You can find part one of this topic here.

Before I begin – hey! I’m in France! Now onto the good stuff … The last two things about Google AdWords that I want to talk about is budgeting and targeting, and then I’ll briefly mention Google Alerts and how I use that service, too.

Budgeting on AdWords has been a bit of a nightmare. Once I figured out how to effectively organize our campaigns, ad groups, ads, and keywords (phew!), I soon discovered just how dangerously popular those ads could be! Even though our click-through rate isn’t so high, there were still enough interested people finding and clicking on our ads that our daily cost exploded and became far more expensive than we could sustain on the budget we were given by OMDC. AdWords allows you the opportunity to set a daily budget for each campaign, but each time I thought I had fixed the problem, by lowering the budget of the highest performing campaign of the day, another campaign would pop up and become super popular. It was getting a little ridiculous.

I solved the problem finally by drastically cutting every single campaign’s budget to an absolute minimum for one day, and then started to slightly tweak based on the campaign’s popularity and how much the ads in each campaign cost. Google AdWords prices ad-clicks based on a combination of how competitive the keyword/location is and how good or poor the ad’s click-through rate is (the poorer, the more expensive).

I don’t entirely regret the very expensive days. For one thing, increasing your budget allows you to get a good feel for which ads perform really well and which perform poorly even with a high budget. It helped me cut ads that didn’t work and learn about what does work in an ad (the most important thing seems to be using your top performing keyword in the ad text).

Having to cut the budget dramatically also helped me to prioritize the campaigns – some of the campaigns (remember the Letterpress campaign?) did extremely well but were also extremely expensive, which we just couldn’t sustain. Keywords like ‘letterpress’ cost $1.20 or more per click, which is just ridiculous in our circumstance considering most people searching for letterpress likely aren’t looking to subscribe to a magazine about all kinds of fine printing and book arts, so we probably won’t see the $1.20 click turn into a sale. Even though the Letterpress campaign was very popular and gained a lot of traffic for the site, it just didn’t make sense to continue the campaign when I thought about what those users were probably looking for (wedding invitations, let’s face it) and how they likely wouldn’t purchase anything once they reached our site.

But how do you find out the conversion rate of your ads (that is, the rate at which someone’s click on your ad actually results in a purchase from your site)? Google AdWords does offer a conversion tracking service, but so far we haven’t implemented it for fear of complications with another Google service we use called Analytics. And anyway, that only shows you what your conversion rate is, not how to improve it. How can you try to increase the conversion rate so that the increased traffic actually results in increased revenue?

Well, I mentioned in part one the notion of website targeting. This is something I’m still very new to and am working to understand. Google AdWords doesn’t only display ads on search results; it can also display ads, if you enable the feature, on what are called ‘affiliate sites,’ which are sites like YouTube but also many smaller websites like blogs and even poetry sites. This is significant because Google also allows you to control which of these affiliates your ads might be displayed on.

I started checking which affiliate sites our ads were being displayed on, and was able to exclude sites like “freepoetry.com” or “freepoetryessays.com” (those are just random examples off the top of my head – not sure if those are real sites, be forewarned!). Any website with the word “free” in the URL is probably not attracting my target audience, since the Quill doesn’t have any free poetry available – we have poetry for sale. For some strange reason that I still haven’t puzzled out, I’m unable to access from France the individual URLs that our ads are being displayed on, meaning I can’t exclude individual websites – but I think that’s still okay because I managed to exclude a lot of unnecessary affiliate sites while still in Canada.

That’s all I have to write about Google AdWords for now. There is still a ton to learn and every time I log on there’s some new surprise, but at least the cost has stayed low and nothing too dire happens if I don’t check up on it for a few days.

Surprise! I think I’ll save Google Alerts for next week; this blog has, once again, become longer than intended. Next week I might even include some photos of my new ‘office’ as a treat! In the meantime, I’ll eat some pain au chocolat and work on the digital marketing sections of Spring 2011 tip sheets.