Skin in the game and management consulting or “we rarely see the outcome of our work”

On ethics of management consulting — how performance measurement clarifies assumptions —

A finance department got re-organised a few years ago by management consultants one of whom honestly admitted — “we rarely see the outcome of our work” (I technically am a management consultant as well, albeit in cyber-security). The consultant who uttered that phrase was stating a simple and perhaps an unfortunate truth — many management consulting engagements are short — some 8–12 weeks whereas the lasting and true effects of an organisational transformation are often realised only much later (see Hawthorne effect, for example).

This is problematic, because:

  1. the consultants don’t have skin in the game — will be paid regardless of the true, long term outcomes; and
  2. consultants are denied the opportunity to learn about the true impact of their advice and to correct their advice when new evidence or circumstances arise.

The first problem is the eponymous “skin in the game” — it is not ethical to give advice or intervene if you do not experience negative consequences (downside) of your action, Nassim Nicholas Taleb has written much about it. In the example, the management consultants might have given poor advice (but hopefully good!) and still got paid. Similar concern is often raised with executive pay, where a short-lived CEO relieves high remuneration when the true outcomes of his or her decisions, for example to acquire an organisation, will only be felt some years later. This then is solved by deferring the CEO’s pay or stock options conditional on company’s performance — often stock. Now, neither the hypothetical CEO nor the hypothetical management consultants control all the variables — perhaps a new manager reverses much of their good suggestions and as a result performance declines which makes it hard to install performance pay. I would argue that for many CEOs the long term vesting stock options are just the cherry on top of their base pay whereas for consultants a performance pay of, say, 25–50% would be a serious risk, one they would compensate by charging even more.

But let’s assume that a client and a group of management consultants agree on performance pay to transform an IT department and start looking at how does a good IT department looks like and how can we measure it to we can collect on our performance pay:

  • Is the job of IT department to maximize system availability, if so — then refusing potentially disrupting changes is one way to game it.
  • What about solving user problems? if we measure total number of tickets then not creating new ones is a way to game it (or opposite — open a ticket for every tiny question if more is better).
  • What about user satisfaction? Users might want shinier laptops when current will do and complain.
  • How about compliance to corporate IT and IT security standards? That sounds really good, but what if standards are too onerous?
  • OK, what about — time to implement new applications? if that’s a priority then using off the shelf software, prefrably cloud will certainly reduce time even when it is risky or strategically unwise (e.g. lock-in).

In short, all these potential performance measurements seem to have drawbacks; perhaps then we could use some weighted model like Russel Thomas (tm) scoring system? Certainly it’s an option, but the key point is that by trying to earn some money, we are forced to to define in terms of measurable outcomes how does a good business unit look like. This is half of the equation, what if there are external circumstances that we struggle to control, for example, pay bands restrict hiring good talent? This is another great find — if our assumption about filling empty boxes is unrealistic in the specified time frame or at all, then why are we making these empty boxes?

So back to our original problem — how can consultants have a bit more skin in the game?

In one sense, consultants live from one successful project to another and so a failed project could probably ruin a revenue stream for a while. On the other hand, clients could keep consultants around for a tad longer (I know it can be expensive), but, say 2 months down the road the consultant and the client bond on a personal level, become perhaps friends. Then hurting a friend by being inconsiderate is a less likely outcome and consultants will weigh their decisions a little bit more. On the other hand, perhaps a consultant will not raise an issue when the friend is at fault.

But on the other hand, I hope that we can get more towards performance pay and by defining the measurements, everyone will become more closer aligned to goal and also spot what is and what isn’t important.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.