Thank you, Madam Chair. I will be brief. I want to avoid your hand signal.
The communications policy does in fact have some elements with regard to the polling, the public opinion research. In fact, we do have to test prior to the advertising in order for us to be able to measure and monitor whether or not the advertising has generated what our objectives were.
For example, as Mr. Wouters mentioned, when we run a set of advertisements with regard to the home renovation tax credit, it's trying to get a sense of how many people know about it beforehand and then trying to get a sense of how many people know about it afterwards. It's the same thing when we run advertising for a particular campaign around emergency preparedness, for example. How many Canadians would know what they should prepare for or stock in the case of an emergency? We then run some tests afterwards to get a sense of whether or not the advertising was effective, or whether or not there has been take-up of a 1-800 number, a service, or a specific program or initiative.
Any advertising campaign, I believe it's over $400,000 or $1 million—