17 June 2018

The Wrong Kind of Accountability

"Accountability" is one of those words that gets tossed about a lot, and it’s considered to be an essential part of the values of a high-performing organisation. However, I sometimes feel that we get it badly wrong, and instead of being a force for good, it is a force for bad, and one of the reasons why some of our best intended projects end up as a highly successful resurfacing and dualisation project for the Highway to Hell.

AC/Countability Live

We set up our projects with Senior Responsible Owners, Project Managers, Project Teams, and clear objectives (we think), yet somehow everything feels as if we’re set up to fail. How can this be when we have clear governance structures - Jimmy reports to Jenny who reports to the Project Board which reports to the Strategic Board which reports to Jethro the SRO and all should be tickety-boo?

A clue might be in how we think about Accountability. I’ve heard it expressed in terms of “I want to have one neck to choke” or “the buck stops here”. In our language around Accountability we seem to be anticipating that everything will go tits-up, and we want to be able to apportion blame ahead of time. You hear terms like “heads must roll” and “somebody senior has to get the sack”, especially from the media, and the notion is that if only we had better people in place, the project would have succeeded.

In practice, what happens when projects fail, it’s some underling supposedly reporting to someone further up the chain who gets tagged with the blame, and the whole thing is fixed when they’re fired, because project failure represents a moral failing on the part of the poor sap left holding the can. (That’s not to say that “failure” is always a bad thing - here's a take by Eoin McFadden on the RIGHT Kind of Failure - a very necessary thing, and positive if we're prepared to learn).

Anyway, the wise-after-the-fact cynics (and Northern Ireland is full of these arseholes, but we’re probably not unique) almost seem to want projects to fail. They have no solutions themselves, and the proposals that they do tend to offer are nearly always completely stupid. Not always always, but that’s yet another story for yet another day.

A large part of the problem, I will argue is the Wrong Kind of Accountability (WKA), which exists in the entire governance model, where Project Failure is almost anticipated, and Project Success is almost seen as a million-to-one shot, but it might just work.

Can do we change WKA into Real Accountability (RA)? How do we anticipate Project Success, and make that everyone’s expectation? How, in the event of failure or delay of certain aspects of the Project do we swiftly analyse the problems and take remedial action in order to get things back on track? How, if the Project itself turns out to be a mess (and this does happen from time to time), do we take the learning from that experience and use it to “fail better” next time around?

Well, I’ll tell you one way we’re not going to learn or analyse, and that’s by finding convenient places for bucks to stop or throats to choke. The Wrong Kind of Accountability model is toxic, and leads our organisations and projects ever-further along the Highway to the Wrong Kind of Hell.

Bugs in the structure

So where’s the bug in this wonderful governance structure? Surely everything should work if we have a nice clean and clear Reports-To hierarchical structure? You know those charts that you see? This one’s from MNB Architects (and I’m sure it was a success), and is fairly typical, and there’s nothing intrinsically wrong with it:
Project organisation structure - example from MNB Architects

All looks pretty ship-shape and standard, and similar models are widely used in the Health Service. But you see those lines that connect the various boxes? They’re “Reports-To” lines, and while there are no arrows on them, there is an implied hierarchical directionality to them the builds in this top-down Wrong Kind of Accountability. The Kitchen Consultant is Accountable to the Project Consultants - that’s the clear directionality - it’s one-way, one start/end point, one lane, and one neck to choke. The point of failure is implicitly located at the nodes (the boxes) and not the lines between them.

Another problem with this structure is that the top is isolated from the bottom - and this is where the real problems hit the Health Service. The only way the upper reaches of the chart get to know about what’s going on “below” is through periodic review meetings. These go under a variety of names - Strategic Board Meetings, Project Board, Assurance and Compliance Review - all that rubbish. So the people who are “ultimately accountable” typically get high-level and sanitised updates of how things are going from those lower in the chain. The periodic nature of such meetings means that those doing the reporting typically cram stuff in the night before (which is often when they get the minutes from the previous meeting, which may have been three months ago), and try to cook up excuses for why their particular workstream is falling behind schedule or not delivering. They also tend to identify where to place the blame - sorry, Accountability - for areas where things have gone wrong.

Part of the reason why this happens is that those higher in the pecking order aren’t looking to understand what’s actually going on - they are looking for that most vacuous token in the whole damned system - Assurance.

And with Assurance, we have hit the nub of the problem - the real villain in the WKA saga. Assurance tells you everything is OK and gives the false impression that you’re on top of things when in fact you haven’t got the slightest clue what is really going on. Assurance increases in potency as we climb the hierarchy - Assurance built on Assurance built on Assurance like a lovely house of cards.

Politicians love Assurance. Senior executives and consultants love Assurance. Often doctors, patients, the media, the public love Assurance. Blessed F*cking Assurance. Everything is OK, AND you have a neck to choke waaaaay down the line when you realise that things haven’t gone to plan.
But this attitude does not lead to Project Success. It builds in a failure mentality, which I see as one of the key reasons why good projects fail or go over budget. Much as I would love to hang the failure of the Northern Ireland Renewable Heat Incentive Project at the “Accountable” (ha!) feet of the former First Minister or the senior civil servants involved, and much as we can identify specific failings, the reason that particular corpse kept stumbling along was that everyone was wallowing in Assurance that things would be fine, and they all had a neck to choke if there was to be a problem.

Maybe we need to be more charitable and realise that complex projects built on Assurance and directed acyclic governance graphs are inherently vulnerable to this sort of malfunction.
Assurance behaves as a firewall, preventing those higher in the governance structure seeing what happens at lower levels. It turns everything below them into a “black box”, where they don’t know the workings, but they assume that for specific defined inputs, they get specific defined outputs.

So to quickly recap, here is the problem:
Project governance structures based on Assurance/Reports-To models instantiate the Wrong Kind of Accountability and contain specific vulnerabilities that leave them open to failure, with everyone looking round for the One Neck to Choke.

Embedding Real Accountability

Having identified Assurance as the corrosive element in our structures, how do we get past this? How do we fix the failure-prone command-from-the-top, assurance-from-below model to give our projects (assuming they are well-conceived - again, that’s another story) the very best chance of success, and make Accountability a mechanism for that success, and distributing praise and reward where it is due?

I’m going to suggest that we need to up our game. Each node (box) in the governance diagram needs to be able to reach beyond the node below to have some idea what is happening at the next level, beyond the firewall, so when people have their project meetings, they’re not just getting Assurance, they are understanding what is going on. They can’t do this through the people who report to them - they need to be involved directly, at least to some degree, with that next level.
In the case of the Senior Responsible Owner of a project - the person at the very top of the structure - she/he needs to be engaged and visible on the ground to those who are actually working at the coal face. This involves a lot of work and commitment. And that person - that SRO - has to embrace and accept that Real Accountability. The same is required for Project Sponsors and other senior people involved.

If you think about it, this is the hallmark of successful projects, as well as those unsuccessful, but well-meaning projects where useful learning has been acquired. Leaders are found at every level of the hierarchy - they are engaged, they want the project to work, and they are prepared to roll up their sleeves and understand things at each level of the hierarchy.
They don’t regard the Project as a series of waterfalls that coincide with project meetings - they see it as a journey where each step is important, rather than jumps between major milestones. They’ll help individual people achieve their goals. They’re not “too big to fail”.

Getting stuff done

Perhaps the best example is the Apollo Program of the 1960s and 70s. This was a ridiculously ambitious undertaking in may ways - mind-bogglingly complicated, involving the rapid (and admittedly very well funded) development of new technologies and approaches. But one of the keys to success was that the people at each level of the programme had an understanding of what was happening at the next level up and the next level(s) down, as well as the ultimate over-arching goal of the entire project.

The NASA cleaner probably never said “I’m putting a man on the Moon” to President Johnson when asked “And what do you do here?”, but you can be sure that he/she was very aware that this was what the whole thing was about. Armstrong, Aldrin and Collins knew in high detail about the minutiae of orbital mechanics and the engineering of their spacecraft - they didn’t just accept the Assurance of their engineers. As the Apollo 1 disaster underlined, their lives were on the line. It wasn’t good enough for them to have one neck to choke.

And the same applies to our health service. We can’t operate in anticipation of failure, or an Assurance model based on the Wrong Kind of Accountability. We need to map Real Accountability onto coal-face engagement, relentless pursuit and criticism of our data, desire to succeed, spreading the word, multi-level leadership, constant, iterative progress, speaking the truth to power (the Right Kind of Assurance), and getting stuff done.

The President needs to, from time to time, be prepared to mop the floors.

Let’s build anticipated success into our models, and deliver Real Accountability.



[Your comments are most welcome - maybe I've got this completely wrong!]

1 comment:

  1. Sounds like many projects have too many levels or tiers in the responsibility/roles tree. Also I see a fundamental problem with anyone in the project management team not actually having a 'skin in the game' or true but in. If not you are not motivated in quickly fixing things & maintaining the momentum it's too easy to throw in the towel; again if it doesn't impact your own work day to day. If you don't have motivation by seeing the problem first or second hand what real incentive do you have in fixing it? Will you be sacked if you don't fix it-no! Will you lose financially (bonuses and share options as per private sector)-no. So motivation and having a vested interest beyond just 'managing it' is critical. This is a major factor in why things fail; lack of invested accountability.

    ReplyDelete