Geeks With Blogs
Ulterior Motive Lounge UML Comics and more from Martin L. Shoemaker (The UML Guy),
Offering UML Instruction and Consulting for your projects and teams.

A business classic tells us that Quality Is Free. The title is intentionally provocative: no, quality isn't free, it just pays for itself. But first, you have to pay for it.

And that, unfortunately, is where we fail in the quality game so often. Corporations seem addicted to the practice of compartmentalized budgeting, or what I think of as "bucket budgeting": you've got a bunch of different buckets you pour money into at the start of the fiscal period; and each bucket can only be spent on a particular purpose. When that bucket is empty, you can't accomplish that purpose any more. Oh, the company may have some reserve you can tap; but you're going to get frowned at for exhausting your budget.

I understand bucket budgeting as a planning tool. I think it makes perfect sense, for the same reason you should make estimates out of small elements. In fact, it should the exact same reason, because these budgets should be estimates, not shackles. You'll correct over time.

And that's where the failure comes in. Somehow, some way, in too many organizations, those buckets are shackles. Your  bucket is what you can spend, what you will spend -- regardless of the bottom-line impact of your spending. Even if every $1 spent out of your bucket brings $1.20 into the company, you only get to spend what's in your bucket. This isn't a new complaint, of course; and smart managers certainly keep an eye out for ways that spending money can save or generate more than they spend. But less bold managers don't like to rock boats. They live within their buckets, because overspending their bucket gets them a bad review. It takes courage to stand up and make a case for more money in your bucket, unless you have a very clear, simple chain between the money you spend and the money that comes back in.

And the quality equation is particularly susceptible to bucket budget shackles. Quality does pay for itself, but it seldom shows up anywhere near the buckets where the costs came from. The cost of quality is measured in extra labor and time up front on preventing defects, along with extra labor and on the back end detecting and correcting defects. It's also training time, which takes time away from "productive" work. It's also management time and communications effort in getting everyone -- execs, workers, and customers -- to understand that seemingly slower time is actually faster in the long run.

The benefits of quality, meanwhile, are in reduced support costs, reduced rework costs, and increased customer satisfaction and loyalty. These do affect the bottom line; but they don't put money in the buckets of the managers who have to pay for the quality in the first place.

A common sarcastic reaction I here among the workforce is "Quality is free, so management thinks they can have it without paying for it."  And sadly, this reaction is often justified. But I don't think most managers are really that clueless. I do think many managers are shackled by bucket budgeting, and unwilling to buck the system for something that won't have an effect on their budgets. The effect may be the same as cluelessness; but if we don't understand the proper cause, we can't devise the proper correction.

And no, I don't know what that correction is. I mean, to me, the answer is simple: start treating budgets as estimates, not shackles; but I don't think little old me is going to change corporate culture that drastically just by saying so.

Final note: this isn't inspired by anything at my current client. Really, it's not: I've been complaining about bucket budgeting for over a decade. But it's true that my client is currently entering a phase where they're trying to invest in quality in pursuit of benefits in the long run, and I want to do my part for that. There are forces that will push against that effort, and forces that will push for it. I'm doing a little writing to help me clarify how I can help push in the right direction.

Posted on Friday, November 28, 2008 10:54 PM Development Processes , Personal | Back to top

Comments on this post: Quality is NOT Free

# re: Quality is NOT Free
Requesting Gravatar...
Having worked in Civil Service most of my life, I certainly don't know what to offer in the way of practical guidance. One thing that we had that is close to your bucket budgeting is "overtime".

We used to marvel at managers who doled out overtime like it was coming out of their pockets. Even though we knew that they were accountable for the amount of overtime in their departments, we still wondered why they demanded 10 hours of work from 8 hours of duty time.

The only way I ever found to get around this tight-fisted mentality was to present a "proof-of-concept" to justify the expense.

Justification is as close as I can come to your idea of the cost of quality. Once we were able to prove that we could deliver desirable results with a bit of "overtime", the managers usually kicked the bucket. :)



Left by Mitchell Allen on Nov 29, 2008 12:01 AM

# re: Quality is NOT Free
Requesting Gravatar...

I think justification is the key, in civil service or academia or the corporate world. I honestly don't think that accounting departments are stupid or malicious; but to go beyond the budget requires justification.

Talk show host Bruce Williams used to talk about easy money vs. hard money. Easy money is what they expect to pay. Hard money is what you have to convince them to pay. He used this term in negotiation and sales, but I think it also applies in management and budgeting. If you're a manager and you have money already allocated, that's REALLY easy. Now when you need more, that's a LOT harder. You had to do a little annual projection to get the easy money; now you have to do real work to get the hard money. You have to expend credibility to get it. Many won't take that chance, ESPECIALLY if the benefits will show up in somebody ELSE's budget.

There needs to be a split: plan by buckets, but assess by the bottom line. EVERY manager should be judged on how the WHOLE company does. But I don't see an easy way to do that.
Left by Martin L. Shoemaker on Nov 29, 2008 12:15 AM

# re: Quality is NOT Free
Requesting Gravatar...

I assert that the opposite measurement needs to be taken: the WHOLE company needs to be judged on how its managers perform.

Sticking to what I know, the Police Department made extensive use of computer statistics (COMPSTAT) to evaluate manpower efficiency. At the district level, the District Commander used COMPSTAT to put more officers in areas that were more crime-ridden. However, the DC certainly had no authority to pull officers from "quiet" districts. (Reassignments, naturally, are a high-level administrative function.)

If we consider this evaluation from the perspective of the bottom line, I think it would be unfair to judge a District Commander by the overall performance of the Department.

On the other hand, if there are Department-wide directives whose implementation is mandated at the district level, we can judge the success of the WHOLE Department by how well EACH of the District Commanders comply with those directives.



Left by Mitchell Allen on Nov 29, 2008 12:38 AM

# re: Quality is NOT Free
Requesting Gravatar...
Well, Mitch, I think that "bottom line" differs for different organizations. It really means "meeting objectives". For a Police Department -- which after al shouldn't be a profit-making venture -- the bottom line is measured in crime stats.

As for unfairness: I think a profit-driven manager would scream "unfair" as well if her performance review depended on the screw-ups of the manager in the office next door. That's why I don't see an easy answer. But bucket thinking isn't an answer. In some sense, it's Tragedy of the Commons.
Left by Martin L. Shoemaker on Nov 29, 2008 9:22 AM

Your comment:
 (will show your gravatar)

Copyright © Martin L. Shoemaker | Powered by: