Game Theory in Management

According to FOOD52, the Japanese cuisine item sushi made its American debut in Los Angeles, California, in the late 1960s[i]. By the 1990s its popularity had exploded to the point that it was not at all unusual for people in the Midwest to go out for sushi, and restaurants that served it were positively ubiquitous in the major cities. Another venue for serving sushi was gas stations, which might strike my overseas readers as strange, since sushi is made up of, basically, fresh, uncooked fish, seaweed, and rice, with a variety of additional ingredients typically available. Uncooked fish doesn’t stay fresh (or even edible) for long, not to mention seaweed. And gas stations aren’t known for their ability to produce, maintain, or sell items that have particularly short shelf-lives, historically opting for foodstuffs that could probably survive a nuclear attack. I don’t mean to disparage entire retail chains selling a particular item – but I would never eat sushi sold at such a venue.

Meanwhile, back in the Project Portfolio Management World…

Referring back to my personal adaptation of the Pareto Principle, that the 80^th percentile best managers with access to 20% of the information needed to obviate a given decision will be out-performed by the 20^th percentile worst managers with 80% of the information so needed, it goes without saying that successful Portfolio Managers either have to be extremely advanced (or consistently lucky), or simply appropriately informed of the issues surrounding and impacted by the decisions they make. Okay, but with various managers, analysts, and information specialists all vying for the Portfolio Manager’s attention in order to hawk their own preferred information streams (which demand a very specific type of expertise, held only by those pushing them, don’t you know), how is our nominal Portfolio Manager to differentiate between the valid ones, and those that are useless, or even detract from the targeted 80%? I mean, Chief Financial Officers have been known to insist that everything that the Portfolio Manager needs to know can be derived from the general ledger – surely our PortMan should listen to him!

Yeah, and if you believe that, then you may also be consoled by the fact that gas station sushi in Ohio looks a lot like authentic Japanese restaurant sushi from Honolulu.

Those proffered information streams come at you fast

The thing about the types of information that the PortMan needs to make better decisions that her competitors is that, like sushi, it has a very short shelf-life. How to arrive at the best strategy to leverage assets, project performance, and market share to attain the organization’s overarching goals is an elusive solution, which lives in a dynamic, fast-moving environment. As I asserted in last week’s blog, there are precious few usable formulaic approaches, and even among those none are effective across a wide range of industries or situations. But, if I can’t tell my readers what’s the consistently-right answer, I can jolly well warn them of the wrong ones.

One sure tell that the person pushing a certain type of analysis doesn’t have a valid basis is if they ask the question “Why wouldn’t you want to have this information?” Management Information Systems (MISs), like anything else of value, take time to set up, feed, and maintain. If a given MIS’s output does not provide timely, accurate, and relevant information, then all the energy that went into setting it up, collecting its data set, and maintaining it is wasted. Not only that, but the valid system that should have been set up in its stead doesn’t go on-line, and that 80% goal slips further and further away. So, when any of the MIS-pushers invoke this line of persuasion, as if management information just kind of exists out there, and it’s free for the taking, then something is definitely wrong with their grasp of efficacy.

Another sure-fire indicator of an invalid MIS has to do with its structure. All valid MISs have this (very general) structure in common: (1) data is gathered, based on some discipline (2) the data is processed into information, based on some methodology (like Earned Value or Critical Path) (3) the information is delivered to the decision-makers in a format they can readily understand. Conversely, invalid MISs are often structured like a poll. A central data repository is surrounded by input/output nodes, and the “information” is pulled from it as needed. The two problems with polls are that (1) someone always has better or more accurate data, and (2) the “data” they contain are usually hearsay or subjective perceptions, which are inherently unreliable from the get-go. Polls’ data goes bad faster than unrefrigerated, uncooked fish.

We know where we’ve been, can we know where we’re going?

One final sure-fire indicator that a proffered MIS is less usable than small units of seaweed-wrapped bait has to do with whether it’s a feedback or feed-forward system. Feedback systems are based on events that have already occurred, so that their data is verifiably accurate. Feed-forward systems, by contrast, are based on projections of what experts (both real and fake) believe will happen in the future, meaning that the data in these systems is highly subjective, and, therefore, highly unreliable. Unfortunately, this is the very type of system that modern-day risk managers use, as both decision-tree analysis and Monte Carlo simulations involve collecting experts’ opinions of how the future might unfold differently from the projects’ baselines, guess estimating the odds of the alternate outcome, and cost or schedule impact.

And larding over these guesses estimates with statistical jargon does not remedy this foundational error.

Ignoring any of these warning signs is analogous to purchasing fine Japanese cuisine from an establishment with a primary purpose of pumping petrol out of underground tanks for use in automobiles. It’s possible that both the information and sushi are ingestible, but you’d be better off with one of those hot dogs that’s been turning around and around on the steel-cylinder heating unit since the last Bush administration.

[i] Retrieved from https://food52.com/blog/9183-the-history-of-sushi-in-the-u-s on October 21, 2017, 13:27 MDT.

Before I get to my promised (from last week’s blog) workable definition of Portfolio Management, I want to pass along a little story from my younger days. There was a car I would occasionally see in traffic that looked like a 1978 Lincoln Continental Mk. V (a luxury coupe), but with a really weird twist: its suspension was significantly raised with oversized wheels and extended suspension so that the bottom of the chassis was at least four feet off of the surface, by my estimation. It also had a custom paint job, with the words (I swear I am not making this up) “High Class” written in script on its running boards.

Being something of a car aficionado myself, I must say this is a remarkable fail. Unless the Mark V’s original drivetrain was replaced, it likely had an engine that cranked out an anemic 225 horsepower into a three-speed automatic transmission, comically inadequate for the job of off-road racing, rock-climbing, or any of the other uses that raised sports vehicles typically perform. On the other hand, if this autofrankenstein’s owner had in mind that he was enhancing a luxury coupe’s function, I have to believe he was gravely mistaken. Just getting into the thing must have required a small stepladder – hardly the automotive accessory bound to impress the girl you’ve asked to the prom (a quick insight into car guys’ thinking: the acid test for any luxury car is whether or not you would want to show up to your prom date’s daddy’s house in it).

Meanwhile, Back In The Project Management World…

In my previous writings where I point out that Asset Management, Project Management, and Strategic Management are three very different disciplines, with different goals, methods, and associated management information systems supporting them, I have referred to the macroeconomic examples of hostile takeovers, vampire advertising campaigns, etc.to support my assertion. But there is another genre where these differences manifest themselves – tool usage.

Even a cursory review of management information tool misusage confirms my thesis. When Asset Managers (read: accountants) seek to generate, say, the Estimate at Completion costs for a given project, they invariably use either spend variances or a regression analysis of actual costs, both of which return inaccurate projections. No matter: the accountants will insist that this ridiculous methodology is reliable, when the valid source of this information is an Earned Value Management System. Attempting to use the general ledger tool to produce a PM product is the management science equivalent of absurdly raising the suspension of an American luxury coupe, and thinking that some improvement had been achieved.

And it’s not just general ledger overreach

On the other hand we have attempts to use PM tools for resource management, as in those cases where Work Packages are developed for organizational breakdown structure elements. Work Packages, of course, capture pieces of scope, which are then costed and scheduled. The generic test for whether or not an element of work should be captured in a Work Package is the question: what percent complete are you? If the WP manager is in charge of a piece of scope, this is an answerable question. Alternately, if the manager is responsible for a function, or a resource, this is an unanswerable question. And, if this is an unanswerable question, then the work being managed (almost always) should not be managed as a project, and any cost or schedule performance data produced is bound to be error-filled.

Need more evidence? Ask your accountant to pull information from the general ledger about how your organization is performing against other companies’ project teams. Surely this is essential Strategic Management information – and yet, the self-appointed keepers of all relevant management data simply cannot deliver in this arena.

What tool misuse can tell us

All of which leads us to a workable definition of Portfolio Management. Begin with the premise that Asset Management, Project Management, and Strategic Management are fundamentally different. What’s called for in intelligent Portfolio Management is a balance among the three types, an acknowledgement that an inherent conflict of epistemological interests is in play. Imagine yourself as a “portfolio manager” in whatever organization, occupying whatever role-title that entails. Unless your Chief Information Officer oversees a hot mess of information streams that conflict, overlap, or serve no purpose, there’s a decent chance that your primary petitioners will want you to decide in favor of one of three goals:

• Maximize shareholder wealth (Asset Managers)
• Improve project performance (Project Managers)
• Increase market share (Strategic Managers)

… which almost always conflict with each other. That’s why no single software platform can provide THE information stream for the portfolio manager – they’re inherently rooted in one of these widely disparate types, and typically project that particular type’s techniques into arenas where they quickly lose efficacy. Portfolio Management cannot be reduced to a formulaic analysis, no matter how complex those algorithms may be.

With all that having been written, here’s the promised workable definition of portfolio management:

Portfolio Management is the pursuit of balancing the organization’s initiatives among Asset, Project, and Strategic goals in order to attain the organization’s overarching, or consolidated objectives.

Yes, I know I’m probably the only person who thinks this way, but all of the other definitions I’ve seen always seemed to be lacking precision, as if airy or inchoate but sophisticated-sounding management science-babble could serve as a structure for advancing portfolio management maturity. You may as well raise the suspension of an iconic luxury car, and call it “High Class.”

Oh, wait, that’s been done already.

A common misperception of Portfolio Management is that it’s simply what Project Managers do, just on a larger scale or from a higher organizational perch. It’s an easy mistake to make in the PM world. Just as a Control Account Manager is responsible for the Work Packages that make up the Control Account, so Project Managers are responsible for the combined Control Accounts that make up the project. So, naturally, the person in charge of a collection of projects is a “portfolio manager,” right?

Well, not exactly, and I just happen to have a story that illustrates this.

A certain scheduling software company was hosting a “train the trainer” course for their recently-released upgrade, which was being sold as a type of portfolio management platform. At the end of the course, I and two of my colleagues were invited to a lunch meeting with this company’s CEO, and some of his associates. We were seated at a long table, my team sitting directly across from the software company’s execs, and over convention center chicken and mixed vegetables we began discussing the essence of portfolio management.

The upgraded software, in addition to doing a robust job of critical path scheduling and resource-loading (this system’s resource-loading was so good, in fact, that my organization regularly used it to generate the Basis of Estimate), introduced modules that would allow for time card entry and the capture of travel expenses. It was on these two modules that I began my analysis.

To the best of my recollection, I started by saying that their software already did a good job over in the Project Management side of things, but if they wanted to advance into portfolio management, they would be far better off capturing Strategic Management elements, rather than poaching a few Asset Management elements from the general ledger. They asked me to clarify.

I told them that no general ledger system that’s worth the name is going to be lacking a payroll function. By trying to peel off timesheet from the general ledger and into the organization’s scheduling software – no matter how robust that scheduling software may be – they would be presenting their platform as a rival to the general ledger. I continued that there’s no intuitive reason why an organization would remove that capability from their Finance and Accounting department, and hand it over to the Program Management Office.

The CEO replied along the lines that they felt that the newly added-on modules represented a logical target in the expansion of the scheduling software: however, as he was speaking (but outside his peripheral vision) I could see the proverbial light bulbs going off over his execs’ heads. They, at least, were getting what I was saying.

I went on to say that, even if they could make the argument that timesheet entry is better performed through the scheduling software, that they would never make a convincing case that travel belongs there. Travel expenses are part of Accounts Payable. The very best they could hope for would be to provide a 100% seamless interface with the organization’s ledger, but, given the wide variety of accounting packages out there, that that would be virtually impossible.

The CEO repeated some rather bland verbiage about how the organization was confident that they were moving in the right direction, and the conversation turned to more jejune topics.

However, in the next release of the software, I don't remember that the time sheet and travel entry options were noticeably absent. I’m sure it was just coincidence.

As to the question of whether portfolio management is simply program management writ large, consider the instance of the United States Post Office. Some years back, the were running an advertising campaign aimed at getting people to write more letters at a time when the postal rates were fixed at a point where they were actually losing money for every letter posted. So, in a real sense, whomever they put in charge of that advertising campaign, or project, was working at cross purposes with whomever was trying to stop losing money for each letter mailed. I wonder if those two (or their teams) ever met each other and, if so, what they had to say.

To summarize, I wish to overturn two common misperceptions about Portfolio Management: it is not just Project Management on a larger scale, nor is it poaching certain information streams from the general ledger into the PM information software. And those organizations that believe to the contrary are working at an inherent disadvantage.

So, what is a working definition of Portfolio Management? I’m glad you asked. First, a few ground rules…

Look at that! Out of blog space for this week. Tune in next week for a clearly articulated, workable structure for extended PM practices into valid Portfolio Management.

As I discussed in my second book, tournament chess players have memorized the series of moves they use to begin a game. These chess openings have names, and their variations have names as well. Even at the junior high school level it is a rare player indeed who does not have committed to memory at least one opening to use as White, and two openings to use as Black (as a response to 1. P – K4 or 1. P – Q4), from the first five to fifteen moves of the game, with two to seven variations for each. When I was playing in tournaments, my favorites were:

• Giuoco Piano (as White, though I did not care for the Max Lange attack variant)
• Sicilian Defense (Najdorf or Dragon variants, as Black against 1. P – K4)
• Queen’s Indian Defense (as Black against 1. P – Q4)

…and, should I come across an opponent who was unaware of the traps and pitfalls inherent in the variation I was playing, I almost always won.

In a way, the memorization of chess openings is analogous to the advantage PMI® gives its members, especially the certified ones, when it comes to successfully managing a project. Project Managers don’t just arrive on-site, with a vague idea of the desired outcome. They know to capture the scope as accurately and completely as possible, decompose it into a Work Breakdown Structure (WBS), and begin to estimate the resources (Cost Baseline) and time (Schedule Baseline) needed to achieve the scope. It’s a moderately structured way of approaching the management of the project, and those who have a competent understanding of these structures will have a nearly insurmountable advantage over those who do not when it comes to successfully completing projects.

Now, for those of my readers who were not so geeky as to be involved in tournament chess, a quick side note about chess notation: when the moves are recorded, either by the players themselves or for publication and analysis later, three punctuation marks are used. A dash (“ –“) means “to,” as in “P – K4” means “pawn to king 4.” Then come the commentary marks: a question mark indicates an error, and two question marks indicate a blunder. An exclamation mark indicates a good move, and two exclamation marks indicate a brilliant one. These two can be combined, as well. An exclamation mark in front of a question mark (“!?”) means a bold but risky move, and the opposite order (“?!”) indicates a wild, perhaps even reckless move (well, as reckless as chess players can get, I suppose).

At the opposite end of the intellectual and entertainment spectrum, we have comedy. Really good stand-up comedians can be funny on a highly spontaneous basis, with very little pre-planned structure to their stories. Sketch comedy is a bit more structured in that the players are following a script, but even here the more hilarious instances come about when the comedians deviate from the original text. But what of those instances where deviating from the original script is called for, not to make people laugh, but to successfully manage a project? Here’s where the management sciences quickly lose efficacy, since, by definition, circumstances are unfolding in such a way as to preclude the lifting of solutions from other, highly analogous situations. The PM finds herself breaking new ground, often in circumstances where the wrong decision can have significant consequences.

In highly volatile project management environs, I’m often reminded of Dwight Eisenhower’s quote “In preparing for battle I have always found that plans are useless, but planning is indispensable.”[i] More often than we are perhaps ready to admit, the original plan has to be modified (not abandoned – that would mean we’re working off of a “rubber baseline,” a distinct no-no in PM space), occasionally dramatically, sometimes rather quickly. So, how to know when such immediate, dramatic baseline modifications are needed to save the project? I recommend using chess scoring notation!

Think about it: in the Variance Analysis Reports (VARs), the common template indicates the Schedule and Cost Variances in both dollars and percentages, along with the Cost Performance Index and Schedule Performance Index (CPI and SPI). If a variance breaks threshold, there are blocks of text discussing the problems’ causes and impacts, and corrective actions. Simply put the corrective actions into a bulleted list, and leave space for the program managers or customers to insert punctuation, as in:

Along those lines, I would like to arrange to have the following Project Management “Opening” named after me:

• Do the whole WBS thing, yeah yeah yeah, but then…
• Get rid of the Risk guys,
• Relegate the accountants to relaying actual costs at the reporting level of the WBS, and then offer no additional advice,
• Don’t budget more than the bare minimum for the communications and quality specialists,
• But make sure you are fully staffed with the Earned Value, Critical Path, and configuration management support.

With any luck, the “Hatfield Opening” in PM will become more famous than the other associations most often attached to my name, which include two train wrecks that happened 100 years apart, and a little misunderstanding with some people named “McCoy.”

[i] Retrieved from https://www.brainyquote.com/quotes/quotes/d/dwightdei164720.html on October 2, 2017, 12:44 MDT.

Are you tired of various media outlets, politicians, and special interest groups inundating you with attempts to “raise (your) awareness?” I know I am. Since this is my last post for the month of September, my readers will be perhaps interested in knowing that this is the “official month” for the following:

• National Italian Cheese Month
• National Self-Care Awareness Month
• Better Breakfast Month
• Classical Music Month
• Fall Hat Month
• Intergeneration Month
• International Square Dancing Month
• International Update Your Resume Month
• National Blueberry Popsicle Month
• National Chicken Month
• National Courtesy Month
• National Honey Month
• National Mortgage Professional Month
• National Mushroom Month
• National Papaya Month
• National Piano Month
• National Potato Month
• National Preparedness Month
• National Rice Month
• National Sewing Month
• Save Your Photos Month
• Self Improvement Month
• Whole Grains Month[i]

…among others. Now, many of you will no doubt wonder why you need to have your awareness raised about, say, blueberry popsicles, much less spend time in the month of September contemplating, or even celebrating them. But I would like you to imagine a scenario where you are in a project team meeting, and, in the middle of evaluating the cost and schedule performance indices, some consultant suddenly interrupts.

“You’re not taking into account blueberry popsicles.”

You respond “That’s right – I’m not.”

“But there have been several articles on the effects of blueberry popsicles on this kind of work in the trade journals. Haven’t you read them?”

“No, but I have never seen a published study establishing, with verifiable data, that blueberry popsicles – or any popsicles, for that matter – have anything at all to do with the kind of information I need to maximize the chances of bringing this project in on-time, on-budget.”

“Well, that’s only because your awareness of the blueberry popsicle effects hasn’t been raised. I can perform an analysis and write your project’s Blueberry Popsicle Plan…”

“I don’t want such a plan. I can manage without it.”

“No, you can’t, and the fact that you are suggesting as such is evidence of your lack of knowledge of Project Management.”

Sounds absurd, doesn’t it? But it is of a piece with the Information Deficit Model, the notion that, for any given field of study, there are experts who have an abundance of knowledge and expertise, and others who are not as enlightened, but should be. In those cases where we’re talking about passing along experimentally-derived findings that support new theories or overturn existing ones, I’m okay with this model. It’s in those instances where management science swerves into pseudo-experts pushing a poorly-supported (or completely unsupported) hypothesis that this “raising awareness” stuff gets truly irksome. In many of these instances, the Information Deficit Model effect is indistinguishable from arrogant pseudo-intellectuals foisting their shallow musings on the rest of us, and demanding recognition and respect for doing so. For Project Management professionals, I believe that a usable litmus test to differentiate between the two should be if the PM advance being asserted is backed up by something more substantial than expert opinion, or even consensus. If the organization’s key decision-makers need to be made aware of a usable technique, theory, or even hack, then let those on the in-the-know side of the divide show a record of repeatable outcomes, usable experimental or real-life data, or results published in a peer-reviewed journal. To see if such a litmus test is usable, go back and re-read the indented paragraph, but substitute the words “risk management” for “blueberry popsicles,” and consider for yourself if it works.

If it does, the people who consider themselves to be the owners of an abundance of Project Management techniques and want to impart their wisdom to the rest of us should either produce their objective, repeatable findings, or else, to borrow a phrase, check their “in-the-know” privilege.

[i] Retrieved from https://nationaldaycalendar.com/september-monthly-observations/ on September 23, 2017, at 13:13 MDT.