

Think about it – providing a potential customer with a good or service that they want without it being monetized or influencing market share is literally volunteer work.

I want to be crystal clear here: Asset Management and Strategic Management are fully legitimate approaches to business, worthy of study and attained expertise. And yet it must be pointed out that they proceed from a proffered good or service, and the management of the creation of goods and services belongs in the Project Management domain.

So, yeah, let’s look at those scantily-clad business models. It’s okay to conclude that ours is the most attractive.

[i] Hatfield, M. A. (1995). Managing to the corner cube: three-dimensional management in a three-dimensional world. Project Management Journal, 26(1), 13–20.

In last week’s blog I cited Hatfield’s Rule Of Management #3, which states:

3. The 80^th percentile best managers who have access to only 20% of the information needed to obviate a given decision will be consistently out-performed by the 20^th percentile worst managers who have access to 80% of the information so needed.

For GTIM Nation citizens who recognize that rule as a derivative of the Pareto Principal, go to the head of the class. I now want to introduce another PM-esque derivative of the Pareto Principal, this one sitting at #29 of the aforementioned Rules Of Management:

29. 80% of the value of a Project Management Office (PMO) is derived from 20% of the information it generates.

Does this sound outrageous? Indefensible? Consider the following: let’s postulate PMO (A) and PMO (N). PMO (A) sets its technical agenda so as to include the following:

• A robust risk management (no initial caps) system, including a risk register (nic) documenting the results of a thorough risk analysis (nic), with an ongoing re-evaluation of risk events, their estimated cost and schedule impacts, and additional risk event scenarios going forward.
• A fully-staffed communications group, who seek out all potential stakeholders, and employ the mechanisms to ensure that their “input” or “feedback” is properly incorporated into the formulation and execution of the Projects’ technical approach and, in some cases, reflected back to the macro-organization in such a way as to actually influence its business model.
• A strong Quality Management sub-team, providing Ishikawa Diagrams of each and every decision point that affects the Projects’ technical approach to resolving the scope, as well as Six Sigma analysis of the accompanying processes or any out-of-threshold variances encountered along the way.

Now let’s take a look at the setup of PMO (N). It includes:

• A robust Scope Baseline capture, including a valid Work Breakdown Structure, with Control Accounts and Work Packages sufficiently detailed to not only serve as the basis for deriving the Cost and Schedule Baselines, but also as a usable defense in the event of a conflict with any subcontractors (or even customers) over performance.
• A basic Earned Value Management System, one that allows for reliable cost performance measurement and accompanying accurate capacity for calculating at-completion cost and duration.
• And, since PMO (N) does this for all of the Projects in the portfolio, it can roll-up its information streams to indicate overall portfolio performance,

…and that’s it.

All other things being equal, I would argue that the organization served by PMO (N) would wayyyyy outperform the organization saddled with PMO (A). Just on its face PMO (N)’s information streams are more valuable than (A)’s, and, in my opinion, it’s not close.

Then we have the nature of the information streams of PMO (A). Consider what the risk managers (nic) alone would need to perform their “analysis.” It includes hours of time with the Work Package / Control Account Managers, asking them (or some other Subject Matter Expert) about the alternative outcomes of the selected Project’s execution strategy, their impact and estimated odds of occurrence, over multiple alternative outcomes. Also consider what happens at the end of such analysis: other than contributing data to the estimate of a potential contingency budget, it’s little more than a list of things that might go wrong, tripped out in Gaussian Curve jargon.

Then we have our friends, the Communications Managers. Don’t get me wrong – as long as these advocates structure a usable communications plan that aids in the Public Affairs aspect of Projects, particularly high-profile ones, I’m completely good with this function. It’s just that some of these experts, in my experience, often start pushing this business about “engaging all stakeholders.” The complete population of “stakeholders” will invariably contain people who are against your Project coming in on-time, on-budget, or even its existence. Giving these stakeholders unearned influence in the setting of the technical approach, appropriation of resources, and the manner or tempo of the work being executed is counter-productive in the best of circumstances. PMs who not only pursue this strategy but actually pay Team members (or even outside consultants) to do so would be well-served to re-examine their nominal approach to performing actual Project Management.

Next let’s look at the resources and time needed for Quality Management experts to perform their analysis and generate their specifically-tailored information streams. Useful for an ex-post-facto analysis of something that went wrong? Absolutely. But unless your Project provides a good or service for which any error could have massive, destructive outcomes, you might want to think twice before funding their budget.

And here’s the kicker: PMO (N) is, in all probability, going to cost less than PMO (A). Maybe not an exact Pareto Principle split of 80/20, but close enough to answer the question in the title. What can 20% of a PMO get you? Turns out, quite a lot, if it’s the right 20%.