Some Results From Our AdWords Experiment

Posted On: May 18th, 2009 | Posted by: Dean

I mentioned in a previous post that a group of MBA students from Kellogg was helping Genlighten with a Google AdWords experiment. They’re part of an annual competition that’s supposed to help teach participants how Google Ads work. Here’s a snapshot of the results they’ve gotten thus far for their ads aimed at recruiting lookup providers:

Provider Campaign Results

Provider Campaign Results

I’ve only shown three of the most popular variations of the ads they created. The columns to the right summarize some of the key ways of measuring the ads’ impact:

  • “Clicks” is the number of times the ad has been clicked on
  • “Impr.” is the number of times the ad has appeared, either on a search results page or on a website that accepts Google ads.
  • “CTR” is the click-thru rate, basically the the number of clicks divided by the number of impressions. Yup, it’s actually that low.
  • “Cost” is how much we’veĀ  been charged for the clicks we’ve received. The cost of these ads is running around $0.30-$0.50 per click, sometimes even less. During the early part of the campaign, the team set the maximum amount they were willing to pay per click at $0.50. They’ve since raised that slightly.

These metrics make up one component of the “return on investment” calculation we need to do to decide whether Google AdWords are a good way for Genlighten to recruit providers. The other component is the number of interested providers who sign up to download our Provider “How To” Guide and receive an invitation to our private beta after clicking on one of these ads.

Depending on how we count them, that number is probably about 5. That’s out of 150 total clicks across all variations of the provider-oriented ads. So the “conversion rate” — the percentage of those who come to the site via our ads who express interest in becoming providers — is only about 3%. At $0.50 per click, that means we’ll need to pay about $15 to recruit a single provider.

Is that a good result? It depends on how many lookups that provider is able to get clients to order, obviously. But based on what I know now, I’d be pretty happy if we could reproduce that number consistently. I typically spend several times that to recruit providers at genealogy conferences such as those put on by NGS, NERGC and Family History Expos.

But is the comparison an appropriate one? When someone visits my booth at a genealogy conference and expresses interest in becoming one of our providers, it’s usually pretty clear to me that they’re a) serious and b) qualified. I don’t have any such evidence for those who come to our site via AdWords. And I won’t really know for sure until I see how they perform as providers. That concerns me.

For now, though, I’m pleased with the AdWords experiment and the tremendous amount of data we’ve been able to gain from it. In a subsequent post, I’ll talk about our results in trying to attract lookup clients via AdWords.

I’m curious… what do you think of the ads the Kellogg students composed? Would you have clicked on them had you come across them on a Google search results page or on a genealogy website? Please let me know in the comments. Thanks!

Leave a Reply