@NowSourcing.Com

Quality scoring is a black box.  Everyone involved knows it’s important, but some people consider it important enough to optimize for it while others go on about their day and accept the score they get, come what may.  Regular readers of these posts may realize that I am one to fall in the latter of the two camps, because I focus on conversions and not click-throughs in almost all cases.  After all, the quality score is how qualified the engines think your keyword and ad is to make them money.

Regardless of the tips I present below, I firmly believe that last statement.  The quality score is there for the engine’s benefit and not ours as advertisers.  It’s the engines grading us on how profitable we are for them.  If we’re not profitable enough with a high CTR, they charge us more.  However, is the quality score graded on a scale or on a curve?  If it’s on a scale then you can do everything right and get an “A” and be happy with your high score.  If it’s graded on a curve then your final grade is dependent on the performance of others.  If you do better than they do, you score high.  If you do worse, you score low.  In other words, your ability to write better ads with higher relevance than your competitor impacts your quality score.

Consider this: You have a quality score of 5 and your competitor with identical keywords and bids has a quality score of 10.  Since the keywords are identical the relevancy to the raw search query should be the same, meaning the only difference is the ad copy.  They have an awesome ad and you don’t, so their quality score is higher.  Typically, this means that they can get a click at the same position as your ad for a lower cost per click.  If they can spend less money per click, they can afford to go after more traffic, which means they have more opportunity to score conversions.  More conversions means more cash to fuel future Adwords spending and continued sales growth.

Do I have your attention now?  Good.  Here’s how to optimize your ad groups for quality score:

Keep your keyword count in ad groups between 25-50.  This includes match type variations of the same term.  This isn’t a steadfast rule, but is a good rule-of-thumb.  If you have more, make sure that they are keywords that focus around a very tight central theme.

Don’t bid on broad match.  Keeping everything to phrase and exact match will limit how likely it is for Google to match your keywords to an irrelevant term.  Remember, Google’s definition of relevancy may not be the same as yours.  (If you really want broad match, put it in a different ad group).

Separate your misspells.  If you use the dynamic keyword insertion in your ads, misspells will show up in the headline.  This will hurt your quality score, not help it.  Keep them separate and use the correct spelling of the term in the ad copy.

Don’t use dynamic keyword insertion in ads.  Surprised?  Why?  If you keep the keyword counts low in the ad group, you can write the ad to include those keywords.  Having an exact match term as a keyword and in the ad text without Google injecting it shows maximum relevance.  Besides, with everyone and their mother using the DKI these days, it doesn’t have anywhere near the CTR boost that it used to.

Now before you go off and redesigning your accounts to optimize for quality score, consider this: some of your terms will already have a high quality score.  If you have a couple terms in ad groups that score poorly (like 5-7) and some terms that score well (8-10), then simply remove those poorer terms and put them in their own ad group.  Copy your better converting ad from the original to include in the new ad group and bam!  You have a higher average quality score already.

One of the most frustrating things I can imagine while managing small PPC accounts is getting priced out of a keyword that has the potential for a lot of traffic and lots of sales. If you’re in that situation, there are ways to compete by getting maximum exposure and going beyond Google’s quality score. The secret: Advanced day-parting.

The original purpose of day-parting was to let advertisers turn their ads off when they don’t want ads within a campaign to be seen. This could be due to call center hours, or perhaps a drop in the conversion rate. Here’s a better idea: Use day-parting across multiple mirrored campaigns to make sure you get exposure at different parts of the day. Let’s walk through it step-by-step:

1) Go into Google’s reporting interface and run an hourly report regardless of date for a sufficient time period at the campaign level, preferably for a single campaign. Be sure to check the boxes labeled “impression share”A sufficient time period is usually at least 30 days, but possibly up to 90, depending on your conversion volumes. Make your report screen look like this:

2) Export to Excel, and sort by hour of day. You should have 24 entries, ranging from 0-23. You need to look for a couple things:

• At what hours of the day is your impression share lowest?
• At what hours of the day is your conversion rate the highest?

Your impression share is what percentage of search queries are you getting for the keywords within the given campaign. Low numbers mean you’re losing out on opportunities. High numbers mean you show up often.

3) For example’s sake, let’s say that our busiest hours are 6:00-7:00am, 12:00-2:00pm, and 5:00-7:00pm on weekdays. For these hours we’d set our ad scheduling to look like this:

The above example is an extreme where you only run the ads during these peak times. However, if you’re using advanced day-parting also allows you to change your max bids during these different periods within a day. So if you know the 12-2 hour is where you’re going to make all your money, you can dictate that those bids automatically be raised any percentage over the Max CPC at the keyword level that you’re comfortable with. During the other hours of the day you can run the ads at a lower percentage than the max CPC, so you don’t get charged as much per click. Keep in mind that your average position will decline.

This seems pretty thorough as-is, so why use mirrored campaigns? For some really competitive keywords, you may not even last through these small peak hours to get the exposure you’re looking for. In those cases you need more than one campaign with different daily budget caps to make sure your traffic is getting spread across the times that you want, and not whenever Google thinks its best.

Expect to spend 2-3 hours of research per campaign in reports before deciding which hours are best suited for your campaign. Due to that time requirement, I’d recommend doing this on only your top 1 or 2 campaigns, and only if you see your daily budget being maxed out constantly, but can’t afford to raise it. Another good sign is when Google suggest that you should at least double your daily budget to get max exposure.

Remember, the goal is not to get more impressions or clicks for the sake of getting traffic, but instead to maximize conversions at peak times of the day. If you execute this tactic and you’re not seeing an overall increase in sales within a month or so, don’t bother with the extra management. Kill the test and go back to what you were doing originally.