From the Minister’s Mind…

I have just come across Dick Roche’s website ( It’s not bad as politico sites go. I was particularly interested in his post re: the Electoral Register issues. To an extent it is more telling in what it lacks than what it contains.

1)      The use of Census Enumerators or other temporary personnel to support local authorities in preparing the next Register as part of an intensive registration campaign to be conducted this summer“.

It is May now. As per my previous posts, Census Enumerators cease to be employed on the 21st of this month. I haven’t seen or heard any adverts recruiting for Electoral Register data gatherers. Irrespective of the lead times involved, recruitment of additional people to run the process is simply scrap and rework and will not deliver a sustained improvement in the quality and control of our Electoral Register. Unless the deficiencies in the processes are addressed any cleansed register will rapidly degrade in terms of its quality. Scrap and rework is not quality improvement, it is not value-adding. It is just scrap and rework – a costly exercise in standing still at best.

2)      Additional ring fenced financial resources will be made available to local authorities to update the Register“.

Excellent. Money will be available to throw at the problem. But in order for that money to be spent wisely, there must be a clear strategy across the local authorities and the Dept of Environment & Local Government as to how that money should be spent. Leadership is required. My suggestion for a quick hit:

·        Select a standard Information Quality toolkit to perform standardisation and matching within local authority datasets and match local authority Electoral registers against the register of Deaths and also the Register of marriages.

Standardisation on a single toolset allows for greater reuse of common functions (such as the standardisation of Irish addresses), the development of transferable skillsets between Local Authorities and economies of scale when negotiating with vendors.

This process can then be automated and an initial level of quality scorecarding can be put in place. However, updating the Register is only the start. It is still Scrap and Rework, albeit with some fancy toys in the mix to improve efficiencies. What is required is the expenditure of resources to identify root causes, assess and improve processes and in general conduct a root and branch overhaul of our Electoral Register processes to bring them to a standard that meets the expectations and needs of 21st Century post-Celtic Tiger Ireland.If the core processes are not addressed, everything else is just scrap and rework which will need to be redone again in a few years (if not months).

3)      Updated and consolidated guidelines to ensure all Local Authorities work to the same template

Standardisation of Information capture processes and procedures is an important first step. Reducing variation in the process is important. This is the nearest thing to a Process Improvement initiative that the Minister has mentioned thus far. But it isn’t enough. It addresses one potential root cause for the dismal level of quality in our electoral register – what are the others?? Will the consolidated guidelines make provision for people who want to update their name without moving address?

Fixing the templates without examining other root causes and reviewing the process and information architecture simply means that the scrap and rework will be better structured. Scrap and rework is non-value adding and does not improve quality in a sustainable way. Repetitive cycles of scrap and rework are costly exercises in admitting failure of governance.

Does the Minister propose here to establish a consolidated Governance Board to ‘own’ the standards around templates, toolsets, Information Architecture design etc. or is this just a memo that will be issued? If it is the former – then this is a positive step. The latter is just lip-service to Quality.

4)      An early start to be made to the local Authority Register Campaign for 2007/2008

Data-cleanse early, data-cleanse often. No matter how early you start your scrap and rework or how much time you allow for it, it remains Scrap and Rework unless there is a review of processes to remove sources of error and root causes of non-quality.

All this proposal does is push the ‘cycle time’ for an Electoral Register refresh out a bit and increase the cost of doing so. That is fine as a once-off to establish a baseline level of quality in your data, but should it become the ‘way things are done around here’ it is simply institutionalised scrap and rework.

Where are the Minister’s proposals for a strategic, long-term plan for continuous improvement of quality in the Electoral Register?

5)      “Local Authorities have been instructed to make use of relevant databases to cross check the Register; Councils in particular will be encouraged to use their own databases more effectively than they have to date;”
Cross-checking = matching. Matching of data between datasets requires an investment in tools and techniques to standardise the data and then match on name-plus-2 identifiers.

If the existing databases that the Councils have were not designed for the purposes of matching against other datasets, they may not hold sufficient data items to support a match. In addition, they may not be complete as surrogate views of reality – for example I don’t think Wexford County Council have any record of me living at my current address as my wife pays the Bin Charges and the car is in her name.

Additionally, the quality of these ‘surrogate sources’ for reality might also be questionable in terms of completeness, consistency, conformity, accuracy etc. and could lead to false positive matches or false negative matches.

Data Protection issues come into play here also and could land Councils in legal bothers if they use a data set for a purpose for which it was not provided.

Furthermore, continuous clashing of multiple data sets doesn’t improve the source processes – it is just another flavour of scrap and rework unless it is in a clear process context. Conducting these clashes using ‘traditional’ IT approaches will likely be costly and time consuming. This raises the need for a common “Second Generation” Information Quality management toolset across Local Authorities that can standardise and match across multiple data source types. There are a number of such tools on the market.

6)      “The possibility to allow local authorities access other databases to cross check the Register is now under consideration”
Access to more data does not lead inexorably to better quality Information. A clear process for what information is to be used and how is needed. Furthermore, technical issues about getting access to the data and integrating it into an Electoral Register master file would need to be considered.

Again – this raises the question of process purpose and function and the tools that would be available to Local Authorities to conduct the processes.

It smells like a Process Improvement, but unless the additional information can be used to address an identified Root Cause of process failure and Information Non-quality, it is simply information for information’s sake and would be another non value-adding form of scrap and rework.

7)      “Approximately 30,000 deaths occur each year: new arrangements to delete deceased persons from the Register have been agreed”
This is a process improvement piece that, at first glance, would address another obvious root cause of error in the Electoral Register.

What are these arrangements? Do they address the technical challenges of standardising data and matching across multiple data sets, particularly given the ‘interesting’ nature of Irish addresses?

Do the arrangements include recommendations on toolsets, techniques and processes for standardisation and matching of data?

If the arrangement is simply to share “electronic files” of data without addressing the process of making use of this information then we return to the point that access to more data does not lead inexorably to better quality Information.

8)      “An intensive information campaign will be mounted later this year”
What information will be given in this campaign? Will the Minister spend taxpayer funds educating people on an Electoral Register process that would seem to have some key deficiencies?

Will the Minister be conducting a Root Cause analysis and Process Improvement project to identify and address deficiencies in the Electoral Register process before engaging in “an intensive information campaign”? If not, then we must accept that we may simply be teaching people how to operate a broken process more efficiently.

Yes, an information campaign is required but in the absence of a review of the processes to identify and address root causes the information campaign would be unlikely to lead to significant sustained improvements in the quality of Electoral Register information.

9)      “Better ‘on-line’ facilities to help  people check the voting Register, including an additional on line facility through the Department of the Environment website”

What are the processes that would sit behind these on-line facilities? How will citizens be able to correct their Electoral Register data through these on-line facilities? How will on-line facilities serve to improve the quality of Electoral data?

One of the lessons from the Business world and initiatives such as customer relationship management (CRM) is that bolting a web front end on defective processes simply exposes the defective processes and information to your customers and the public.

There are any number of technical and detailed questions on the proposed information architecture for these on-line facilities that I could ask… but my reader’s eyes are probably glazing over a bit already…

10)  “A strengthening of the controls at polling stations at election time. New guidelines in this area will issue before the next Election”.
 The focus of quality management and quality improvement should be on building quality into the process, not inspecting defects out. While an improvement on the controls at the polling station is to be welcomed, a root and branch reform of the processes and governance for the Electoral Register would deliver more sustained quality improvements.

Catching someone personating on the day is good. But denying them the opportunity in the first place is even better. A statistical report showing the continued increase in accuracy of the register is a nicer headline than a report of the numbers of persons caught in the act of corrupting our democracy. Apart from anything else, the question people will ask is how many weren’t caught for each one that was?

The Minister pats himself on the back for his practicality, sense and realism. I question the completeness of the thinking and the long-term sustainability of any quality that would arise from his proposals. Nothing that is proposed looks to identify and eliminate root causes of error. Nothing that is proposed seems, to my reading, to amount to anything more than reactive scrap and rework.

Minister – prove me wrong.