There’s no such thing as … not “an IT project,” although there isn’t. Something else there isn’t is a perfect organizational chart.
You’d think that after centuries of structuring organizations for optimal performance, someone would have figured out a “best-practice” — another thing there isn’t, as it happens, but also not this week’s topic, which is, in an oblique way, something you can do to mitigate your org chart’s limitations once you’ve given up on designing a perfect one. What it is: establishing so-called Centers of Excellence (CoEs) or their grassroots cousins, Communities of Practice (CoPs).
Start with the org chart itself and why there’s no such thing as a perfect one — important because if we don’t understand the problem we’re solving we’ll only solve it by accident. And so …
Org charts are taxonomies. As a general rule, taxonomies partition each level based on the same logic as higher levels, so if the company’s top level is functional — Operations, Finance and Accounting, Human Resources, Information Technology, and so on — it usually makes sense to organize the next level by function as well, so that the CIO breaks IT down into Application Support, IT Operations, IT Business Office …
Then the CEO gets word that some multi-million-dollar project is in trouble, and realizes it isn’t the first, nor is it the second project management disaster in the making. In fact, expecting big projects to crater has become baked into the company culture.
Crater is the default expectation. Late being better than never the CEO recognizes the root cause of project management failure: There’s no “one throat to choke” — no one organization whose head ends up on the chopping block (if you’ll forgive my mixing of executional metaphors).
The topic goes on next the ELT (executive leadership team) meeting agenda, where the CIO recommends the “best practice” solution, namely, to establish a PMO (project management office), to be given responsibility and authority for defining, establishing, and overseeing project management best practices throughout the enterprise … and here’s the punchline … crossing all organizational boundaries.
The ELT, that is, has just recognized the fundamental limitations of taxonomy design: No matter how you break things down from trees to branches, to leaves, to veins in the leaves, some responsibilities will never fit neatly into just one box.
Which is too bad, because whether it’s something seemingly simple like adding a PMO or some other box, or something more complicated like a full-bore organizational restructuring, when you’re done drawing boxes and connecting them with lines, some responsibilities just don’t fit.
The point of reorganizations is to fix problems by removing barriers, but every barrier removed is really a barrier that’s just moved.
Reorganizations, that is, fix what’s broken by breaking what’s fixed.
Which is why, a year after our mythical ELT establishes the PMO and commissions an independent review of its outcomes, they discover that, when compared to pre-PMO times:
- Projects fail at the same rate.
- Actual project management practices are still scattered and inconsistent.
- Project managers face twice the administrative burden … they have, that is, twice the paperwork.
- Everyone involved complains that the PMO has turned into little more than a stifling, choking bureaucracy whose main purpose seems to be establishing a bunch of flaming hoops everyone else has to jump through.
It isn’t the PMO’s fault, either. It was chartered to prevent failure, not to foster success. Of course it became a bureaucracy. Try to prevent failure and you’ll do what it did: List every reason projects fail, and establish reviews and checklists to make sure none of those reasons happen here.
Sadly, what they miss is that, taken together, their reviews and checklists become new causes of project failure.
Which is why CoEs and CoPs often work better than attempts to solve problems through more oversight groups.
CoEs’ and CoPs’ purpose is to foster techniques that work well, and to find ways to improve them further, in domains that cross organizational boundaries.
Crucially, they’re chartered to have little (CoE) or no (CoP) authority, so they have to achieve their goals through persuasion and influence — they can’t compel compliance, only encourage participation.
The difference between them: CoEs are official bodies. A project management Center of Excellence has someone or a small team, part of whose official responsibility is to improve project management practices throughout the enterprise.
A CoP does not. It’s a grassroots organization composed of enthusiasts who participate because they want to.
And gee … what a shame. That’s all the time we have this week. Next week we’ll pick up where we’re leaving off.