Photo credit: stephen_woolverton When launching a new Google Adwords campaign there are a few default settings that you need to change right out of the box in order to maximize your campaign's performance. Let's get started, beginning with the most important setting:

 

1. Search network / Content network search-content-setting By default, Google sets up your Adwords campaign to advertise on both the search network and the content network. In offline advertising, this would be the equivalent of carefully crafting three television ads, handing them to the studio and then having them tell you "hey, we're gonna take the audio from these and run some radio ads on our affiliate station and charge you for those too, cool?" No, it's not cool. Just like radio and TV, the search network and content network are inherently different animals and unique care needs to be taken when creating a campaign for either network. Don't hand Google more money until you know exactly why you're doing it and what you'll get back. While search ads are only displayed to people who are typing the keywords you've targeted into Google and are actively searching for whatever you're advertising, content network ads are true "interruption" marketing and are shown to people all over the web who have displayed no intention of seeking out your product or service other than visiting a website with some of your targeted keywords on it. Not only do you need to create separate plans of attack, the statistics and success metrics for the search and content network are completely different. By creating a single campaign and having it run on both, you're combining two giant advertising platforms into one even bigger mess of data that will be incredibly frustrating to sift through. On the search network, click-through-rates (CTR) generally hover between 1-5%. On the content network, a CTR of 0.3% could mean you're making a lot of money. See why you don't want both of these networks dumping stats into the same place? Plus, you want to know where your conversions are coming from. Maybe you lose money hand-over-fist on the content network, but the search network gives you a 320% ROI. It will be harder to know without splitting up the campaigns. Don't listen to Google. Keep the ratio of networks and campaigns at 1:1.

 

2. Standard/Accelerated Delivery adwords-accelerated-deliver When it comes to Adwords, just like all other marketing methods, testing is everything. In my opinion, the thing that makes Google Adwords the number one online advertising platform is Google's incredibly large reach/market-share that lets you get the fastest marketing test results available in the world today. Thanks to Adwords you can quickly know if a product is worth creating, doubling down on or ditching just by targeting some relevant keywords and writing a few ads. No clicks? Forget about it... either your idea about who your target market is is wrong, there is no demand for your product or service, or both. (Or maybe you're just not good at picking keywords and writing ad copy. If that's the case, let me do it for you as part of my all-inclusive PPC management services package). Google excels here because Adwords compresses the one factor that is the enemy of all marketers: time. The faster you get statistically significant test results, the faster you can make adjustments to your business or improvements to your ad copy. If you make these changes faster than your competitors, you win. If it's vice-versa, you'll be left in the dust. That is the reality of the current Adwords landscape for many industries today. Accelerated ad delivery helps with this by telling Google to run your ads as quickly as possible. It's an instruction from you to take the speed governor off of your Adwords campaign and just let it fly and see what happens. If I had to guess, I think Google defaults new campaigns to "standard" delivery because they don't want new advertisers blowing their ad budget in the first couple hours at the start of every day, getting discouraged and then closing their accounts. You've heard the saying "you can sheer a sheep many times, but skin it only once"? Google doesn't want to skin you your first day in the game; if your campaign isn't performing, they want to give you time to learn what you're doing wrong and correct it before your entire budget is in their pocket and you walk away from the table forever. Good poker players know you never take a sucker for all of his cash in a single sitting. I could be wrong. Regardless, if you're serious about Adwords, hit the accelerated delivery button and don't look back.

 

3. Optimized / Even Ad Rotation adwords-standard-rotation As I stated above, in order to beat your competitors on Adwords you need to be constantly improving your campaigns. In my opinion the best way to do this is by continually split testing two or more ads against each other for every ad group in your account, then acting on the test results on a daily basis. (This is also one of the many things I take care of for you under my PPC management services package). This provides a great opportunity for a smart advertiser like you  (hey, you're reading my blog, right?) to get ahead of your competitors, because Google's data shows that the majority of advertisers on Adwords aren't split testing their ads and aren't making daily adjustments once the tests are done. I know this because the default ad rotation setting for new Adwords campaigns is "Optimize," which lets Google pick a winning ad for you. If you keep Google's default and let them "optimize" your ads for you, after a short period of time Google will declare your split test is over, pick the ad with the highest CTR at the given moment, and then show it way more often than any other ads you've written for your ad group. Google's "optimized" setting doesn't give your ads time to fail on their own. This poses a big problem for intelligent pay-per-click marketers who wait for statistically significant test data before making changes to their ads. If you write two ads and after 24 hours you look at the reports and see that the first ad received 4 clicks while the other received 2, would you call off the split test and crown the first ad the winner? Of course you wouldn't, because you don't have enough data to show that the first ad will beat the second in the long run. It would be like flipping a coin, watching it land on heads 3 times in a row and then stating you're certain that heads will beat tails in the long term. You have to wait to collect enough data so that you can be 95% statistically certain that one ad will outperform the other if they both ran for eternity. The only way you can be absolutely (okay, 95%) certain that one ad will beat another is by running their stats through this equation: split-test-equation Look at that equation and you can see that split testing your ads by hand, and doing it the right way, can be a lot of work. Google doesn't trust you to do this work, and how could they? Thankfully I don't do this by hand. The tool I use every day for this is called WinnerAlert. This thing is invaluable to my business. Every day it gathers statistics for every single ad in every ad group in every campaign in my account (and my clients' accounts), runs them all through that scary looking statistics equation, and then alerts me via email only when a test has been completed. You can see how powerful this is. It only emails me when I need to take action by writing another ad. Once I do, WinnerAlert automatically starts the test all over again. However, there is a small monthly fee for this service. If you spend over $200 a month on Adwords I would argue that it's well worth it, but I want to provide you with a free tool that does the same thing, although a lot slower and not on autopilot like WinnerAlert (you have to input the data yourself): SplitTester.com. Why doesn't Google build a tool like WinnerAlert? Maybe they're working on it. And maybe, once it's finished, they'll trust you to split test your ads the right way. Until then, be smart: go against Google's recommendation and handle your own split testing. Your campaign's click-through-rate and return-on-investment will thank you for it. Have any questions? Post them in the comments!

Share with others...

Comments (6)

Track comments via RSS 2.0 feed. Feel free to post the comment, or trackback from your web site.

  • David on Aug 26, 2009

    Great article!

    I’m not sure I agree with using accelerated delivery. If you have unlimited funds and can afford serving ads at breakneck speed all day long, then it seems like a good idea, however, what if your funds are limited? You could burn through your daily budget before lunch and miss all the users that only search after noon. Your thoughts?

  • John Jorgensen on Aug 26, 2009

    David,

    Good point, I should mention something in this post about the Google Adwords “dayparting” feature that lets you control what time of day your ads run. For example, you could set it to pause your campaign until normal business hours so someone at your office is around to answer the phone. I would personally use dayparting to control what time of day your ads show, not your delivery options.

    Have you found dayparting useful at all for your business? Do you see your CTRs/conversions fluctuate a lot based on time of day? Thanks for the question.

  • AdWords Split Tester on Aug 28, 2009

    There is a great tool for calculating statistical significance of your data – AdWords Intelligence. It has a Split Tester feature that will make the statistical analysis. It can also easily be linked with your AdWords ads, so it automatically downloads the performance data and makes defined actions (delete or pause ads) directly to your account (if you can declare a winner with high statistical significance).

  • David on Aug 30, 2009

    John,

    I’ve not tried dayparting. I know it exists, but have always been concerned that I’ll miss that engineer that is working into the wee hours of the night. Additionally, Dajac products can sell in many different timezones. Restricting based on time of day in the Eastern timezone doesn’t make sense in relation to our customer base.

    I wasn’t aware I could break keyword statistics down to time of day. How do I do that?

  • Hey, Adwords Advertiser: Google Thinks You Suck At Split Testing Ads | Internet Megaphone on Sep 1, 2009

    [...] recommend that as an advertiser you shouldn’t listen to Google’s recommendation to optimize your [...]

  • John Jorgensen on Sep 1, 2009

    David,

    Working on a post about this right now..

Leave a Comment

Share This Post

Subscribe Subscribe to RSSStay up to date with my blog

Twitter Follow me!Follow me or get followed

Facebook FacebookCheck my Facebook profile

linkedIn linkedInAdd me on linkedIn

Hi, I'm John Jorgensen

Hello! I'm an internet marketing consultant / geek living in Manhattan Beach, California. I also grill a mean steak. Have an online marketing question? Contact me.