One of our clients did just that and asked us:

“We are seeing QS below 5 for some of our top spending keywords. Should we optimize these keywords and try to increase our QS?”

It’s a great question, so let’s do a quick refresher. QS is still mostly made up of CTR, and there is often a give-and-take between quality score and conversion rates. In other words, ads with higher CTRs show lower conversion rates.  You may be able to write a more click-friendly ad, but the marginal clicks you will receive are from less qualified folks. Now when you mouse over the little bubble (on the Keywords tab) you will get a bit more detail regarding QS. For example, one of our main keywords has an AdWords QS of 4, which means “average” for CTR, ad relevance and landing page.

Let’s say we want to test a price in the 1^st line of ad text (ie, $24.95), which Google also will include in the headline of the ad. By putting price upfront, we might exclude folks who want to see something like “up to 50% off”. Sure, a 50% off ad may have a better CTR (QS), but it may not be as profitable for you (the advertiser).

To sum it up, we generally do not “optimize for quality score” and instead prefer to optimize on harder metrics such as CPA or ROAS.

Old News

We recently began reporting to a new hire for one of our long-time clients. Since we have had this client for several years, we naturally have considered dayparting from time to time. This new direct report suggested we reconsider using dayparting in our AdWords account, citing it as a way to increase bids during periods of heightened impression activity. This is a poor idea. Let me explain why.

The key thing to remember is that we want to maximize value; impressions alone denote zero value, and they may even have a negative correlation with value. Value is comprised of revenue (or some other conversion metric) and click costs.  If we talk only about revenue, conversions, or conversion rates, we are missing out on a key component of value.

For instance, it may be the case that revenue and revenue-per-click (RPC) increase greatly from 2pm to 5pm, but unless cost rises at a slower rate than does revenue, increasing bids is probably not a great idea.  If, however, average CPCs for some reason stay the same while RPC increases 20%+, then a daypart bid multiplier could be a great idea.

Dayparting as a response to increased or decreased impression activity misses the point that the goal is centered around value. Further, to calculate value you have to take costs into account. We do want to increase revenue or conversions, but within a certain efficiency metric. For a conversion-oriented account, CPA (Cost Per Acquisition) is the efficiency metric. For a revenue-oriented account, ROAS (Revenue Over Ad Spend) may be the efficiency metric.  But to talk about impressions or clicks without talking about revenue or conversions, not to mention costs, will lead to poor decisions.

We all love feeling as though we’ve dodged a bullet, right? In this post we provide some practical tips and real-life examples for using AdWords tools without them using you.

The AdWords Search Funnel

• Top Paths report: we especially enjoy the Transitions reports.
• Assisted Conversions report: understanding assist roles may change your bidding approach.
• Path Length report: this will tell you how many folks are NOT converting on the first click alone.

Head over to Trada and read the post! http://www.trada.com/blog/advanced-adwords-tools/

• Hitting your budget and losing out on available impressions
• Missing CPA targets
• Not realizing an account has been paused or is no longer getting traffic
• Failing to install a conversion tag on new landing page
• … and more

The first step in setting up alerts in Google AdWords is to dream up some worst-case scenarios. In our agency, for example, we need to know when a campaign spends 95% of its budget because if we hit up against budgets on a consistent basis, we are likely losing impression share and have some bidding work to do. That might throw up a big red flag to our client. So… red flags = alerts!

We might set up this alert like the following: 

Other worst-case scenarios might be:

• a website going down
• conversion rates dropping below a target
• cost per conversion going above a target
• someone forgetting to un-pause a campaign

Using alerts to signal problems early on can be a real savior for any advertiser. Make sure to do these at the MCC level (if possible) and alert more than one team member for redundancy.

Once you have some alerts set up, let them run for a few days and tweak as needed. But keep in mind, you cannot Edit an existing alert in Google Adwords! Instead, you’ll need to create it over from scratch. Another good practice is to advise your team on how to handle alerts that signal problems. Put together a contingency plan and assign responsibilities accordingly. Worst-case scenarios can and should be managed as long as someone with authority is paying attention!

Does ad position affect click-through rates (CTR)? Yes.

Does ad position affect your campaign’s cost? Yes.

Do ads in the top ad position get more conversions? Yes.

Does ad position affect conversion rates? No.

Can you guess the most important question above? Yes, it’s the one answered No. This also happens to be one of the most misunderstood points about ad position. Listen to what Google’s chief economist, Hal Varian, has to say:

“We have used a statistical model to account for these effects and found that, on average, there is very little variation in conversion rates by position for the same ad.”

That’s pretty straightforward, and you should trust Hal. To be more accurate, it’s totally possible that for any given account ad position may actually correlate with differences in conversion rates. But in aggregate – and in virtually all of the PPC accounts we have managed over the years – we have not found this to be the case.

So, to recap: if you want more clicks, put your ad at the top. If you want more conversions, go for the top. But if you do that, you’re going to spend proportionately more money. Just don’t expect those clicks from the top to convert at a higher rate.

It’s all the same when it comes to ad position and conversion rates.

In the past, when we’d run Dimensions reports by Day, the date would appear but the actual “day” would not. We’d see something like: Oct 11, 2011… Oct 12, 2011… etc. Not very helpful if you want to analyze weeks of data to look for similaries across, say, Saturdays and Sundays.

Now what we see (screenshot below) is the day just before its date: Tue, Oct 11, 2011… Wed, Oct 12, 2011… etc. When we download this report, we find it much easier now that we don’t have to whip out the calendar to make additional notes on what day falls on which date. (Note: We did find ourselves doing some quick excel magic to extract the day from the string, allowing our pivot table to show the right stuff.]

We’d like to thank the AdWords team for not only listening, but also for helping us save time and improve our day-to-day workflow. Hey agencies, did you notice this, too?

We stumbled upon one of those improvements and thought we’d share it with you.

In adCenter Desktop, when you have a large group of broad match keywords that you would like to create phrase and exact match versions for, the long method to do this is to select your keywords, copy and paste into your ad group, change their match types, and repeat the process until you have what you need.

A much faster way to accomplish this, thanks to a nice shortcut in adCenter Desktop — which Google’s AdWords Editor cannot boast — is to right-click on a keyword or group of keywords, choose Add Match Type, and click All. The result is pretty slick: instant keyword match type expansion! But wait… it gets better.

adCenter was keeping track of search queries, categorizing them as broad, phrase or exact, and assigning data to each match type without you even knowing it. Now you can view performance statistics to see data for each of the new match types. For example, splitting out a broad match keyword that earned 100 conversions last month could reveal 45 conversions came from the exact match version, 25 from the phrase match, and the rest from broad. This is great information to have, and all you needed to do was “unhide” it.

Before you call it a day, you need to do one more thing. Since AdCenter automatically assigns the ad group’s default bid to these new match types, you’ll need to set new bids accordingly. It’s a bit tedious, but well worth the time — your account will love you!