Inside the Biz with Jill Dyche

The Quality Gap: Why Being On-Time Isn’t Enough

In which Jill advocates behavior changes, starting at the top. But not for you. No, you’re fine. Just for other people.

The biggest problem in business today is that everything is date-driven and not quality-driven.

There, I’ve said it. And a few dozen of my past and current clients are now sidling up to their laptops to shoot me an e-mail asking me if this blog is about them. (The answer is Yes It Is And You Know Who You Are.) Seriously, this is a problem everywhere, yet despite the wholesale crises it precipitates it’s actually getting worse.The root cause of this problem is the old bugaboo of perception. “Progress” is usually measured by speed-of-delivery, not fitness-for-purpose or conformance to requirements or streamlined processes or any of those other quality maxims. We’ve all heard a variation of the following:

“Just [complete the work] so I can get it into my status report for this [week/month/year].”

So that project manager’s boss is now satisfied that the team is getting things done instead of abusing flex time or work-from-home policies. Meantime the campaign went to a saturated segment. Product prices no longer match across divisions. Account managers are confused about territory assignments and are cannibalizing sales. And the Feds have just left a message for your CFO.

You have to name it to claim it. (I learned this from watching The Biggest Loser the other night.) So herewith, the five main reasons for this phenomenon:

1: The measurement conversation isn’t baked into project initiation. In our BI, MDM, and data governance projects we make sure that this is part of the requirements phase. But practitioners aren’t the only people who need to have this conversation. Business executives and managers should be proactive about their quality criteria during ideation or (at the latest) in the business case. This not only elucidates delivery steps, it makes scoping so much easier.

2: The “effort delusion”, that anachronistic WASP-y assumption that many people working hard will yield positive results. There’s a monkeys-on-an-island analogy here that I’ll refrain from making. But as many before me have aptly observed working hard simply isn’t enough.

3: No one closes the loop. Imagine how many companies have invested in quality programs, data quality automation, business analysis skills, TQM and SixSigma training and other improvements yet continue to fail to reconcile the project’s original objectives from its delivered outcome. Instead, projects endure scope creep or are delivered as a shadow of their original vision. Closing the loop between original vision and ultimate deliverable is a learned and practiced behavior. Instead, mediocre projects drive a flurry of fix-and-maintain activities that would have been unnecessary had they been delivered right the first time, and ultimately far more costly.

4: Failure to define realistic delivery increments. This is Project Management 101. Or is it? The problem here is that the people doing the scoping are often not those on the hook to deliver the goods. I’ve seen project managers idling in the doorways asking programmers to rattle off their tasks, then randomly assigning time-to-completion. It doesn’t work that way. Or it shouldn’t.

5: The economic climate has made people paranoid. There, I’ve said it. And a few dozen of my past and current clients…oh, nevermind. You’ve seen it yourself. People go into delivery hyperdrive and start producing at all costs. (“Just load the data into the database—we’ll worry about whether it’s usable later.”) Worse, they cover their collective asses while spinning stories of their productivity, backing into post-facto project plans and pointing fingers when people ask questions.

Maybe if we understand the root causes, we can fix the problem. (Thanks for that one too, Bob and Jillian!)  Or maybe we’ll just stay on the couch and keep chomping away, occasionally groping for the remote control so we can get something different just by pressing a button.

Could Data Governance Help the War on Terror?

In which Jill wags her finger at the silos—ours, theirs, and yours.

“Our intelligence community failed to connect those dots, which would have placed the suspect on the no-fly list."
-- President Barack Obama

President Obama addressed the Christmas day terror attempt in his White House briefing this afternoon. He acknowledged that the U.S. government had botched terrorist tracking and that the system had failed. He admitted that each agency took responsibility for its own stuff. He also confessed that there was a lack of information sharing, that various government agencies had failed to pull it together. The subtext was that current practices and the best of intentions aren’t enough when it comes to the war on terror.

Obama sounds like a lot of corporate executives I’ve met. They admit that the individual departments are working hard and may be meeting their individual goals. But then something stupid happens. The company loses its largest account or a competitor beats them to the punch on a new product. The hard fact emerges that the company could have used its information better. And that siloed organizations beget siloed data.

The alleged terrorist, Umar Farouk Abdul Mutallab, had been in Nigeria, Yemen, the United Kingdom, and The Netherlands before flying to the U.S. Each of these governments had a slice of information about him. The UK had enough information to put him on a watch list. But the information was never integrated into a rich enough profile of Abdul Mutallab for the FBI to mark him as a person of interest. As Informatica marketing vice president Chris Boorman wrote in his excellent blog post, this would have been an apt use of identity resolution software. Various government agencies could have matched partial information to form a holistic view of Abdul Mutallab, one that could have been shared, thus pinpointing him as a suspect. With the right collaboration, this could happen at both federal and international levels.

There are over a million names on the U.S. terrorist watch list, and in fact the intelligence community had information that an attack on the U.S. might be imminent. White House Press Secretary Robert Gibbs insisted that since apprehending Abdul Mutallab, they had gleaned “usable, actionable intelligence” from him. Given their track record, you can’t help but wonder how American intelligence agencies will use this information. 

“It is increasingly clear that intelligence was not fully analyzed or fully leveraged,” Obama conceded. “That’s not acceptable, and I will not tolerate it.” As much as you believe him, you’ve got to wonder what he intends to do. Sure, he inherited a bunch of crap, but it’s crap we hired, er, elected him to fix.

The conversation in the press has been the same conversation that occurs in corporate conference rooms across the world after a major screw-up: “Who did the oversight?” and “Who should get fired?” “Growing questions about if and when heads will roll!” trumpeted CNN’s Wolf Blitzer. In a crisis, attention seems to shift to the people issues.

Instead, the government should be addressing process issues. Indeed, the real conversation should be how to move forward. These questions should be asked now: “How should we bring identifying data together? What are the key sources? How should integration, access, and usage policies be formulated? What would a sustainable process look like?” Those questions aren’t “who” questions, they’re “how” questions, and they should be front-and-center in the national security conversation.

Maybe they should also be part of your next conversation with executives or business users? The Christmas day terror attempt is one of many, many examples that the opportunity cost of not having data governance could be high, and the consequences nothing short of catastrophic.