Contact details for our Australian point man (and the webmaster for our main site) are on the IAIDQ website (www.iaidq.org) if you’d be interested in talking to him.

I actually co-chaired a conference in Sydney back in 2005. The wife still hasn’t forgiven me for buggering off to Oz for a week and leaving her in Ireland.

My concern with these committees and reviews is that the new Secure Flight system will never be able to pass no matter what they do. For example – “what are the chances someone could hack the customer list?” “About 1 in a billion – if they had a set of supercomputers and the best hacking team on the planet” “oh, so the system can be hacked? Declined”. Never mind that the existing systems are unsafe or don’t work!

I once read a story about a plan for a new wind farm in Eastern Victoria (Australia) which was rejected by the local government because a particular type of rare bird might be killed by the blades. The environment report said that the rare bird had never been sighted in the area and even if it was it would probably not fly into the blades. However they couldn’t prove conclusively that this stupid bird wouldn’t fly into these blades and it gave the politicians a reason to shoot down the plan.

Since you cannot prove that any system is perfect it will be easy for political opponents of the new Flight software to keep shooting it down.

Thanks for taking the time to comment. I have it on my list of ‘things to do’ to get back to you on ‘Have you anything interesting to say about Data Quality’, because the answer is that I do, as does the Association I’m VP Publicity for – the IAIDQ (www.iaidq.org).

Looking like Greg from the Wiggles isn’t (last time I checked) a risk factor for travel according to the Dept Homeland Security, so your friend should count himself lucky.

However, you have hit on a number of the key problems that can result in IQ Trainwrecks and which make resolving some of these issues complex.

1) Training, education and measurement of front-line staff is an issue.

In Call Centres (where I began my career) people are measured on speed of call answer, length of call, average wait times etc. However these things often result in a reduction of information quality as corners are cut to hit these measured targets (IQ is rarely if ever a metric on scorecards… YET).

In Passenger flight screening there may be targets and goals (“no terrorist flies” type slogans). There may be a culture of applying the letter of the rule, which means that kids with the same name as people on the no-fly lists are subject to security checks or just not let fly. One suggestion… don’t put that type of check in the system. Give autonomy to the front-line people to make assessments based on information that presents itself to them… “Yes, the name flags a hit but the person presenting is 3 years old so is unlikely to be able to restrain a flight attendant.”

2) Conflicting legal rules and objectives

Related to number 1 is the need for the project teams and people defining processes to understand and accomodate the often conflicting legal issues surrounding information. This doesn’t just apply in the case of this type of information, but can affect everything from CRM to records retention.

One way of viewing it is that law is society’s expectation of how you conduct business. As with most ‘end user’ expectations it is complex and seemingly contradictory. However the quality of your deliverable is measured against those expectations.

3) Effectiveness of process

The key problem reported by L’il Johnny’s mother was not that her son was on a list, but that she couldn’t follow the process to take him off the list. Having tried to follow it myself, I suspect that the problem is that Johnny might have to fill out a form to allow his mother to act on his behalf (or someone may be telling Johnny’s mother that he needs to fill out the form).

Quality of information stems from quality of process. It is not a fatal problem of information quality if you have mis-matches or duplicates in your data, but if your processes to correct those errors are ineffective or just don’t work you wind up with an IQ Trainwreck.

My experience has been that you can throw money at systems until the cows come home, but until you clearly define your information subjects (what is a ‘passenger’? are all ‘passengers’ adults? if passengers can be non-adults, how do we identify this?), and have defined holistic end-to-end processes for how/when the information is to be captured, updated, maintained and destroyed then you will not achieve improvements in quality of your information.

I remember a few years back I presented at a CRM conference in Dublin and a delegate from a large insurance company confided in me that they had spent €xx million on a CRM system but “for all the good it does we might as well have just set fire to the money”. The root cause he identified was that the organisation hadn’t taken the time to understand what the processes and information that were key were and as such wound up blowing their budget for a system that didn’t meet the expectations of the Business and was deemed to be ‘not quality’.

So, stakeholder expectations (legal issues), training & culture (front-line staff), processes and definition of information subjects are all critical to delivering quality information.

Until Safe Flight gets the mix right on the above it will continue to eat budget and fail to deliver. If some of these issues (such as the corrections process) were reviewed an incremental improvement might be possible on the existing processes to improve the ‘customer’ experience.

aw dang.. I should have saved all of this for an article!