There has been an interesting case in the news recently about Canada Post asserting its intellectual property rights over postal codes. You may have read that Canada Post is suing an Ottawa-based company called Geolytica, which operates the website GeoCoder.ca. At issue is whether anyone has the right to sell or give away lists of Canadian postal codes. Canada Post claims that postal codes are their intellectual property. One interesting aspect of this case is that Geolytica insists that they have never copied any Canada Post data and they have merely gathered a list of every postal code that has ever been entered into their system by members of the public. Their database is “crowd-sourced”, not a copy of a Canada Post dataset and furthermore, Geolytica maintains that postal codes are facts and aren’t subject to copyright.
It will be interesting to see how this case plays out in the courts. Regardless of the eventual outcome, this case raises a data quality issue over and above the intellectual property question at stake. In particular, when it comes to postal codes, who do you trust?
If Geolytica has compiled a list of addresses captured from user entry over many years, then how reliable can this dataset be? Anybody who’s ever taken a close look at address data quality knows that address data capture is extremely problematic. Without reliable address quality controls, very high error rates are common. Completeness, accuracy, standardization – all of these are obstacles when capturing address data manually. So is a crowd-sourced dataset good enough to use as your standard of what is correct, complete and standardized?
Another issue is timeliness. There used to be a time when one of the most frequent complaints we heard was that some particular new postal code wasn’t found in the Canada Post data. This tended to be an issue in new subdivisions. Happily, since Canada Post introduced changes to their internal address data maintenance systems several years ago, we almost never hear this complaint anymore. There is about a 6% annual change rate within the Canada Post address reference data. Can you expect crowd-sourced datasets to keep up in terms of completeness and timeliness?
One other thing to think about is the stability of your data source. You don’t have to spend very much time with your search engine of choice to find dozens of sources of postal code files. We know of one organization who was using a Canadian postal code file from Zip Code World. Canada Post has successfully asserted that Zip Code World is not licensed to distribute a Canadian postal code file and Zip Code World is no longer supplying Canadian postal code data. One of the hallmarks of these grey (or black) market sources of postal code data is very low cost. Canada Post licenses postal code data for anywhere between $8,000 and $40,000 per year, depending on how you use it. If you’re using someone who is selling it for less than one tenth of this price, you might find your data supply suddenly cut off.
The best way to make sure that you are getting the best Canadian address data quality available is to use Canada Post SERP-recognized software like nCode. Canada Post is the only source for complete, accurate and timely address reference data and the best of this data is only available through SERP-recognized software.
If you’d like to understand more about address data quality and the issues of timeliness, completeness and accuracy, contact the address quality experts at Nova – we’re here to help.