Wednesday, July 25, 2012

The Decline and Fall of Some IT Groups

We are information systems consultants. As part of our typical engagement we work with our clients' internal IT groups. For this reason, our clients' end users consider us to be experts on their organization's IT groups. Recently a depressingly large number of our clients' end users have come to us to ask why their internal IT groups are doing such a bad job. They ask us this question because it is not an easy question to ask of a colleague; it is easier to ask an outsider.

This barrage of questions has forced us to consider this question, although we usually try to stay out of internal IT politics as much as possible.

When we took a step back to consider the different IT groups in different organizations, a pattern began to emerge. We are seeing an alarming number of them in the following situation:

  1. They are being asked to "do more with less"
  2. Positions are being filled only after long delays if at all
  3. Functions are being taken in-house to save money
The first point is a function of the recession: every one is trimming their budgets. Pity the poor IT group: unless it has technically savvy senior management, the right things are not being cut. We see this directly all the depressing time, budget cutting by size of line item and not by return on investment or the marginal utility of the line item.

The second point is a function of the recession as well: there are so many applications for any given job that managers are tempted to wait for the PERFECT candidate. Here is an excellent column from James Surowiecki's reliably interesting and useful column in the New Yorker.

The third point is a function of muddled managing: if you have cut your IT group's budget, and failed to hire new people, then adding work to their plate is very unlikely to produce good results. The most common outcome we see is a decline in service to the users and a decline in morale within the IT group.

So if your formerly-good IT group is in steep decline, you might want to lobby for their budget to be increased or their headcount to go up. But if that is not an option, then perhaps you can temper your irritation at the decline in service with pity for the IT group itself: it can't be any fun to be on an ever-sinking ship.

Wednesday, July 18, 2012

Managing Consultants

I am currently paid on a consulting basis.

I put it this way, rather than "I am a consultant" because I do not have any great attachment to the business rules under which I get paid. I want to do good work for reasonable people and in return I want to be paid a reasonable amount.

For many reasons, many of them both tedious and beyond my control, consultants are being demonized; we are the new lawyers: a professional group we all love to hate and can agree to mock. For example, consider this poster from the often-hilarious

While I agree that lawyers are often used ineffectively, I do not agree that lawyers are useless busybodies, just as I don't think that HR exists to making hiring difficult or facilities management exists to stop you from having nice office furniture.

(I blush to admit that I was going to use "first thing we do, let's kill all the lawyers" until I looked into it and found that I had that backwards: Sigh. Another blow to my feelings of omniscience.)

Leaving aside the inherent problems with demonizing any group, I have a problem with demonizing consultants. To me, the big difference between generalizing about lawyers and generalizing about consultants is that lawyers are a fairly well-defined group providing a fairly well-defined service, while consultants are a ridiculously varied group with only a compensation methodology in common.

Mercifully, to make my point today I can leave aside the question of whether or not consultants are evil, because it does not really matter: at this point, almost no medium-sized or large companies can do without them.

(In my experience, small firms benefit most from consultants, but small firms seem better adjusted about this issue: happier and more accepting.)

So pick a side: consultants are a useful tool because you can rent expertise that you need for only as long as you need it, or consultants are parasites whose only goal is to addict you to their services. Either way, there they are.

I assert that in this day and age, managing consultants is a key skill, just as managing women was forty years ago, when women started showing up outside the steno pool.

I accept that consultants are not a panacea: institutional memory is good and employee loyalty is good.

I agree that consulting engagements need to be reviewed periodically to ensure that renting expertise is still the best way to accomplish whatever goal toward which your consultants are working.

What I cannot abide is the current mindless swing toward "No Consultants!" as a policy. This is so rarely practical in today's business environment. I say this not because such policies cut into my paycheck: hilariously, they don't cut into my paycheck, at least not so far. What they do is drive up my paycheck as I stand aside and then wait for the frantic emergency calls that always come--unless I hear, instead, the sickly silence of despair as organizations caught in the death spiral of "doing more with less" until they go out of business or fail.

Love them, hate them, but make sensible use of consultants. Don't ban them and don't cling to them forever. Instead, assess your needs, consider both cost and benefit and do what makes sense. That is more work than issuing clear, simple edicts such as "no consultants!" but that is why managers get paid the big bucks, isn't it?

Wednesday, July 11, 2012

In Praise of a Touch of Humility

I am not a humble person. I don't think that I ever have been. So at first blush it may seem strange that I would be praising humility. In fact, I claim that this is a natural situation: I pride myself on being effective and accurate. In order to be effective and accurate, I need good information and feedback. In order to get good information and feedback, I need to seek it, which requires embracing the fact that I make mistakes and therefore trying to avoid future mistakes and fix past mistakes.

I hasten to point out that I do not think that I suffer from the over-confidence that infects so many leaders in so many workplaces. (I have already ranted about the childish pretence of endless certainty in a previous post but this is a new, slightly different rant.)

As a practical matter, this means that I approach software development in a specific way: I do things more than once. I write "watch dog" programs to monitor processes. I write and use consistency checks. On the plus side, many of my mistakes are caught. On the minus side, many of my mistakes are exposed to public view. However, I sleep better at night knowing that my confidence in my work is based on more than my high self-regard.

As a theoretical matter, consider this interesting Harvard Business Review blog entry: Less-Confident People Are More Successful by Tomas Chamorro-Premuzic. His thesis is this:

There is no bigger cliché in business psychology than the idea that high self-confidence is key to career success. It is time to debunk this myth. In fact, low self-confidence is more likely to make you successful.
After many years of researching and consulting on talent, I've come to the conclusion that self-confidence is only helpful when it's low. Sure, extremely low confidence is not helpful: it inhibits performance by inducing fear, worry, and stress, which may drive people to give up sooner or later. But just-low-enough confidence can help you recalibrate your goals so they are (a) more realistic and (b) attainable. Is that really a problem? Not everyone can be CEO of Coca Cola or the next Steve Jobs.
As a side note,  I point out that a friend of mine offered a gloss on this article for me: he would say that we are talking about humility here, not clinically low self-esteem. I would define humility as the trait of being willing to consider negative feedback; I suspect that there are many working definitions. The dictionary definition I found at is this:

 modest opinion or estimate of one's own importance, rank, etc.

I would say "unexaggerated" is a key component of this concept, but perhaps that is only my personal sense.

Regardless of the fine print on the definition of the opposite of over-confidence, I fervently hope Chamorro-Premuzic is right in his assertion that not-overly-confident people are more successful. In my bitter experience, boundless and baseless confidence are richly rewarded, without regard to the outcomes of those manly, clear-cut and confident decisions when complex, nuanced, multi-stage decisions would seem to be required. (With lots of checking to make sure that the path along which we are all running is the right path.)

It would be fabulous to put the Era of Empty Assertion behind us and move on to a more results-based, fact-based, reality-based, merit-based workplace, at least in IT. May it be so.

Wednesday, July 4, 2012

Broke, Broker, Brokest

In response to my last post an old friend who is an organizational psychologist pointed out something that has been on my mind for a while: the tendency of many people to live the dictum "if it ain't broke, don't fix it."

(By the way, the history of this phrase really surprised me: check it out on Wikipedia if you are so inclined.)

I can see why this dictum is useful in large organizations: endless thrashing and reorganization is often the bane of their existence and they need some way to make sure that "fixes" are actually needed. (They also need to follow up those fixes to make sure that they worked and continue to work: see an earlier post.)

But I am seeing many instances of throwing out the baby with the bathwater in this regard. I see simple-minded adherence to this principle which eliminates all of the constant small decisions and leaves only the rare big decisions.

In a nutshell, there is more to business decision evaluation than "broken" vs "not broken". Very few processes either work unacceptably badly or perfectly. Surely we can do better than just these two gradations?

The determination of whether or not an issue or set of issues is worth addressing involves some possibly subtle cost versus benefit analyses. I assert that the deciding test of "Is this process hopelessly broken?" is simply not good enough to produce an effective or efficient organization.

I hear from managers that there is an additional complication: their organizations have tightened the purse strings so much that the level of effort required to push through a small investment seems too great: better to save your strength and only push through large projects. Since the projects are large, the scrutiny is great, so the projects have to be "no-brainers" so better to wait until your department is on the verge of collapse and then propose a giant, sweeping solution.  I do not have a good answer to this logic: if the premises are true, if the organization is that badly out of balance with respect to its decision-making, then perhaps only doing the big, dire things is the way to go.

And yet.... Especially in these recessionary times, with staffing levels so far below idea, don't we need to be as efficient as possible? Mindlessly trying to avoid expenditure while assuming that fixed labor costs will produce greater output under the magic incantation "do more with less!" will produce nothing but burnout and turnover.

So the next time someone suggests incremental refinement and modest investment, try to imagine a future that gets slowly better and better with modest cost and reasonable return.